ML20207F074

Nuclear Power Plant THERMAL-HYDRAULIC Performance Research Program Plan

ML20207F074
Person / Time
Issue date:	07/31/1988
From:	NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES)
To:
References
NUREG-1252, NUDOCS 8808180209
Download: ML20207F074 (84)

Text

. .

j.

NUREG-1252 i Nuclear Power Plant Thermal-Hydraulic Performance Research Program Plan U.S. Nuclear Regulatory Commission Office of Nuclear Regulatory Research p e nouq

. /

5 0. .s ,I

%;..,...f

+

g 81 g g 880731 1252 R PDR

e

, s NOTICE Availability of Reference Materials Cited in NRC Publications Most documents cited in NRC publications will be available from one of the following sources:

1. The NRC Public Document Room,1717 H Street, N.W.)

Washington, DC 20555

2. The Superintendent of Documents, U.S. Government Printing Offica, Post Office Box 370"e Washington, DC 20013 7082

3. The National Technical Information Service, Springfield, VA 22161 Although the listing that follows represents the majority of documents cited in NRC publications, it is not intended to be exhaustive.

Referenced documents available for inspection and copying for a fee from the NRC Public Docu-ment Room include NRC correspondence and internal NRC memoranda: NRC Office of Inspection and Enforcement bulletins, circulars, information notices, inspection and investigatior: notices; Licensee Event Reports; vendor reports and correspondence; Commission papers; and applicant and licensee documents and correspondence.

The following documents in the NUREG series are available for purchase from the GPO Sales l Program: formal NRC staff and contractor reports, NRC-sponsored conference proceedings, and 1 NRC booklets and brochures. Also available are Regulatory Guides, NRC regulations in the Code of I Federal Regulations, and Nuclear Regulatory Commission Issuances.

I Documents available from the National Technical Information Service include NUREG series  :

reports and technical reports prepared by other federal agencies and reports prepared by the Atomic Energy Commission, forerunner agency to the Nuclear Regulatory Commission.

Documents available from public and special technical libraries include all open literature items, such as books, journal and period; cal articles, and transactions. Federa/ Register notices, federal and state legislation, and congressional reports can usually be obtained from these libraries.

Documents such as theses, dissertations, foreign reports and translations, and non-NRC conference proceedings are available for purchase from the organization sponsoring the publication cited.

Single copies of NRC draft reports are available free, to the extent of supply, upon written request to the Division of Information Support Services, Distribution Section, U.S. Nuclear Regulatory Commission, Washington, DC 20555.

Copies of industry codes and standards used h a substantive manner in the NRC regulatory process are maintained at the NRC Library, 7920 Norfolk Avenue, Bethesda, Maryland, and are available there for reference use by the public. Codes and standards are usually copyrighted and may be purchased from the originating organization or, if they are American National Standards, from the American National Standards Institute,1430 Broadway, New York, NY 10018.

1 i

4 ,

l NUREG-1252 I R4  ;

1 l l

Nuclear Power Plant Thermal-Hydraulic Performance Research Program P an l

l i

Manuscript Completed: June 1988 Date Published: July 1988 Division of Reactor and Plant Systems Offico of Nuclear Regulatory Research r U.S. Nuclear Regulatory Commission Washington, D.C. 20555 i l

Wp" m,,,)

I

ABSTRACT The purpose of this program plan is to present a more detailed description of the thermal-hydraulic research program than that provided in the NRC Five-Year Plan so that the research plan and objectives can be better understood and evaluated by the offices concerned. The plan is prepared by the Office of Nuclear Regulatory Research (RES) with input f rom the Office of Nuclear Reactor Regulation (NRR) and updated periodically. The plan covers the research sponsored by the Reactor and Plant Systems Branch and o Defines the major issues (related to thermal-hydraulic behavior in nuclear power plants) the NRC is seeking to resolve and provides plans fnr their resolution; o Relates the proposed research to these issues; o Defines the products needed to resolve tnese issues; o Provides a context that saows both the historical perspective and the relationship of individual projects to the overall objectives; and o Defines major interfaces with other disciplines (e.g., structural, risk, human factors, accident management, severe accident) needed for total resolution of some issues.

Although thermal-hydraulic phenomena pervade the entire range of safety issues, including accident prevention and mitigation and the consequences of severe accidents, the thermal-hydraulic research covered in this program plan is primarily concerned with accident prevention and mitigation before significant damage occurs to the-reactor core. A separate plan covers research concerned with accident management after damage occurs to the core.

This plan addresses the types of thermal-hydraulic transients that are normally '

considered in the regulatory process of licensing the current generation of light water reactors. This process is influenced by the regulatory require-ments imposed by NRC and the consequent need for technical information that 4.

supplied by RES through its contractors. Thus, most contractor programmatic work is administered by RES. Regulatory requirements involve the normal review of industry analyser of design basis accidents, as well as the understeding of abnormal occurrences in operating reactors. Since such transients often involve complex thermal-hydraulic interactions, a well planned thermal-hydraulic research plan is needed.

The relationships and responsibilities of offices within the NRC for thermal hydraulics can be summarized as follows. The lead role in defining issues and objectives and performing technical review is assumed by NRR. The technical basis and the ana!ytical tools for decisions involving thermal hydraulics are largely supplied by RES, which has the statutory authority and responsibility to develop recommendations for, engage in, and contract for research deemed necessary for performance of NRC's licensing and regulatory function. RES has iii

also assumed a major role in the use of this research for the resolution of safety issues and development of regulations. Compliance with these regulations is monitored by NRR. The Office for Analysis and Evaluation of Operational Data

-(AE00), which is independent of the routine regulatory process, identifies issues for review based on plant operational data. AE00 also uses the products of the thermal-hydraulic research to meet their analysis needs for-incident I review and tochnical training. In general, all offices.become involved with I addressing short-term issues, while RES has the additional responsibility to I

anticipate future issues and plan research for their resolution.

l l

i l

iv

TABLE OF CONTENTS P_ag iii ABSTRACT .-..............................................................

vii EXECUTIVE

SUMMARY

1-1

1. INTRODUCTION .......................................................

1.1 Research Purpose and Goals .................................... 1-1

1. 2 Products of Thermal-Hydraulic Research ........................ 1-3 1.3 Historical Perspective of Thermal-Hydraulic Research . . . . . . . . . . 1-7 1.4 Current Thermal-Hydraulic Research Needs ...................... 1-9 1.5 Interrelationships (NRR, AE00, Regions, and RES) .............. 1-11 1.6 Coordination with U.S. Industry ............................... 1-13 1.7 International Cooperation in Thermal Hydraulics ............... 1-13 1.8 Future Plans for Thermal-Hydraulic Research . . . . . . . . . . . . . . . . . . . 1 2. THERMAL-HYDRAUL IC ISSUES AND REGULATORY NEEDS . . . . . . . . . . . . . . . . . . . . . . 2-1 2.1 ECCS Rule Revision ............................................ 2-1 2.2 Code Scalability, Applicability, and Uncertainty (CSAU)

Evaluation ............................................ ....... 2-2

2. 3 Evaluation of Operational Performance of B&W Plants Versus Pl ants of Other Vendors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3 2.4 Response of B&W Once-Through Steam Generators to Operational Transients and Accident Conditions ............. 2-4 2.5 Response of S&W Plants to Small-Break LOCA .................... 2-4

2. 6 Response of W/CE Plants to Small-Break LOCA ................... 2-5 2.7 Long-Term Cooling After a LOCA ................................ 2-6 2.8 Boron Mixing in BWRs After an ATWS ............................ 2-8 2.9 Primary System Depressurization to Avoid Direct Containment Heating ........................................... 2-8 2.10 Downcomer Fluid Temperature Under Pressurized Thermal Shock Conditions .............................................. 2-9 2.11 Water-Hammer Diagnosis ........................................ 2-9 2.12 Safety of Advanced LWRs ....................................... 2-10 2.13 Operator Guidelines and Improved Simulators ................... 2-10 2.14 Containment and Balance of Plant .............................. 2-11

3. RESEARCH PROGRAMS ..... ............................................ 3-1 3.1 Major Programs Planned for Completion in 1991/1992 Time Frame.. 3-1 3.1.1 20/3D ............................................... 3-1

, 3.1.2 ROSA-IV .... ........................................ 3-5 3.1.3 Multiloop Integral System Test ....................... 3-7 3.1.4 International Code Assessment Program ............... 3-10 3.1.5 TRAC-PWR ............................................ 3-11 3.1.6 TRAC-BWR ............................................ 3-13 3.1.7 R E LA P S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3- 14 3.1.8 C S AU Imp l eme n ta ti o n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-16 v

l l

TABLE OF CONTENTS.(Continued)

P,_ ate

3. 2 Ongoing Base Programs ......................................... 3-20 3.2.1 Thermal-Hydraulic Technical Support ................. '3-20 l

3.2.2 . Nuclear Plant Analyzer............................... 3-24

-3.2.3 Nuclear. Plant Data Bank'............................. 3-25 3.2.4 Basic Studies ....................................... 3-25 3.3 New Programmatic Initiatives .................................. 3-27 3.3.1 B&W OTSG Testing .................................... 3-27 3.3.2 Continuing Experimental Capability .................. 3-28 3.3.3 Support to NRC Operations and Technical Training Centers ........................... ....... 3-29 h 3.3.4 Support to Regional Inspectors ......... ............ 3-29 3.3.5 Containment and Balance of Plant .................... 3-30 3.3.6 "Front-end" Support for Accident Management . . . . . . . . . 3-30 APPENDIX A Recent NRR User Need Requests and RES Responses ........... A-1 APPENDIX B Thermal-Hydraulic Technical Support Center ................ B-1 APPENDIX C Acronyms ..................................................'C-1 l

r I

o l -

vi

EXECUTIVE SUMfiARY One of the five major programs in the Office of Nuclear Regulatory Research concerns preventing damage to reactor cores. The program consists of four elements: plant performance, human performance, reliability of reactor systems, and accident management. The subject of this plan is plant performance, and the emphasis is on prevention of severe accidents. The programs concerning human performance and accident management have separate plans. This research program encompasses the operations of the reactor as a system, including control of power l level, maintaining water in the reactor system, core cooling and heat removal, j and maintaining proper coolant temperatures and pressures. It pertains to both normal and abnormal conditions, including accidents. The results are used to ensure that plant equipment, operational procedures, and training are adequate to deal with operating events and prevent serious accidents.

The resaarch described in this plan provides info.mation on thermal-hydraulic response (fluid movement and heat transfer) of the reactor coolant system when the plant is subjected to transients and malfunctions caused by pipe breaks or other abnormal events. The information is used to develop and assess computer codes that help set standards on equipment performance and to confirm the effectiveness of various operating procedures to prevent core damage. This plan provides a technical basis for planning and conducting current and future thermal-hydraulic research. The plan is consistent with assumptions and guidance in the NRC Fiva-Year Plan published in March 1988. The plan fulfills the commitment in NUREG-1244, "Plan for Integrating Technical Activities Within the U.S. NRC and Its Contractors in the Area of Thermal Hydraulics," for a thermal-hydraulic research program plan.

The plan discusses 14 issues being addressed through research activities and the regulatory needs they are to meet. The plan then describes eight programs that have a defined endpoint. The programs are referred to as 2D/3D, ROSA-IV, MIST, ICAP, TRAC-PWR, TRAC-BWR, RELAP5, and CSAU. The plan continues by describing four programs that have a perceived long-term need and thus do not have a defined endpoint. These programs provide technical support, nuclear plant analyzer improvement, plant data bank support, and basic studies. Finally, the plan describes new programs that are being developed. These programs cover once-through steam generator (OTSG) testing, nuclear plant analyzer utilization, water-hammer guideline development, and containment-related and accident-related thermal-hydraulic studies. A separate plan (NUREG-1236) describes in more p detail the research related to B&W plants (MIST and 0TSG).

For each of the above programs, an explanation is given as to why the research is needed. Future (expected) results and intended end products are specified.

The anticipated use of specific end products is described. Expected completion dates are included for significant parts of the programs, vii m

1. INTRODUCTION 1.1 Research Purpose and Goals The NRC has the responsibility to ensure that nuclear power plants are designed, constructed, and operated in a manner that will ensure the protection of the health and safety of the public. The NRC operates according to the principle that safety of plant design, construction, and operation is the responsibility of the licensee. Nevertheless, the NRC staff must have the ability to independently assess plant designs and safety analyses submitted by license applicants and to review operating experience. This requires that a sound understanding be obtained of the important physical phenomena that may occur during transients in operating power plants.

The purpose of the thermal-hydraulic research program is to produce the informa-tion necessary for the NRC to carry out its regulatory function by providing the NRC staff with the capability to accurately predict the expected behavior of the fluid systems in light water reactors (LWRs) during all normal and upset conditions. This ensures that regulatory actions affecting design, operation, and maintenance of those systems can be made on a firm technical basis. In practice, this requires the ability to understand and describe the processes of heat transfer and fluid flow for the range of transients that may occur.

Transients may be grouped into three classes of events: (1) design basis accidents required to be analyzed in license applications; (2) events that have actually occurred in operating plants; and (3) other transients that may be postulated, usually based on risk assessment studies, for the different plant designs in operation, under construction, or envisaged in the United States.

Design basis accidents consist of an initiating event such as a pipe break or failure of an active component (pump or valve) followed by a single failure of an active component. These events, shown in Table 1, are documented in Chapter 15 of the Final Safety Analysis Report and reviewed as part of the licensing process.

The second class of events ranges from the Three Mile Island-2 accident (March 1979) through the loss-of-feedwater event at Davis-Besse (June 1985) and the overcooling transient at Rancho Seco (December 1985). Other significant operating transients have occurred at Arkansas Nuclear One (1980), Ginna (1982), Hatch (1980), and elsewhere.

The third class of events falls outside the design basis review. Most thermal-hydraulic research over the past decade has concentrated on events within the design basis envelope, with much less attention paid to beyond design basis scenarios. Such events include some combination of multiple failures and/or operator eriors and are revealed through probabilistic safety studies and op-erating experience. Thermal-hydraulic research for events outside the design basis envelope includes the investigation of the effects of different phenomena in the reactor coolant system and pressurized water reactor (PWR) secondary systems and LWR ba131ce of plant such as water hammer, fluid / fluid mixing, fluid-structure interaction, auxiliary feedwater flow distribution, steam generator heat transfer, liquid inventory distribution between reactor vessel and loops, and direct contact condensation phenomena.

1-1

i Table 1 REQUIRED ACCIDENT AND TRANSIENT ANALYSIS PWR BWR Decrease in feedwater temperature Decrease in feeuwater temperature Increase in feedwater flow Increase in feedwater flow Increase in steam flow Increase in steam flow Inadvertent opening of steam generator safety valve Steam line breaks Closure of main steam isolation valve ,

Loss of external load; turbine trip Loss of external load; turbine trip loss of nonvital AC power Loss of nonvital AC power Loss of normal feedwater loss of normal feedwater Feed line breaks Loss of forced reactor coolant flow Loss of forced reactor coolant flow Startup of idle reactor coolant pump Startup of idle recirculation pump or increase in recirculation flow Reactor coolant pump motor seizure or Recirculation pump motor seizure or  !

shaft break shaft break .

l Boron dilution from CVCS malfunction Increase in inventory from inadvertent Increase in inventory from inadvertent actuation of high pressure actuation of high pressure injection or CVCS overfeed core spray, high pressure injection, or reactor core isolation cooling system Inadvertent opening of pressurizer Inadvertent opening of ADS valves relief valve Loss of coolant accident loss of coolant accident Reactivity and power distribution Reactivity and power distribution anomalies (boron dilution, rod anomalies (rod withdrawal) withdrawal) 4 1-2

New models for these phenomena are required based on separate effects experiments. Simulation of these events also requires more complete modeling of the balance of plant, including the control systems, than may exist in current computer codes.

The specific functions the NRC staff performs in the area of thermal hydraulics are to:

o Confirm safety margin in licensee analyses by performing audit calculations; o Investigate and resolve safety issues, e.g. , anticipated transient without scram, pressurized thermal shock, feed and bleed; o Evaluate the impact of design and operational-related changes; o Evaluate the safety of new, advanced reactor designs; o Examine the adequacy of operator guidelines and accident management strat-egies; o Understand operating reactor transient events and their broad implications; o Evaluate proposed changes to technical specifications and setpoints; o Answer questions from licensing boards, the Commission, Congress, and the public; and o Provide information on the risk-dominant accident sequences and the early phases of postulated severe accident scenarios. For PWRs this includes loss of all feedwater, station blackout, anticipated transient without scram, transient-induced loss-of-coolant accident (LOCA) with loss of injection, and small-break LOCA with loss of injection. For boiling water  ;

reactors (BWRs) this includes transient with loss of injection or loss of decay hgat removal, transient with stuck-cpen safety relief (ADS) valve and loss of injection or loss of decay heat removal, and anticipated trans- .

ient without scram.  !

1 As utilities gain experience in the operation of their nuclear plants, requests I for relief on technical specifications and operating limits are received, sup-ported by analyses. NRC may perform audit calculations to confirm the tech-nical basis for any improvements. Since various technical specification improvements may be proposed, their integrated effect on safe plant operation requires careful investigation.

1. 2 Products of Thermal-Hydraulic Research The principal products of thermal-hydraulic research are analytical tools (com-puter codes) to understand and predict plant response to deviations from normal operating conditions. The codes model the plant behwior by describing the processes of heat transfer and fluid flow that occur. The simulation of the expected behavior of the fluid systems in LWRs resulting from transients is normally performed with computer codes since it is too costly and unsafe to test the response to severe events using actual power plants.

1-3

l Whenever possible, physically based modeling is used in the development of the codes. Empirical modeling developed from thermal-hydraulic experiments is used to characterize important phenomena. Such experiments are conducted in integral and separate effect facilities. An integral facility is a scaled representation of a plant with all the major components present to provide information on overall system response and the interactions of different phenomena. The interrelationships of models that comprise the code are assessed using integral experimental facilities. In contrast, a separate ef fect facility studies a particular component or phenomenon in greater detail.

Each experimental facility has its own particular limitations associated with scaling and other design compromises that preclude direct extrapolation of experimental results to the full-scale plant. Therefore, the computer code provides the required link. Furthermore, the code allows different power plant transients to be studied without the need to embark on experimental programs for each separate case.

The code development and experimental programs operate according to a feedback process. As different scenarios are encountered or postulated and potential code deficiencies identified, particular experiments are run to obtain data necessary to establish the code accuracy or to improve the code. The inter-linkage of code development and experimental programs is such that one cannot exist without the other. Events involving new phenomena are periodically encountered in operating reactors for which the code applicability has not been verified. This requires a continuing separate effect and integral experiment capability to ansess the code's applicability to such scenarios.

This dual analytical and experimental approach was routinely used to help resolve safety and licensing issues over the past decade.

The codes have been formulated to be sufficiently flexible and accurate to analyze the range of postulated accident scenarios for the LWR designs in the United States, including the four vendors of nuclear steam supply systems:

o Westinghouse - two, three, and four loop; o Combustion Engineering; o Babcock and Wilcox; and o General Electric - BWR-2, 3, 4, 5, and 6.

The best estimate analysis of transients using thermal-hydraulic computer codes provides a means of understanding plant response. In this way, NRC is able to assess plant design and operation and develop appropriate rules and guides.

The proposed revision to Appendix K to 10 CFR Part 50 to allow for improved operation of plants is one example. The pressurized thermal shock rule, based on a program of experiments and computer code analysis, is a second example.

A third example is the analysis of decay heat removal in Babcock and Wilcox plants, which assisted in establishing the technical basis for continued opera-tion following the 1985 Davis-Besse loss-of-feedwater event. These are some examples of the use of research products to provide the technical bases for decisionmaking. Table 2 provides several examples of this dual approach.

1-4

Table 2 EXAMPLES OF DUAL EXPERIMENTAL / ANALYTICAL APPROACH TO RESOLVE ISSUES Issue Experiments Analyses Margin of Conservatism in LOFT, Semiscale, 20/3D, Analyses of Test Appendix K; Revision to TLTA, SSTF Facility Data and Appendix X (LOCA) Full-Scale LWRs Pressurized Thermal Shock Creare, Purdue, and TRAC-PWR and RELAPS Semiscale MSLB and SGTR. Analysis of Oconee UPTF, HDR, Finland, and Calvert Cliffs and H. B. Robinson.

Small-Break LOCA and Natural MIST, OTIS TRAC-PWR and RELAP5 Circulation in B&W Reactors Analysis of Data Small-Break LOCA in BWRs FIST, TLTA TRAC-BWR Analysis of Data Feed-and-Bleed Procedures Semiscale S-SR-1, 2, and SASA '.nalyses for Decay Heat Removal S-PL-3. LOFT LP-FW-1.

in PWRs Performance of Upper Head Semiscale S-UT series, COBRA / TRAC, TRAC-PWR and Upper Plenum Injection 2D/3D in W Reactors Performance of Upper Plenum CCTF, SCTF TRAC-PWR Injection in W 2-Loop Plants Small-Break LOCA with no ROSA-IV, LOFT LP-SB-3, RELAP5, TRAC-PWR, High-Pressure Injection Semiscale S-NH series SASA Analyses Liquid Holdup in Steam Semiscale S-UT-6, S-VT-8, RELAP5, TRAC-PWR, Generators During Small- S-LH-1, S-LH-2, ROSA-IV NOTRUMP Break LOCAs Steam Generator Tube ROSA-IV, Semiscale TRAC-PWR, RELAPS Rupture (SGTR)

Anticipated Transients LOFT L9-3, L9-4, RELAP5, RAMONA-38, Without Scram (ATWS) in Semiscale S-PL-7, FIST TRAC-BWR PWRs and BWRs Iodine Behavior Following MB-2, ORNL, Northwestern CITADEL, TRAC-PWR, SGTR University RELAP5 1-5

Table 2 (Continued)

EXAMPLES OF DUAL EXPERIMENTAL / ANALYTICAL APPROACH TO RESOLVE ISSUES Issue Experiments Analyses ECC Bypass and Penetration Creare, BCL, UPTF, CCTF, TRAC-PWR, RELAP5, LOFT SOLA Fluid Structure Interaction HDR, SAI K-FIX (FLX) on Reactor Core Barrel and Vessel Internals after LOCA BWR Containment Pressure MIT, GE, Livermore, PELE, SOLA Suppression Pool Loads Marviken Stability Marq'ns for BWRs DRESDEN, FRIGG NUFREQ TMI-2 Accident Semiscale S-TMI series, TRAC-PWR, RELAP5 MIST, OTIS Plant Transients TRAC-PWR, RELAPS ANO-II LOFT L6-7 Crystal River Semiscale S-PL-3 Ginna LOFT L6-8C St. Lucie Semiscale S-FS series Davis-Besse MIST, OTIS Steam Binding CCTF, SCTF TRAC-PWR l

Effects of Reactor Coolant LOFT L3-5, L3-6, TRAC-PWR, RELAPS Pump Operation During PWR LP-SB-1, LP-SB-2 Small-Break LOCA Semiscale S-SB-P1,

-P2, -P7 1-6

The accuracy of the thermal-hydraulic transient codes can only be estimated through a sound assessment program that considers the phenomena the code must be capable of representing and the ability of the various experimental facil-ities to provide the necessary information for assessment. Since test results from experiment facilities are not intended for direct application to full-scale plants, the codes are then the means by which plant behavior can be represented using experience gained from scaled experiment facilities.

The final products of thermal-hydraulic research are computer codes whose accu-racy and applicability have been assessed against sufficient and relevant data.

The assessment provides an evaluation of how accurately the code can reproduce and simulate the observed transient behavior of the coolant system in a variety of scaled test facilities. Given a proper noding representation and a set of initial and boundary conditions, the code should be able to accurately repro-duce the behavior of transients, which it has been designed to address, within a quantifiable error band. It can then be confidently applied to predict the consequences of postulated scenarios. This assessment is continuing through several cooperative agreements with industry and foreign governments, but is now more important since best-estimate analyses are expected to replace overly conservative analyses in the regulatory process.

Thermal-hydraulic research is distilled and made accessible in a comprehensive.

integrated, and usable format through assessed codes. User-friendly features must be included to make t'

• codes usable by nonspecialists. This includes comprehensive and underst. Jable visual displays of data, ability to interact during calculation of the transient, and fast-running time on the newest com-puters, which feature parallel processing. Fast-running computer codes are needed to reduce the cost of calculations and increase the timeliness of their use by regulatory personnel.

1.3 Historical Perspective of Thermal-Hydraulic Research The earliest nuclear power plants built were relatively small and remotely sited. The assurance of safety was primarily attained through the ability of the containment building to accommodate core melting that might occur and through the low oopulation density surrounding the sites. During the mid-1960s, however, plant designs increased in size from the order of 300 MWt to 3000 MWt and the sites selected were often closer to the electricity users, entailing higher population densities. The increase in power lowered the probability that the containment could accommodate postulated core melt sequences. Consequently, a greater emphasis was placed upon prevention of core melt. Attention was focused in three areas: (1) quality of design and construction to reduce the probability of ruptures of the reactor coolant system piping; (2) emergency core cooling systems to supply water to the reactor vessel in the event of a rupture; and (3) engineered safety features to provide containment cooling and remove fission products in the event of a l rupture. Research on thermal-hydraulic transients was expanded to provide a basis for judging the adequacy of the emergency core cooling system (ECCS) for preventing core melt. Criteria were established such that the core must remain adequately cooled for any size rupture up to complete rupture of the largest pipe in the reactor coolant system.

1-7

In view of.the relatively limited information existing in the early 1970s, con-servative criteria were formulated in Section 50.46 and Appendix K to 10 CFR Part 50 for license applicants to follow in performing safety analyses. There was a strong need to develop a sound technical basis to reduce the large uncer-tainties associated with LOCA analyses and to develop more appropriate regu-latory guidelines and criteria. Concurrently, therefore, the safety research effort under way was greatly expanded to provide information to reduce the uncertainties in understanding the phenomena that may occur during LOCAs.

The research was comprised of experiments to study the phenomena that may occur during transients and LOCAs and to obtain a physical understanding and description of the phenomena. Computer codes could then be based on informa-tion generated by the experimental programs to model the complex interactions that occur.

The thermal-hydraulic computer codes developed, TRAC-PWR, RELAP5, TRAC-BWR, COBRA, and RAMONA-3B, provide a best estimate of plant response for use by the NRC staff in performing safety analysis. Their development was focused ini-tially on the analysis of large-break LOCAs since these were the most severe transients postulated as the design basis. The WASH-1400 study of reactor safety, however, highlighted the need to focus increased attention on small-break LOCAs and transients with multiple failures or operator errors. This redirection in emphasis was hastened by the Three Mile Island-2 accident that occurred in 1979. The change in emphasis has been from large-braak LOCAs to less severe, but more probable, small-break LOCAs and transient events. Such trancients may include multiple failures, common mode failures, and operator errors as highlighted or revealed by probabilistic safety studies and actual operating experience. The focus on transients extended the range of events that the codes must be capable of analyzing and increased the requirements for code accuracy in modeling plant behavior.

Through the years, the demands on the codes have grown from the original empha-sis on large-break LOCAs to a myriad of transients and uses such as severe ac-cident sequence analysis, probabilistic risk assessment studies, and evaluation of operating experience. There has also been a shift in emphasis from conser-vative to best-estimate analysis since, especially for small-break LOCA and transient scenarios, conservative becomes difficult or impossible to define or achieve. As a result of code development, the codes can now calculate a spec-trum of transients and breaks in a best-estimate manner, including secondary-side breaks and control systems effects.

Many of the safety research programs that were started have achieved their planned objectives and are now complete. For example, experimental testing in the loss-of-fluid-test (LOFT) facility was conducted by the NRC from 1978 through 1983 and under international OECD sponsorship until 1985. The program is now complete and the facility is decommissioned. Experiments on BWR thermal hydraulics included the Two Loop Test Apparatus (TLTA) and the Full Integral System Test (FIST) facility in cooperation with the Electric Power Research Institute (EPRI) and the General Electric Company. This work is also complete.

Also completed are domestic experimental programs carried out at Semiscale, FLECHT-SEASET, MB-2, SSTF, and other facilities. A summary of these data is provided in NUREG-1230, "Compendium of ECCS Research for Realistic LOCA Analysis." Corresponding analytical efforts included the development of 1-8

TRAC-PWR at Los Alamos National Laboratory (LANL) and TRAC-BWR and RELAP5'at Idaho National Engineering Laboratory (INEL). The code development spanned a period of more than 10 years, and the major part of the work is now complete.

With the completion of various programs, the thermal-hydraulic research budget has decreased in successive years from a peak of about $97M in FY 1980 (1980 dollars) to about $12M in FY 1988.

The research conducted since the adoption of 10 CFR 50.46 and Appendix K in 1973 has resulted in greatly improved understanding of the phenomena that may occur in a LOCA. With the constraints imposed by Appendix K, a great deal of effort expended on LCCA analysis is based on assumptions that do not closely resemble the physical phenomena as they are now understood. Moreover, distor-tions created by the use of artificial conservatisms may be a net detriment to the overall safety of plant design and operation. Safety is best served through use of best-estimate analysis, with appropriate accounting of uncertainties.

Revision of 10 CFR 50.46 and Appendix K will allow best-estimate rather than artificially conservative analysis. This is expected to result in improved allocation of resources by the NRC and the nuclear industry. The revised rule may allow relaxation of plant operating limits; reduced number of plant trips; core loading schemes for improved fuel utilization; reduced surveillance requirements; lower rates of embrittlement of pressure vessels; and operation at increased power, within the limits of turbine and condenser capacity.

The economic benefit is expected to be significant, and both the NRC and the nuclear industry will be able to concentrate on more important safety issues.

1.4 Current Thermal-Hydraulic Research Needs Research plans cover the period of 1988-1992 and are aimed at maintaining and developing the technical bases for regulatory decisionmaking. This includes experimental programs and computer code improvement to fill remaining gaps in knowledge and refine the codes.

Previously, when a large effort was under way, a number of national labora-tories and research organizations were involved in carrying out the research.

Due to greatly decreased expenditures, an effort is now in progress to consol-idate a good part of the remaining thermal-hydraulic research resources at INEL.  :

Specialized capabilities that would be impossible to replace are also being  !

supported at other laboratories, and support is being devoted to a limited 1 number of university programs to assist in maintaining and improving thermal-hydraulic expertise. With the closure of domestic experiment facilities, an evaluation is needed on how best to retain continuing experimental capa-bility to respond to unforeseen future issues and to maintain a minimuni level of technical expertise.

The NRC has a number of specific needs for additional information in the area of thermal hydraulics, including experimental data and modeling, summarized as follows:

1. Analysis of transients in operating power plants (e.g., the June 1985 Davis-Besse event). The NRC requires the capability to quickly analyze transients using computer codes and to perform experiments to investi-gate relevant phenomena in a well-instrumented experimental facility.

1-9

1 l

l Particular emphasis is now being placed on safety issues associated with Babcock and Wilcox plants.

2. Study of small-break LOCAs and transienL in Babcock and Wilcox reactors.

.Much research has been directed at West 1tghouse and Combustion Engineering designs, and much of this is applicable to Babcock and Wilcox plants.

There are, however, two principal features of Babcock and Wilcox designs that require specific experiments aimed at their study: (a) vent valves.

between the vessel upper plenum and downcomer; and (b) once-through steam generators (OTSG) and auxiliary feedwater (AFW) spray. The experiment data are being provided through the Integral System Test Program, which includes MIST, and several new programs specifically investigating the OTSG and AFW spray. The analysis includes use of TRAC-PWR and RELAP5.

3. Evaluation of upper plenum injection, associated with six Westinghouse 2-loop plants in the United States.

4. Evaluation of PWR secondary system transients and breaks. Most research has been oriented toward primary system transients and breaks. Additional experiments and analysis of secondary system effects are in progress or planned for study.

5. Summarization of past research results. A large amount of information exists in the form of separate reports, papers, journal articles, and theses. Much of this work must be summarized to make it more readily usable to regulators. Major results will be incorporated into code applicability documents for major vendor designs.

6. Assessment and improvement of the thermal-hydraulic computer codes.

Significant improvements have been made since the previous versions of the codes were assessed. The current versions are now being assessed to quantify their accuracy, to correct software and logic errors, and to determine which models need to be improved. Improvements in the running time and user friendliness of the codes are also needed. Current calcula-tion speed results in significant computer costs and in some cases may limit the number of sensitivity studies that can be performed. Improve-ments in numerical methods and code structure are expected to yield significant improvements in running time. Additionally, improvements.are needed t: ssist the preparation of input and the interpretation of output.

7. Study of advanced LWR designs. Should new plants be proposed for construction, the NRC must be prepared to expeditiously evaluate the designs with a view to approving construction permits. Advanced BWRs are being designed in Japan in cooperation with General Electric. There is also a cooperative design being pursued by Japan and Westinghouse with  ;

marketing planned for the late 1980s. According to analyses, limiting l design conditions for the new BWRs are not determined by LOCAs but by j system transients. It may be expected that U.S. utilities may start ordering advanced reactors in the early 1990s. NRC computer codes will be kept up to date to analyze various limiting transients with different l sdety systems to provide the technical basis for the Office of Nuclear Reactor Regulation (NRR) in licensing of any advanced reactors.

1 1-10 l

l

8. Integration of thermal hydraulics, risk, and human reliability with special application to Babcock and Wilcox designs.

The procedure followed by the Office of Nuclear Regulatory Research (RES) staff for resolution of research needs and safety issues is as follows:

1. Identify and prioritize the cegulatory objectives or safety issues and establish the specific RES-NRR-AE00 coordination through the Regulatory Research Review Group (RRRG), as well as coordination with other offices, as appropriate. Ensure that specific tests, analyses, and reports are associated with specific regulatory issues or regulatory objectives.

Ensure that RES is kept informed of regulatory issues and objectives.

2. With the assistance of an NRC contractor, identify existing information available from domestic and foreign sources, assess the total research support needed, and formulate research program and schedule.

3. Ensure that the NRR or RES group using the research is also the group approving its initiation, continuation, and conclusions. Ensure that continuity of purpose and objectives is maintained, with allowance for new safety issues or priorities based on operating experience.

4. Contract for any needed research.

5. Monitor technical progress and schedule; integrate relevant infprmation by including results and conclusions of related research through the contractor.

6. Ensure that the research is technically sound and that progress is being made toward achieving the stated objectives. Provide for peer review of research results through formal review groups and publication of results in technical journals.

7. Summarize and disseminate results of completed research to NRR and other users. Regulatory integration is to be an explicit part of every research program. Provide summary topical reports when completed research achieves closure on regulatory issues or objectives, and transmit to NRR.

8. Incorporate the results in the regulatory process. including revising i rules and regulatory guides where appropriate. )

l

9. Periodically review with the Advisory Committee on Reactor Safeguards )

(ACRS) the regulatory objectives or safety issues, technical progress, I research results, and incorporation of information into the regulatory j process. l l

1.5 Interre p onships (NRR, AE0D, Regions, and F{SJ NRR has the principal responsibility for ensuring that the nuclear power plants it licenses are safe for operation. As such, NRR is the primary user of thermal- i hydraulic information and the principal determinant of research objectives and priorities. NRR generates much of its own required information and obtains information through technical assistance programs and through vendors and licensees via topical reports and responses to information requests. If NRR 1-11

decides that new research information is needed in the form of a computer code, modifications to a code, or experimental data, it will request that RES develop and provide the necessary information. Then, RES has the responsibility for providing validated codes and experimental data to ensure this validation.

The principal products that NRR uses are experimental data, computer codes, and results of code analyses. A requirement, therefore, is for assessed computer codes that are sufficiently accurate for the purposes for which they are used.

In addition, NRR requires a sound understanding of the important physical phenomena that may occur during transients in operating power plants and for specific safety issues. RES provides support to NRR for first-time use of the codes for new regulatory issues. NRR then applies the codes, as required, for subsequent analyses.

When urgent regulatory issues arise, NRR analyzes and assesses these issues.

Frequently, technical assistance resources are employed to assist in this ef-fort. If needed, NRR requests assistance from RES and its contractors to do follow-on studies after the initial urgency has passed. NRR funds its own technical assistance work, while RES funds follow-on work requested by NRR plus any additional research deemed necessary.

The Reactor and Plant Systems Branch of RES is responsible for planning and carrying out the thermal-hydraulic research necessary to obtain the information required by NRR. RES performs research (using contractors) according to tech-nical need established by RES staff and in response to NRR requests and ACRS recommendations. RES plans research with NRR sdvice and concurrence.

Resultant information must be summarized and transferred to NRR in su:h a man-ner as to be readily usable in decisionmaking and the regulatory process. RES holds periodic meetings with its contractors to review progress and plan future work. Results are documented in periodic research reports, in summary reports at the conclusion of major milestones or at the end of the project, and in research information letters.

The Office of Analysis and Evaluation of Operational Data (AE00) analyzes and evaluates operational safety data and proposes generic NRC actions that should be taken to maintain or improve public health and safety. AE0D provides major findings, suggestions, and recommendations to NRC program offices (including RES) for consideration and action. Thus RES and its contractors may do follow-on studies after the initial AE00 assessment.

The Regions execute established NRC policies and assigned programs relating to inspection, licensing, emergency response, et al. Two objectives for improved RES interactions with the regional staff are (1) to facilitate transfer of completed regulatory products and technology to inspectors and (2) to familiarize RES staff with operational problems.

The severe accident program uses thermal-hydraulic and other computer codes to calculate consequences of the risk dominant sequences. The Probabilistic Risk Analysis Branch in RES performs probabilistic safety studies using computer codes and the physical understanding developed through thermal-hydraulic re-search. St.ch studies provide a method of cataloging and arranging in order of significance the accident sequences representing serious threats to the fuel, 1-12

the reactor coolant syst 1., or the containment. The research reau W au ist NRR in evaluating industry risk assessment studies. This risk assessment y spective coupled with thermal-hydraulic infermation provides insights on saf t issues related to plant design, plant diagnoses for incident response and operator action, and procedures based on symptoms observed. These inc ig' cs can be used in determining priorities for thermal-hydraulic research. j The NRC is separately developing an accident management plan. Results d thermal-hydraulic research will play an important role in this plan, especia F y in the area of accident preverition and mitigation.

1.6 Coordination with U.S. Industry Where appropriate, the NRC cooperates with U.S. industry in carrying out re-search prcgrams. The establishment of Sint prcgrams is normally t.ought as a means of sharir.g the costs involved ard in broadening the technical scope of given projects. The industry groups most often involved are EPRI. the nuclear steam supply system (NSSS) ventiors, dnd utility owners groups. A current example of such cooperation is the Multiloop Integral System Test (MIST) program, which is sponsored by the NRC, EPRI, Babcock and Wilcox, and the Babcock and Wilcox Owners Group. The MIST program is managed by a prcgram management group consisting of representatives of each of the patties involved.

Similar past examples include FIST, FLECHT, SSTF, MB-2, and TLTA.

In general, the RES staff maintains contact with the industry on prcjects of interest. The industry representatives receive program reports and are invited to research 'nformation meetings. In the 2D/3D project. for example, the NSSS vendors normally attend the regular meetings and comment on the plan of work and analyses of results. Such comments are considered by RES in aefining the experiment specifications and plan. The NRC also sponsors the annual Water Reactor Safety Information Meeting that includes safety research results from 1 NRC contractors, industry, and foreign institutions.

Much thermal-hydraulic work of interest to NRC has been performed by EPRI and ,

the vendors. NRC receives all EFRI thermal-hydraulic reports and iias periodic l meetings to coordinate research plans and discuss research programs of mutual interest. NRR also receives information directly from the NSSS vendors and licensees via license apolications, responses to requests for information, and topical reports. RES intends to continue research collaboration with industry )

whenever common objectives are identified and agreement can be obtained on the conduct of joint programs.

1. 7 International Cooperation in Thermal Hydraulics RES places importance on cooperation with other countries in the area of ther.dal-hydraulic research. Such m eration can be bilateral or multilateral. The NRC has obtained bilateral Cannal-hydraulic exchange agreements with nearly all noncommunist block countries with nuclear power programs. The most important of these include France, Federal Republic of Germany, Italy, Japan, Sweden, and the United Kingdom. These agreements form the basis of the International Code Assessment Program (ICAP). Under this program, the NRC has provided its advanced, i best-estimate thermal-hydraulic codes TRAC-PWR, RELAPS, and TRAC-BWR in exchange '

for assessnent and applications information and experimental data. The NRC 1-1%

serves as coordinator of this joint international project on the quantitative assessment of NRC codes. The ICAP takes in essentially all major nuclear thermal-hydraulic experimental facilities in the world.

Through a bilateral agreement with Japan, the NRC has access to all ROSA-IV results. The trilateral 2D/3D program between the NRC, Federal Republic of Germany (FRG), and Japan is a major joint project investigating large-break loss-of-coolant accidents. The OECD LOFT project is a program of experiments in the LOFT facility. The project has 10 member nations, and the U.S. member-ship includes the NRC, EPRI, and the Department of Energy (DOE). These projects are governed by boards consisting of representatives from each member nation.

The NRC participates in the Committee on the Safety of Nuclear Installations (CSNI) of the OECD Nuclear Energy Agency. Within CSNI, the subject of thermal hydraulics is covered by the Principal Working Group 2 on Transients and Breaks.

This group provides a forum for multilateral discussion and information exchange.

It also provides a r.ethod for carrying out international standard problem exercises. Information is distributed to interested NRC staff and contractors through trip reports. In addition, the International Programs group in the Office of Governmental and Public Affairs receives copies of all CSNI documents directly.

1.8 Future Plans for Thermal-Hydraulic Research 1.8.1 Current and Future Users It is anticipated that the needs of NRR for thermal-hydraulic research will continue. There will still be a need to audit industry submittals of analyses related to regulatory concerns. Final versions of the major thermal-hydraulic codes (TRAC-PWR and RELAPS) are to be prepared by June 1989. They should be capable of auditing forseeable industry submittals with reasonable accuracy and speed. These final versions will also have user-oriented features to facili-tate use by the NRC staff. The codes will also be available help NRR resolve specialized issues as they arise.

Another activity for which NRR has requested assistance is review of industry computer codes, for ECCS rule submittals as well as for other purposes, with emphasis on code quality assurance and determination of code uncertainty.

Finally, in response to NRR requests, synthesis reports will be prepared on completed research to aid in current regulatory activities. These synthesis reports are designed to achieve closure on thermal-hydraulic technical issues for which extensive research has already been completed, but for which conclu-sions are not readily identified because results are scattered among may reports.

Recent examples of such reports are on feed-and-bleed phenomena and on natural circulation.

The Office of Analysis and Evaluation of Operational Data (AE00) appears to be a future user. The Technical Training Center has expressed an interest in the nuclear plant analyzer (NPA) to supplement their training programs geared to training simulators. Possible link-up of the NPA to these training simulatcrs is being investigated in order to improve the software capabilities. The NPA is also used in Incident Response Center drills.

1-14

The NRC regional offices have similar needs, and these will be explored in the future. Currently, for example, a handbook on water hammer is being prepared for regional inspectors.

RES will continue to use thermal-hydraulic codes to support major research programs. Accident management programs will use codes for front-end analyses and for providing conditions during the core melt portion of postulated severe ,

accidents. <

l With the advent of advanced LWR designs, it is anticipated that new code and experimental capabilities may be required to resolve regulatory concerns.

These needs will be reviewed periodically to determine any new programmatic requirements.

1.8.2 Branch Structure and Research Emphasis r

In the past, the RES branch responsible for administering NRC thermal-hydraulic research was structured into analytical and experimental sections. Currently, i it is structured into one section that-manages research, while the other sec-tion manages application of completed research to resolve regulatory issues.

Applications include use of a multidisciplinary approach (PRA, human factors, equipment reliability) as needed to ensure total resolution of the issue.

Recent examples of such an approach include completed work on the pressurized thermal shock issue and the recent program on development of an integrated methodology to evaluate safety of B&W plants. Applications also include utili-zation of thermal-hydraulic expertise for accident management.

1.8.3 Major Programs To Be Completed by the End of CY 1991 Most major current programs are planned for completion by the end of CY 1991. '

These include the International Code Assessment Program with 14 participating l countries; the 20/30 program with the FRG and Japan; the ROSA-IV program with .'

l Japan; the MIST and the once-through steam generator programs with U.S. industry; and final versions of all major thermal-hydraulic codes.

1.8.4 Continuing Baseline Programs Between now and 1992, the emphasis and resources on code development and i assessment will decrease. Nevertheless, a base program effort will be maintained in this area to make necessary code improvements based on new information, to retain a cadre of Experts, and to address new issues as necessary. Basic studies will continue, as needed, to assist in issue resolution. The nuclear plant analyzer will be the major vehicle for making thermal-hydraulic software products easily accessible to users within NRC and other users. The thermal-hydraulic Technical Support Center will continue to be used to assist NRC in efficiently accomplishing both immediate and long-term goals.

1. 8. 5 New Programmatic Initiatives New programmatic initiatives that are starting up or being planned include the following:

1-15

1. An international consortium to assist in developing and assessi$ tne final planned versions of TRAC-PWR and RELAP5.

2. Review of existing software that can be used to perform accident management studies and a plan for integrating and improving such software into a user-friendly package.

3. A joint program with B&W, the B&W Owners Group,.and EPRI to investigate the performance of once-through steam generators under accident conditions, especially starting from full power conditions and including the use of auxiliary feedwater.

4. Expanded use of the NPA for AE0D use in both the Technical Training Center .

and the Operations Center. I I

5. Support to regional inspectors through handbooks and improved training. I l

6. Investigation of procedures to improve the performance of NRC training l simulators. i

7. Restructure of work orders and contracts to include a multidisciplinary approach to issue resolution.

8. Use of results from MIST and the University of Maryland facility to test scaling concepts for use in future facility design and operation.

1.8.6 RES/NRR/AE00 Task Group on Future Thermal-Hydraulic Research:

Recommendation on Thermal-Hydraulic Codes Thermal-hydraulic computer code development is coming to a logical and orderly conclusion. Five thermal-hydraulic codes will be maintained and improved by RES in the future consistent with user office requests. These codes are RELAPS, TRAC-PM , TRAC-BWR, COBRA-NC, and RAMONA-38.

The reactor systems safety senior research program steering group appointed a task group to make recommendations on thermal-hydraulic code plans. The task group, composed of two members from RES, two from NRR, and one from AE00, met twice in April 1988. Their recommendations were presented to the senior steering group in May 1988. The task group cor.sidered both development and maintenance plans for the codes. Code development is defined as completing planned efforts for model improvements. Code maintenance is defined as correcting errors, providing user conveniences, improving modeling (as needed), and developing input decks (as needed).

The task group recommendations were:

1. The NRC needs independent expertise and analysis capability. Both PWR and BWR analytical capability is required.

2. The codes by themstives are not useful. They ust be accompanied by plant input decks, including balance of plant modeling.

1-16

i l

3. Complete TRAC-PWR and RELAPS through ICAP Consortia in 1989, ensure completion of 3D-2 STEP Algorithm in TRAC-PWR.

4. Maintain TRAC-BWR for active use. Deemphasize TRAC-BWR only if RELAPS is upgraded to have comparable best-estimate BWR capability and TRAC-BWR input decks are converted to RELAP5.

5. Maintain COBRA-NC for analyzing containment superheat (NRR).

6. Maintain RAMONA-3B for active use.

7. Consolidate codes to one or two institutions.

8. Provide resources to maintain codes overall, with periodic emphasis shifting (among codes) based on regulatory needs. In this way, some codes could temporarily by "on the shelf" until a need for their use arises from one of the user offi:es.

9. Conduct a critical review of nuclear plant analyzer practicability for headquarters staff.

In a June 7, 1988 letter from D. Ward to L. Zech, the ACRS endorses this general plan for finishing thermal-hydraulic code development.

1-17

2. THERMAL-HiORAULIC ISSUES AND REGULATORY NEEDS The following are a current set of thermal-hydraulic issues being addressed through research activities and the regulatory needs they are to meet.

2.1 ECCS Rule Revision 2.1.1 Issue Section 50.46 of 10 CFR Part 50 requires that calculations be performed to show that the emergency core cooling system (ECCS) will adequately cool the reactor core in the event of a loss-of-coolant accident (LOCA). Appendix K to Part 50 sets forth certain required and acceptable features that evaluation models, used to perform these calculations, must contain. The results of these calcu-lations are used to determine the acceptability of the ECCS performance.

In many instances, these calculations result in technical specification limits on reactor operation (e.g. , peak local power) in order to comply with the 2200 F cladding temperature limit and other limits of Section 50.46. These limits restrict the total power output and optimal operation of many reactors in terms of efficient fuel use, maneuvering capability, and surveillance requirements. Removing unnecessary restrictions on operation would allow increased U.S. electricity production, worth several hundred million dollars a year, without detriment to public health and safety. In some cases, removal of these restrictions would result in benefits to safety. For example, diesel generator reliability would be improved by requiring less rapid start times.

Also, by allowing a higher neutron flux at the core center, less neutron leak-age at the vessel wall would result, thus reducing vessel weld embrittlement and the likelihood of crack propagation due to pressurized thermal shock.

NRC, 00E (including the AEC and ERDA), U.S. industry, and foreign research on ECCS performance now provides a technical understanding showing that the existing ECCS rule restrictions are too stringent. The NRC staff has, there-fore, proposed to revise the ECCS rule to allow use of this completed research, as an alternative to the currently required features of Appendix K, when calculating ECCS performance. The proposed revision would allow realistic calculations, based on the best information available, to be used to show that the ECCS performance meets the criteria of Section 50.46. An accompanying evaluation of the uncertainty of the calculation would be r q uired, and this evaluation would be considered when comparing the results of the calculations to the criteria.

A notice of proposed rulemaking was published for comment in the Federal Register (52 FR 6334) on March 3, 1987. A draft regulatory guide and a draft report, "Com.oendium of ECCS Research for Realistic LOCA Analysis," NUREG-1230, summariz Wg the ECCS research supporting the rule revision were also released for public comment. The NRC staff has completed review of the comments on the rule, regulatory guide, and NUREG-1230 in preparation for recommending a final rule to the Commission and publishing final versions of the compendium and regulatory guide.

2-1

l 2.1.2 Regulatory Need Several echnical activities are under way to support the final rulemaking.

T b buik of the research, summarized in NUREG-1230, that supports realistic calculat),ns of ECCS performance during a LOCA has been completed for several years. Completion of the last of this experimental research will be conducted under the 20/3D program (see Section 3.1.1). The remaining experiments include full-scale tests of ECCS bypass phenomena and experiments investigating the behavior of certain plants (i.e., Westinghouse 2-loop plants) with ECC systems that inject water in the upper plenum of the reactor vessel. The results of these experiments are not expected to change our current understanding of ECCS performance but will be used to reduce the uncertainty in the calculated plant response. While realistic or best-estimate computer codes have been used for a number of years to calculate ECCS performance, the requirement to quantify the uncertainty of the calculations is relatively new. As part of the effort to revise the ECCS rule, RES initiated a task to quantify the uncertainty of NRC-developed computer codes used for calculating the plant response to the LOCA and other transients (see Sections 2.2 and 3.1.8).

The final ECCS rule revision is scheduled for completion in the summer of 1988.

At that time, a final regulatory guide and compendium are also scheduled for publication. The code uncertainty methodology was completed in 1987 and the preliminary uncertainty estimates for calculations of a LOCA in selected plants was completed in 1988. This will provide a demonstration of the feasibility of uncertainty quantification and peer review of the technology prior to imple-mentation of the final ECCS rule revision. Completion of the final 20/3D experiments is not scheduled until 1990. These results will be used to reduce the uncertainty of calculations and can be used in the application of the rule.

2.2 Code Scalability, Applicability, and Uncertainty (CSAU) Evaluat1on 2.2.1 Issue The NRC has proposed a revision to the Section 50.46, "Acceptance Criteria for Emergency Core Cooling Systems in Light Water Nuclear Power Reactors," issued in 1974. It requires that:

1. An acceptable evaluation model have sufficient justification to show that the analytical technique realistically describes the behavior of the reactor system during a LOCA, and

2. The uncertainty of the calculations be estimated and accounted for when comparing calculated results with temperature limits and other criteria so that there is a high probability the criteria would not be exceeded.

2.2.2 Regulatory Need There is a regulatory need for independent audits of industry submittals in order to ensure that these submittals meet the revised acceptance criteria.

This entails a methodology to:

1. Determine whether or not the code has the capability to model and scale  ;

processes from test facility to full-scale nuclear power plants. l l

2-2

1 l

l 1

2. Determine whether or r.ot the code has the capability to address a LOCA scenario, and

3. Quantify uncertainty of a code to calculate a particular parameter j important to a LOCA. l l

The program discussed in Section 3.1.8 addresses this need. l

2. 3 Evaluation of Operational Performance of B&W Plants Versus Plants of Other Vendors 2.3.1 issue l Operating events at some B&W plants raised concerns over the sensitivity of B&W l Plants to operational transients and led the NRC to conclude that there is a

! need to reexamine the basic design of B&W plants. The B&W Owners Group (BWOG) has taken the lead in this evaluation. As part of the NRC independent assess-ment, RES sponsored a study that developed an integrated methodology to assess the operational performance of these plants and that applied this methodology on a trial basis. The study used both thermal-hydraulic and human factor methods and compared results with information used in probabilistic risk assessment (PRA). The cases studied were a complete loss of all feedwater (LOFW) transient, a small-break LOCA, and a steam generator overfill with emergency feedwater.

The objectives of this analysis were to:

1. Evaluate the extent to which the plant systems and/or operator actions are effective for transients in B&W plants.

2. Compare differences in the operational performance capabilities and limitations of a representative Combustion Engineering (CE), Westinghouse, and B&W plant based on the results of the assessment performed to satisfy the first objective.

3. Evaluate the significance of the findings in terms of risks identified in existing PRAs for the selected plants.

2.3.2 Regulatory Need NRC needs the capability to independently assess plant and operational changes ,

recommended by the BWOG as a result of their reassessment of B&W-designed plant I safety. A capability is needed to demonstrate that individual recommendations  !

provide improved safety (or at least do not degrade safety) as _well as to )

demonstrate that the total of the recommended changes does actually provide a significant improvement to total plant safety or level of risk. A key element in this reevaluation is operator reliability, particularly since transients in l B&W plants involving the steam generator proceed more rapidly than in other PWR I designs, thus potentially placing additional requirements on operators of B&W plants.

l 1

1 2-3

2.4 Response of B&W Once-Through Steam Generators to Operational Transients and Accident Conditions 2.4.1 Issue-There is a lack of experimental data to validate the B&W once-through steam generator (OTSG) models used in NRC best-estimate codes such as RELAPS and TRAC-PF1 for B&W transient analyses. First, data are needed on the thermal-hydraulic aspect of wetting of tube walls and flooding at tube support plates by auxiliary feedwater (AFW) in a B&W OTSG. The other data need is on the thermal-hydraulic aspect of 0TSG performance for transients that are initiated from full power (e.g., steam line break, feedwater line break).

2.4.2 Regulatory Need A thermal-hydraulic research plan for B&W plants (NUREG-1236) was prepared in response to a memorandum from the Director of NRR to the Director of RES on B&W reactor testing needs, date( 10/31/84, requesting RES to obtain thermal-hydraulic operational transients data for B&W plants at the MIST facility. The NUREG presents a plan for thermal-hydraulic research for B&W-designed reactor systems.

It describes the technical issues, regulatory needs, and the research necessary to address these needs. The plan also discusses the relationship between ongoing (NRC as well as EPRI) and proposed research and provides a schedule to complete the required work.

2. 5 Response of B&W Plants to Small-Break LOCA 2.5.1 Issue The Three Mile Island Unit 2 (TMI-2) accident in March 1979 was a small-break LOCA that led to (. ore damage. Following the accident, the NRC reviewed the analytical predictions of feedwater transients and small-break LOCAi, for the purpose of ensuring the continued safe operation of all operating reactors.

As a result of the review, NSSS vendors and fuel suppliers were required to provide experimental verification of the various modes of single phase and two phase natural circulation predicted to occur in each vendor's reactor during small-break LOCAs. This requirement is delineated in hdREG-0737, "Clarification of TMI Action Plan Requirements" (task action plan Item II.K.3.30, "Revised small-break LOCA methods to show compliance with 10 CFR Part 50, Appendix K").

The experiment data on small-break LOCAs from the Semiscale and the LOFT facil-ities were applicable to the Westinghouse (W) and CE-designed reactors but not to the B&W design. The B&W OTSG presents a significantly different configura-tion from the U-tube steam generator employed in the W and CE designs. Neither the Semiscale nor the LOFT facilities modeled the B&W hot leg configuration of the OTSG and, as a result, did not simuisto the appropriate natural circulation conditions. In particular, there was uncertainty about the effects of two phese flow, noncondensible gases, and the validity of the boiler-condenser mode of heat removal. In addition, the hydraulic stability, effects of high point vents, and internal reactor vessel vent valves were items of interest.

2-4

2.5.2 Regulatory Need A memorandum from the Director of NRR to the Director of RES, dated December 30, 1981, requested the conceptual design of an experimental facility and identified l data needed to confirm the thermal-hydraulic behavior of B&W plants for small- l break LOCA.

2.6 Response of W/CE Plants to Small-Break LOCA ,

2.6.1 Issue A number of issues related to W/CE plants during small-break LOCAs have been identified and are currently being studied using results or planned test results from the ROSA-IV experimental program. These issues are described below.

1. Core Level Depression and Liquid Holdup Phenomena During the course of the Semiscale experimental program, a transitory core heatup was shown to occur earlier than expected (based on water inventory). Core uncovery resulted as liquid was held up in the steam generator U-tubes, and this formation of a loop seal produced a hydrostatic pressure imbalance such that the core liquid level was depressed. (The liquid was forced from the core into the downcomer.) However, when the loop seal cleared and the break began to pass a high void fraction flow, the liquid inventory moved back into the core and terminated the early core heatup. Subsequently, the core uncovery phenomena were extensively studied and confirmed in another Semiscale experiment (S-LH-1), with improved instrumentation and representative boundary conditions.

Core level depression and liquid holdup phenomena have also been confirmed in the large-scale test facility ROSA-IV. The issue is whether the phenomena will occur in full-scale W/CE plants and whether excessive core heatup will occur as a result of these phenomena.

2. Pressure Vessel Lower Plenum Instrument Line Rupture A seismic event that leads to the rupture of one or several in-core instrument tube lines (installed in the bottom of the pressure vessel and plumbed to a seal table) may be postulated. A W study has indicated ', hat no more than three such lines (older W PWRs have 58 instrument lines) can fail during a seismic event. The issue Ts whether there are conditions under which the ECCS cannot keep the core covered after the rupture of several instrument tube lines.

3. Small-Break LOCAs With No High-Pressure Injection Flow The issue of small-break LOCAs with no high pressure injection flow is that a  :

condition of inadequate core cooling might exist as a result of the loss of coolant before the primary system pressure decreases to the accumulator and low-pressure injection system (LPIS) pressure setpoints. If extended dryout of the core should occur, fuel damage may result before the fluid from the LPIS can reflood the core.

2.6.2 Regulatory Need l Resolutions of the above issues will allow the NRC to evaluate plant safety system design and operator guideline changes made to prevent the core from ,

l I

2-5 l l

being uncovered or core heatup during small-break LOCAs. Large-scale data obtained from the ROSA-IV experimental program together with Semiscale test results can be used by the NRC to resolve these issues. The ROSA-IV test  ;

results are being used (1) to assess the capability of the TRAC and RELAP codes '

to predict plant behavior, and (2) to understand the mechanism of the phenomena associated with these issues.

2.7 Long-Term Cooling After a LOCA 2.7.1 Issue  !

Long-term core cooling following a large cold leg break, is obtained by boiling water in the core. The resulting steam flows through the hot legs, the steam generators, the crossover pipes from the steam generators to the reactor coolant pumps, and out the break. Makeup water is provided to the core via the head of water maintained in the downcomer upper annulus due to ECCS flow, while excess ECCS water flows directly out the break.

The crossover pipes connecting the steam generators to the reactor coolant pumps are voided during the initial portion of many LOCAs. These pipes, which contain the loop seals, are then assumed to remain voided for the rest of the accident scenario. This assumption rnay be based on the presence of weirs in the discharge pipes of the pumps, on the elevation of the pumps above the remainder of the cold leg pipe, on the flow of steam through the crossover pipes, and/or on the restricted flow areas in che pumps. Any of these may prevent water from flowing from the cold legs into the crossover pipes.

In certain cases, this assumption may be incorrect, and the weirs in the bottom of the pump discharge pipes will not prevent water from refilling the loop seals. In addition, steam flow may be insufficient to prevent backflow of water several hours after initiation of the LOCA, and it will not be sufficient if the reactor power level is low before the accident. Under these conditions, water may refill the loop seal and block the free exiting of steam from the i reactor coolant system, thus raising a question regarding the correctness l of the design basis assumption. This is referred to as loop seal plugging.

Filling of the loop seal would block the steam flow path, depressing the vessel water level, and lead to core uncovery. Thus, unanticipated and unanalyzed pressurization and core uncovery in a PWR may occur following an initially successful operation of the ECCS.

A small decay heat generation rate may be more serious than a large one under some of these circumstances. This is opposite to the conservative position that is always applied to the early portion of LOCA analyses. Low decay heat will result in a low steam generation rate, which could allow water to enter the crossover piping. It also could result in quiescent boiling with an accompanying lack of level swell and carryover of water droplets. The effect would be to make it more difficult to keep fuel cladding temperature acceptably low in the uncovered portion of the core.

The decrease in water level in the core could continue until a sufficient vent path existed to relieve steam, or some other means of heat removal was estab-lished. In CE and W plants, the level could drop until the bottom of the U-shaped crossover pipe cleared and steam was able to flow toward the break.

If all other conditions were the same, the problem would be greater in }{ PWRs 2-6

I because the crossover pipes extend to a lower elevation relative to the core than they do in the CE plants. In B&W plants, the reactor vr sel vent valves would open and provide a path for steam flow. This would prevent significant pressurization of the upper vessel relative to the cold legs, thus eliminating i the issue for B&W plants.

Moderate core uncovery while decay heat is high (as would be the case within a ,

few hours of shutdown from 100 percent power) is not of significant concern if '

one is investigating realistic plant response. The boiling process is relatively violent, and there is significant frothing and liquid carryover. This is suffi-cient to keep the upper portion of the core cool, even for a condition where the core collapsed liquid level is approaching the center elevation of the core.

Core uncovery with a small heat generation rate, such as a long time after shut-down or shutdown from less than 100 percent power, is of more concern. A stable l collapsed liquid level in the vicinity of mid-core with a froth height of only l

a few inches could result in high core temperatures in upper regions of the core.

l

! 2.7.2 Regulatory Need The concern is a potential failure to maintiin adequate heat removal following initial recovery from a cold leg break LOCA. (Initial recovny constitutes satisfactory operation of the ECCS to either reflood the ccre or to maintain the core in a water-covered cond' tion during and/or immediately following blow-down of the reactor coolant system.)

Design-basis analyses of LOCAs must be in compliance with Section 50.46 of Title 10 of the Code of Federal Regulations (10 CFR Part 50). Criterion (5) of Section 50.46 is the requirement to maintain long term cooling following successful initial operation of the ECCS. Design-basis analyses generally are terminated when the analyst establishes that maximum temperatures have been reached, that the core i3 fully covered by water, or that this will occur, and that temperatures are lower and fuel damage is less than regulatory limits.

The remaindar of the accident has not been subjected to the close scrutiny that was applied to the early stage of the accident. This is also true of extensive tests conducted to investigate LOCA behavior. Virtually all tests have been terminated before the time of concern here and therefore have not directly addressed the concern. Thus, there may be a lack of suitable test data to verify existing analysis techniques that should be considereo for investigating the concern.

Currently NRR is determining the most appropriate way for the industry to address the problem. RES has not received an official NRR user need to provide technical support. However, we believe that since this relates to the current ECCS rule revision activity (see Section 2.1) RES should continue exploratory analysis. Through the Technical Support Center (see Section 3.2.1), initial scoping analyses have been performed. Preliminary findings indicate that at high decay heats with high steam flow (generally occurring within 1 day following a LOCA from full power operation), loop seal plugging is unlikely. If steam flow is too low to prevent plugging, it is not likely to clear subsequently.  ;

Uncertainty in core froth level calculations is significant, and internal bypass  :

flow significantly affects core level depression. l i

2-7 I l

l

2.8 Boron Mixing in BWRs After an ATWS 2.8.1 Issue Following an articipated transient without scram (ATWS) in a BWR, the reactor i coolant recirculation pumps are tripped automatically, usually by a high pressure signal. The reouced natural circulation core flow results in a decrease in reactor power because of increased voiding in the core. Core power is further reduced by procedures that require that the water level in the vessel be lowered to further reduce core flow. Heat is removed from the system by steam venting to the suppression pool through relief valves. The reactor power is ultimately shut down by the injection of boron into the vessel. This injection of boron is a slow process and must be initiated early enough so that the reactor is shut down before the suppression pool is overheated, which could result in j overpressurization of the containment.

There is uncertainty concerning the behavior of the boron solution, which is more dense than the reactor coolant water. The dense boron may not mix well with the reactor coolant under the low flow conditions following an ATWS and may settle in the lower plenum. ATWS procedures, therefore, call for the operator to increase the coolant level, and thus increase flow, after sufficient boron has been injected, to ensure that the boron is circulate 1 to the core and shuts down the reactor.

2.8.2 Regulatory Need Current computer codes used by NRC to calculate the behavior of BWRs during an ATWS (TRAC-BWR and RAMONA) include the capability to account for boron injection and the influence of boron on core om.;er. However, these computer codes assume that the boron is perfectly mixed with the reactor coolant and cannot account for separation and settling of the more dense boron solution. Thus, significant uncertainty in the calculated behavior of the plant during an ATWS exists. A research program is air.ast completed at the University of California at Santa Barbara (see Sectien 3.2.4) to study boren mixing to improve the computer codes and reduce the uncertainty in understanding plant behavior during ATWS.

2.9 Primary System Depressurization to Avoid Direct Containment Heating 2.9.1 Issue SNrce terms used in current regulations do not accurately reflect the best onderstanding of the phenomena leading to radioactivity releases. The question of how best to prevent severe accidents or to mitigate their effects has not been fully answered. Direct containment heating (DCH) is one of the subissues included under the general issue of severe accidents. In accident sequences such as those initiated by station blackout or small-break loss-of-coolant events, core degradation and melting may take place while the reactor coolant system remains pressurized. If the reactor vessel is breached in such a condition, the molten core debris could be ejected under high pressure. The ejected debris might be dispersed out of the reactor cavity into the surrounding containment volume as particles that would transfer thermal energy to the containment atmosphere.

Metal contents of the debris could react with oxygen and steam and generate

, chemical energy, further heating and pressurizing the containment. The l pressurization may lead to an early containment failure. DCH and early contain-l ment failure may be avoided by depressurizing the reactor coolant system.

2-8

2.9.2 Regulatory Need NRR needs to develop guidelines for severe accident management procedures. The procedures will be developed based on today's knowledge of severe accident phenomena with all attendant uncertainties. These guidelines will be used by licensees in developing their severe accident management plans so as to prevent and/or mitigate consequences of severe accidents.

I 2.10 Downcomer Fluid Temperature Under Pressurized Thermal Shock Conditions 2.10.1 Issue l

During some transients in a PWR, emergency core cooling water is injected into the cold leg of the reactor while the fluid in the vessel is at high pressure.

There were concerns that this system overcooling transient may result in "thermal shock" of the reactor pressure vessel.

2.10.2 Regulatary Need In order to determine the effect of size on thermal mixing, experiments were performed in this country and abroad, in facilities of various scales, that is, in CREARE (1/5-scale), IVO (2/5), PURDUE (1/2), CREARE (1/2), HDR (1/2), and in UPTF (full-scale). Two computer programs, REMIX and NEWHIX, were developed to model thermal mixing processes in the cold leg and in the downcomer. The as-sessment of these codes against test data obtained in facilities listed above demonstrate that they can be used with confidence to model thermal mixing for conditions of interest to reactor applications. The results of this 5 year experimental and analytical effort are being summarized in a topical report to be issued in FY 1988 by the University of California at Santa Barbara.

2.11 Water-Hammer Diagnosis 2.11.1 Issue Based on Licensee Event Reports and published compendia of events, 188 water-hammer events at nuclear plants have been reported to the NRC. The majority of these events have had little or no impact on plant operation and have been traced to well understo,d and readily analyzed causes such as rapid closure of a valve or startup of a pump. Certain events due to condensation-induced water hammer are more complex physically and harder to diagnose. These condensation-induced water-hammer events are the ones primarily responsible for significant damage or impact on plant operation. Interpretation of such events is of current concern.

2.11.2 Regulatory Need When an inspector arrives at a plant as a result of a damaging event, he/she may face any of a large number of potential situations with respect to water-hammer events. Plant personnel may have already diagnosed the event or they may still be trying to determine what happened. The event itself ma/ be quite simple to understand or physically complex. The exact time of occur-ence may be known, or damage may have gone undetected for ar unknown period of time. Rather than consider each possible situation separately, a general procedure is needed to assign cause and effect durirg the event. A diagnostic guidebook would be useful 2-9

wheaever a particularly complex or severe water-hammer event has occurred in a power plant. The NRC inspectors can ascertain the probable loads sustained by apparently undamaged plant components so that safe plant restart can be ensured.

2.12 Safety of Advanced LWRs 2.12.1 Issue The nuclear industry has initiated programs for the development and construction of advanced LWRs. In general, the advantages sought are simplification and standardization. Because current plant designs have been developed by adding on systems to meet changing safety requirements, their designs may be unneces-sarily complex. Advanced LWRs start with a new concept (with the goals of inherent safety, simplicity, maximum use of passive safety systems) and use operating experience gained in current plants. The concepts include large units up to 1000 MWe and smaller units of 400 to 600 MWo. The industry goal is to use proven technologies and extensive R&D so that commercial plants can be built without the need for expensive, time-consuming demonstration plants.

2.12.2 Regulatory Need One of the approaches NRC will use to evaluate the safety of advanced LWR concepts will be thermal-hydraulic analysis of system response to e wide range of transients using NRC best-estimate codes (RELAP5, TRAC-PWR, and TRAC-BWR).

To use these codes with confidence, it is necessary that their applicability to the transient conditions and phenomena and to the reactor component designs be established. It may be that existing models in the codes are applicable to the new system concepts and that the range of conditions for which they have been validated include those expected for applicable transients in the advanced LWRs.

If this is not the case, it may be necessary to establish the applicability of the codes using industry-supplied data (or data generated by NRC research programs) or even to modify and validate these codes before they can be used for independent safety analysis of advanced LWR designs.

2.13 Operator Guidelines and Improved Simulators 2.13.1 Issue The operator is recognized as an important factor in safe recovery of plants from certain operational transients. Guidance for an operator in diagnosing the transient type taking place and determining the appropriate corrective actions needed to stabilize and recover the plant to a safe condition is provided in emergency operating procedures. The correctness and optimization of these procedures are improved as the analysis basis for them more closely approaches the actual response of the plant and as the range of plant transient results available is increased. These analyses form the bases both for formulation and later validation of these procedures. Performance of operating crews, under the stress of an actual reactor transient, can be materially improved by accu-rate, well-founded procedures, and through practice on plant simulators that faithfully represent both the plant / operator interface as well as the plant transient characteristics.

2-10

.~

2.13.2 Regulatory Need The NRC needs a sound technical basis for promulgating requirements and guidelines related to nuclear plant operator guidelines and training. An ex-tensive data base that demonstrates both the practicality and effectiveness of possible isoprovements would benefit both NRC and industry in this effort. The NRC needs an evaluation of the practicality of approaches to provide improved plant analyses for use in procedure development and validation, operator training, etc. In addition, the NRC needs data to support implementation of these require-j ments and guidelines, e.g., improved safety, possible cost benefits of implementa-tion versus cost of plant shutdown and repairs, and improved plant availability.

Research support will be developed in the programs described in Sections 3.2.1, ,

3.2.2, 3.2.3, and 3.3.3 of this program plan, which will satisfy these needs I that relate to plant thermal-hydraulic behavice.

2.14 Containment and Balance of Plant 2.14.1 Issue Section 50.46 of 10 CFR requires that safety-related electrical equipment should remain functional during and following design basis events to ensure (1) the integrity of the reactor coolant pressure boundary, (2) the capability to shut down the reactor and maintain it in a safe shutdown condition, and (3) the capability to prevent or mitigate the consequences of accidents that could result in offsite exposures defined in 10 CFR Part 100 guidelines. The environ-mental qualification of the equipment depends on heat transfer and thermal-hydraulic phenomena occurring within the containment after the reactor blows down in a LOCA. The interim position of the NRC staff on condensing heat transfer coefficients after a superheated steam blowdown during a steam line break is described in NUREG-0588. A report from the Brookhaven National Laboratory, dated June 23, 1987, reviews the current state of the art in condensing heat transfer coefficients and recommends improvement of such coefficients currently used in licensing audit calculations.

2.14.2 Regulatory Need The condensing heat transfer coefficients (under superheated steam blowdown conditions) used in licensing audit calculations are conservative. The need exists to quantify existing margins. In order to quantify these margins, it is necessary to improve applicable correlations and to maintain and assess the COBRA-NC computer code. )

l l

l l

l 1

i 2-11

l

-3. RESEARCH PROGRAMS 1 The principal purpose of the NRC's thermal-hydraulic program is to improve the understanding of, and ability to predict, nuclear power plant behavior durit.g accidents and transients. This capability is needed to provide an assessment of the adequacy of LWR designs and operations to ensure that transients will not-lead to acre serious accidents and to modify regulations as required to ~

ensure _-safe operation of plants. _ The RES role-includes putting research into regulatory form, i.e., applications. The program continues to be_ based on both experiments and computer codes. Table 3 provides a matrix _ relating the specific program areas to be described-in Section 3 with issues and needs previously described in Section 2.

For_each program area, an attempt is made to address why research is needed and why NRC is doing all or part of it. A summary is provided as to what will be done in regulatory space when the research result (s) are available. The antic-ipated use of the specific end product (s) is descrit,ed. In the program-descrip-tion sections, the program strategy planned to solve the need is specified. The ,

integration with and relationship to other programs in government, industry, i foreign countries, and universities are shown. In the-results, products, status,  ;

and plans sections, the research results and intended end product (s) are speci-i fled as well as research publications, rules, regulatory guides, standard review plan changes, and research information letters to be produced. Current results are provided only to the extent that they help define or characterize future (expected) results. Also, expected completion dates are included for signif-icant parts of the programs.

• 3.1 Major Programs Planned for Completion in 1991/1992 Time Frame The following sections describe eight programs that have a defined endpoint.

The first three use large-scale test facilities. The programs are referred to as 20/3D, ROSA-IV, and MIST. The other five involve computer codes. The pro-grams are referred to as ICAP (International Code Assessment Program), TRAC-PWR, l TRAC-BWR, RELAP5, and CSAU (Code Scaling. Applicability, and Uncertainty).

3.1.1 20/3D 3.1.1.1 Purpose and Need In 1978 the NRC initiated a cooperative research program with the Japan Atomic Energy Research Institute (JAERI) and the German Bundesminster fur Forschung and Technologie (BMFT - Ministry for Research and Technology) to study the 1

behavior of light-water reactors during the refill and reflood phases of hypo-thetical LOCAs in a PWR. Of particular interest in this program has been .the two- and three-dimensional thermal-hydraulic behavior in the core, downconer,

and upper plenum; hence, the name "20/3D Program" arose.

Safety research objectives of the 20/3D Program are to:

I j 1. Resolve licensing concerns for effectiveness of core cooling provided by ECCS for PWRs during large- and medium-break LOCAs, and 3-1

. Table 3 RELATIOP. SHIPS OF PROGRAM AREAS TO ISSUES AND NEEDS ISSIES PROGRans (Report Section)

(Report Planned Ceepletion by 1992 Degoing Base Prograss Iles Initiati.es Section) 2013D RJ54-IV MIST ICAP TRAC-Pult TRAC-BUR RELAP5 CSAU TSC NPA IPDB Basic 0156 CEC TTC&EDC Age Inne Ctatn30P Act nest (3.1.1) (3.1.21 (3.1.31 (3.1.41 (3.1.51 (3.1.61 (3.1.7) (3.1.8) (3.2.11 (3.2.2) (3.2.31 (3.2.41 13.3. D (3.3.21 (3.3.33 (3.3.43 (3.3.5) (3.I.61 5.'i;.i, O O O oO o O A, n.'i' O oo O oooo o o n.is <2.3i O o o O -

EEs""2 l) O o o O O O

"-l!::

.CA O o o O O E "t2 ~"6 O o O O o O re,.

t Cool's (2.7)

O Baron Hising ATUS (2.51

$"'.'<!". , o o O O O PTS (2.10) O O

.at "

Diaga (2.111 O O Safety A

LURs (2.123 o lL'u"3; O O O O o Centainosat

& BOP (2.14) n v

2. Provide large-scale data fc assessment of scaling capabilities of computer codes to predict the accident response of PWRs during large- and medium-break LOCAs.

In particular, the 20/30 Program provides experiment data and computer code 1 (TRAC) analyses to support the ECCS rule-revision (see Section 2.1) and esde l

uncertainty evaluation (see Section 2.2) in the areas of ECC bypass around the i downcomer, coolability of a partially damaged core, steam binding effect during l reflood, multidimensional flow effect in the core, and effectiveness of upper plenum injection and vent valves.

! In addition, the 20/30 Program provided full-scale test data on mixing of the .

l high pressure injection fluid with the primary coolant fluid already present in the cold legs to help resolve the pressurized thermal shock issue (see Sec-tion 2.10). The 20/3D Program also provided full-scale test data on counter-current flow between steam and water to help resolve whether a stable flow pattern exists in a reflux condenser mode of cooling during a small-break LOCA (see Section 2.6).

3.1.1.2 Program Description Under the 20/3D Program, JAERI constructed and completed several test series in two large-scale test facilities called the Cylindrical Core Test Facility and the Slab Core Test Facility. Also, the Federii Republic of Germany (FRG) constructed and is now operating a full-scale test facility called the Upper Plenum Test Facility. The NRC provided advanced two phase flow instrumentation and is providing computer code analyses using the Transient Reactor Analysis Code (TRAC) to the three facilities. Each of the above are described below.

1. Cylindrical Core Test Facility j The Cylindrical Core Test Facility (CCTF), operational since 1979, is located in Tokai, Japan, and is opere'ei by the JAERI. The NRC provided advanced instrumentation for insta11atian in the facility and computer code analysis in support of the testing. The CCif it a full-height, scaled-cross-section model of an 1100 MWe PWR with four primary loops. The principal reference PWR is the Trojan Nuclear Generating Station, although certain aspects of the Ohi reactor in Japan are also incorporated.

The CCTF was designed to simulate the thermal-hydraulic behavior of a commercial

, 1100 MWe PWR during the refill and reflood phases of a large-break LOCA with the break in the cold leg. The first series of experiments (Core I) was designed to investigate system thermal hydraulics and core response during the refill and alternative ECC system tests. A second core was used because of the limited j

life cycle of the electric heater rods. The entire test series of CCTF has been completed, and the d&ta are being evaluated at the present time.

2. Slab Core Test Facility The Slab Core Test Facility (SCTF), operational since 1981, is located in Tokai,

! Japan, and is operated by the JAERI. The NRC provided advanced two phase flow i instruments for installation in the facility and computer code analysis in I support of the testing.

)

i

! 3-3

l l

l The objectives of the SCTF testing are to study two-dimensional hydrodynamics and heat transfer in the core and the performance of the ECCS during the end of l blowdown, refill, and reflood phases of a LOCA in a PWR. The SCTF test program 1

includes the testing of three simulated cores. Core I simulates a blocked core of a Westinghouse PWR in which all heater rods in two fuel bundles out of the eight have blockage sleeves simulating fuel rod ballooning. Core II simulates an unblocked core of a Westinghouse PWR. With exception of the fuel rod blockages, Cores I and II are nearly identical. Multiple cores were used because of the limited lifetime of the heater rods. Core III simulates a KWU PWR (FRG).

• Modified upper core and upper plenum hardware are used in Core III along with the upper core support plate injection / extraction system to simulate

, the German combined injection system. The entire test series of the SCTF has also been completed, and the data are now being evaluated.

3. Upper Plenum Tes'. Facility The Upper Plenum Test Facility (UPTF) is located in Mannheim, FRG, at the site of the Grosskraftwerk Mannheim coal-fired power station. This facility is supported by the BMFT and operated by Kraftwerk Union (KWU). The NRC provided advanced instrumentation for installation in UPTF and computer analysis in support of the testing. The UPTF is a full-scale simulation of a 1300 MWe German KWU PWR, specifically the Graftenrheinfeld plant.

The UPTF is designed to test the upper core and upper plenum behavior during the end of blowdown, refill, and reflood phases of a large-break LOCA, as well as downcomer effects, particularly the bypass of ECC water injected through the cold legs during the end of blowdown and refill. About half of the tests in the planned 30-test matrix use combined injection (hot leg and cold leg injection) tr ical of German PWRs. The other half use cold leg injection typical of U.L PWRs. Separate special tests covering the fluid / fluid mixing during a postulated pressurized thermal shock event and the hot leg counter-current flow during the reflux cooling mode of a small-break LOCA are also included in the matrix. All 30 tests are scheduled to be completed by 1989.

4. NRC Contribution to 20/3D Program As indicated earlier, the NRC contribution is in two areas: instrumentation and analysis. Instrumentation was provided to measure two phase flows. The supplied instruments include 13 drag disks, 57 turbine flowmeters, 24 gamma densitometers, 24 liquid level detectors, 779 fluid distribution grid sensors, 40 tie plate drag bodies, 98 liquid breakthrough detectors,. 3 video probes, 58 film probes (for thickness and velocity), 80 impedance probes (for void fraction and velocity), 50 dif ferential pressure transducers, 6 reference conductivity probes, and additional probes. An additional 13 spool pieces containing a turbine flowmeter, a drag disk, and a gamma densitometer were provided. tdditional complementary instruments were provided by the test facilities. These instruments include thermocouplos, differential pressure cells, and nressare transducers.

The analysis part of the NRC contribution is in three categories; facility design assistance, pretest planning, and posttest data analysis. The first two categories took a small fraction of analysis effort. The majority of effort was spent for the last category, posttest analysis, which was directed to: '

3-4

a. Provide understanding of important phenomena of safety concern that are associated with the refill and reflood process of a PWR LOCA,

b. Provide information on how well TRAC predicts the important phenomena of safety concern as compared with the test data, and

c. Identify the areas where TRAC modeling needs to be improved in order to be l i

able to predict the important phenomena of safety concern.

3.1.1.3 Results, status, and Plans This program enabled the United States to obtain large-scale test data for the purpose of resolving upper plenum injection (UPI) Ifcensing issues. 20/30 data are being used by the NRC licensing staff for evaluating the ECCS evaluation models submitted by the UPI plant licensees. In addition, UPI data have been distributed to the UPI plant licensees. This program also provides large-scale test data on LOCA refill and reflood phenomena. These data serve as the best data source for revising 10 CFR Part 50 Appendix K rules. In addition, this program makes a significant contribution to determining the uncertainty of TRAC predictions and to assessing the applicability of TRAC to a full-scale reactor.

As the TRAC computer code begins to be used in the licensing arena, the uncertainty in the TRAC-computed results must be defined. Through this program the predictive capability of TRAC is assessed by comparing the TRAC-calculated results with large-scale test data. In addition, the TRAC computer code was used in the design of large-scale test facilities in Japan and Germany and in the formulation of operating procedures in these facilities as part of the NRC contribution to the 20/30 joint research program.

Testing has been completed in both Japanese facilities (SCTF and CCTF). About half of the planned 30 tests in the German UPTF have been completed. An agree-ment extension to September 30, 1990, will allow for completion of all UPTF tests and subsequent analyses.

3.1.2 ROSA-IV 3.1.2.1 Purpose and Need The NRC and the Japan Atomic Energy Research Institute (JAERI), which is the Japanese government entity in charge of LWR development, have participated in several joint programs of LWR safety research. One of these programs is the ROSA-IV program. Under the ROSA-IV agreement, the NRC is provided access to all test results obtained from the ROSA-IV test facilities. The test data are used by the NRC for code assessment and for further understanding the transient phenomena of PWR small-break LOCA and transients. In addition, the ROSA-IV data can assist the NRC staff in expediting licensing matters and can enhance the credibility of licensing decisions.

l 3.1.2.2 Program Description The ROSA-IV Large Scale Test Facility (LSTF) is the largest nonnuclear facility j in the world for the study of small-break LOCAs and transients in PWRs. The facility is scaled 1:50 in terms of power and volume compared to a full-size 3-5

plant, and 1:1 in terms of vertical scale. The size of ROSA-IV is similar to LOFT, bu' whereas LOFT was oriented toward the study of large-break LOCAs, ROSA-IV is aimed at the study of less severe, but more probable, transients.

The ROSA-IV LSTF data complement LOFT and Semiscale results. The data are used i

to assess and develop more realistic thermal-hydraulic codes to predict plant performance under accident conditions. The data are particularly useful to evaluate the scaling capabilities of the codes since TRAC-PWR and RELAPS were i developed mainly using separate effect and smaller-scale integral facilities. '

The NRC's research efforts on the ROSA-IV program are conducted at two labor-atories. TRAC-PWR code assessment is carried out in a coordinated manner at the Idaho National Engineering Laboratory (INEL) and the Los Alamos National Laboratory (LANL). RELAP5 code assessment is carried out at JAERI in Japan.

INEL and LANL also provide comments and recommendations on the ROSA-IV LSTF test matrix to assist the NRC and JAERI in determining which LSTF tests should be conducted and analyzed to ensure that current safety and licensing issues are addressed in a timely manner.

The NRC program provides JAERI with technical support, including instrumentation refurbishment and spare parts, instrumentation training, and data acquisition and qualification support. INEL has designed, fabricated, and installed the ROSA-IV advanced two phase flow instruments for ROSA-IV beginning in 1984.

3.1.2.3 Results, Status, and Plans All NRC-supplied instruments have been delivered, installed, and qualified and are in operation in the ROSA-IV facility. JAERI has used RELAPS effectively in performing ROSA-IV pretest and posttest analysis work. Since the commissioning of the LSTF, a total of 25 tests have been conducted. Eleven data u ts have been received by LANL and INEL for TRAC-PWR assessment and for possible resolu-tions of the issues identified in Section 2.6.1. Three sets of 5 percent small-break LOCA test results from ROSA-IV together with the Semiscale S-LH-1 test (a 5% small-break LOCA) result and one W PWR calculational result are being analyzed for the resolution of the core level depression and liquid holdup issue. Other issues to be resolved with ROSA-IV test results include the instrument tube "

line rupture and small-break LOCAs with no high pressure injection flow.

The initial ROSA-IV agreement expired on January 31, 1988, and was renewed for an additional 4 years. Posttest analysis work concerning the liquid holdup and core level depression phenomena during a 5 percent small-break LOCA is ongoing at LANL and INEL. The experimental work in ROSA-IV is on schedule.

Upon completion of the TRAC posttest analysis work at LANL and INEL, a research information letter concerning the core level depression phenomena will be issued.

Various technical subjects will be studied in the second phase of the ROSA-IV experiment program. In particular, the anticipated transient tests (such as loss of heat sink, overcooling of steam generator, and station blackout), steam generator heat transfer tests, alternate ECCS tests, natural circulation tests, and tests of plant recovery techniques for various transients and small-break -

LOCAs will be studied. Whenever appropriate, research information letters will be issued for regulatory applications. The expected completion date of this  !

program is FY 1992.

3-6

3.1.3 Multiloop Integral System Test Since early 1980, discussions have been ongoing between the NRC and licensees of B&W plants relative to various licensing issues, including the resolution of NUREG-0737, Item II.K.3.30, Revised Small Break LOCA Methods to Show Compliance I with 10 CFR Part 50, Appendix K. In September 1982 an industry / government group known as the Test Advisory Group (TM) was formed specifically to address these issues. The TAG consisted of representatives from NRC, B&W Owners Group  ;

(BWOG), Electric Power Research Institute (EPRI), and B&W. The TAG responsi-bility was to identify experiment data needs (including existing data and how well they addressed the technical issues related to a small-break LOCA and natural circulation) and to recommend future programs. The TAG successfully completed the work in June 1983 and defined an Integral System Test (IST) pro-gram consisting of selected benchmark analyses of GERDA1 test data; a GERDA upgrade test program known as the Once-Through Integral System (OTIS) program; and the development of a facility specification, design, and associated test program of a 2x4 B&W loop known as the Multiloop Integral System Test (MIST) program. All these were to be carried out at the B&W Alliance Research Center in Ohio. The culmination of the TAG effort was the execution of a triparty agreement in June 1983 among NRC, EPRI, and B&W to carry out the IST program.

In 1987, agreement was reached to conduct additional tests in MIST. The MIST program will be completed in 1988.

1 3.1.3.1 Purpose and Need The technical istues with respect to the small-break LOCA include the require-ment to maintain coolant inventory in the core and to remove core decay heat.

There are a large number of scenarios that can be postulated involving a small-break LOCA. Controlling phenomena for the small-break LOCA change depending on the size and location of the break, the operation of the reactor coolant oumps, assumptions regarding high pressure injection, low pressure injection, and accumulator injection, and whether or not the break can be isolated. The abil-ity of the steam generator to provide decay heat removal capability at any point in the transient depends on the relative temperatures of the primary and secondary systems, the primary and secondary side inventories, and auxiliary feedwater cooling.

Simulation of small-break transients in integral facilities provides data for assessment of thermal-hydraulic codes. The small-break transient is ideal for this assessment because of the large number of phenomena involved. The complex interaction of the controlling phenomena provides a test of the capability of the code to transition from model to model. Because of the large number of phenomeria involved in a small-break LOCA, it is possible to get the correct answer in a calculation for the wrong reason. Therefore, well qualified data that quantify phenomena in each component are necessary to provide an assess-ment base for the individual code models and the integrated system calculation.

2GERDA is an acronym for Geradrohr Dampferzeuger Anlage meaning straight-tube steam generator (test). The GERDA facility was a single loop experimental '

facility of the B&W raised loop design and was m operation at the B&W Alliance Research Center in Ohio. ,

3-7

Natural circulation is an important mode of heat removal during a small-break LOCA. Natural circulation provides a mechanism for removal of core decay heat without the use of reactor coolant pumps. The concern is the ability of the operator to recognize from the existing plant instrumentation those conditions within the reactor coolant system that may cause interruption of the natural circulation process. Natural circulation may be interrupted by voids in the hot leg U-bend, once-through steam generator (OTSG) driven oscillations, and flow oscillations between the cold legs. Subsequently, the operator may be required to stabilize the cooling process and implement means to restore natural circulation, such as using high point vents to remove steam and noncondensible gas from the hot leg U-bend. It should be noted that ..terruption of natural circulation does not imply that core damage is immiaent.

Since the major modes of natural circulation in the B&W design include single-phase, two phase, and boiler-condenser mode (BCM), the codes should be assessed for their ability to accurately calculate the thermal-hydraulic characteristics for all these modes. As such the effect of AFW on BCM cooling needs to be assessed. Unique design features of the B&W plants including the "candy cane" hot leg, the reactor vessel vent valves, and the vertical OTSG require that natural circulation data be obtained specifically for the B&W plant geometry.

Computer codes must predict plant response with reasonable accuracy and without artificial conservatisms being imposed for sake of margin, since sometimes these arbitrary conservatisms mask the important phenomena involved. Data allow assessment of the NRC's best estimate code capabilities for predicting B&W plant transients.

3.1.3.2 Program Description The integral experiment facilities currently available to address small-break LOCA phenomena and r.atural circulation in B&W plants are (1) the MIST facility, (2) the University of Maryland at College Park (UMCP) 2x4 Loop, and (3) the Stanford Research Institute (SRI)-2 facility.

The MIST project was conducted in four phases. Under the first phase, experi-ments were performed in an experimental facility, known as the Once-Through Integral System (OTIS), designed to obtain small-break LOCA and natural circula-tion data for B&W plants of the raised loop design. The second phase of the project was the design and construction of the MIST facility. MIST-III was the testing phase recently completed. MIST-III was designed to address primarily small-break LOCA transients and natural circulation phenomena and was limited l to decay power. HIST-IV is a follow-on to MIST-III and included configuration  !

changes to allow simulation of non-LOCA transients but still limited to decay power levels.

The other two integral facilities where similar experiments will be conducted are the UMCP 2x4 Loop facility and the SRI-2 facility funded by the NRC and EPRI, respectively. They are both low pressure facilities with different geometric scaling ratios and were designed to address scaling atypicalities in the MIST facility in addition to providing code assessment data.

l It is important to realize that the experiment data produced are used to assess how well computer codes predict transient phenomena in B&W plants. From the ,

inception of this program, NRC has initiated an analysis program at LANL to 1

3-8

assess the TRAC-PF1/ MODI code against the data. Similarly, the BWOG will assess the B&W version of the RELAP5/M002 code. Pretest predictions for MIST as well as posttest predictions for all the test facilities are performed throughout the course of the experimental program.

Under the IST program, the cooperative effort between the industry and NRC is very valuable in conducting an experimental program that will provide data of specific interest to the individual parties. During the course of the IST program, Florida Power Corporation and Sacramento Municipal Utility District used the CTIS facility to obtain data to support requests for exemption to the i requirement of high point vents for the reactor vessel head to maintain adequate j core cooling as delineated in 10 CFR Part 50.44(3)(iii). Additionally, the '

Toledo Edison Company also decided to independently fund tests in MIST to support a design change at the Davis-Besse nuclear power plant. The test was pursued as a result of NRC staff concern regarding the accuracy of the computer codes used to support the design change. None of these tests by the utilities would have been possible if the IST program were not in place.

3.1.3.3 Results, Status, and Plans The RELAP5/M002 and the TRAC-PF1/ MODI codes have been benchmarked against data of a small-break LOCA test and a steam generator steady-state test from GERDA. <

Fifteen tests were performed at the OTIS facility. From the OTIS tests, the understanding of the thermal-hydraulic behavior during a post-small-break LOCA event applied to the B&W raised-loop plants was reaffirmed, such as depres-surization to loop saturation, intermittent circulation with repressurization, BCM cooling, refill with high pressure injection (cooling via the leak), and post-refill circulation, cooldown, and depressurization. Tests obtained by the Florida Power Corporation and the Sacramento Municipal Utility District at the OTIS facility led to the NRC exemption for the utilities on the requirement of vessel head vents as delineated in 10 CFR 50.44(3)(111).

Testing in the MIST facility under the phase III program was performed from May 1986 to August 1987, followed by analysis of the data. A total of 50 tests were performed. Although most of the tests were related to small-break LOCA, a limited number of tests were conducted on feed and bleed and steam generator tube rupture. Under the MIST phase IV program, experiment data will be obtained for (1) full-size plant counterpart for MIST, (2) the effects of primary-side natural circulation due to auxiliary feedwater, and (3) risk-dominant accident sequences (e.g., station blackout with loss of auxiliary feedwater and small-break LOCA without high pressure injection). All testing and analysis of the MIST integral experiment data will be completed in 1988. j Integral experiment data and analysis for small-break LOCA transients, feed and bleed operation, as well as all modes of steady-state natural circulation conditions will be completed at the UMCP 2x4 Loop in FY 1988. This will be followed by experiments on steam ynerator behavior and modifications of the facility to conduct operational transients in FY 1989. In FY 1990, testing for operational transients will be completed. A research information letter will be prepared in FY 1989 to discuss application of the data and analysis to full-scale plant behavior.

I 3-9

Most TRAC-PF1/M001 analyses of the integral experiment data from various IST facilities and some counterpart full plant transient calculations will be completed by FY 1988. These analyses.are expected to provide NRC with a validated code to perform B&W transient calculations for small-break LOCA, feed and bleed, steam generator tube rupture, and risk-dominant accident sequences.

)1

. 3.1.4 International Code Assessment Program 1 l 3.1.4.1 Purpose and Need The subject of independent assessment is being addressed as part_of an interna-tional program. For a number of years, different countries and organizations j have been using earlier versions of the TRAC-PWR and RELAPS codes and communi-cating their user experience to the code developers on an informal basis.

During 1984-1985, the NRC undertook to organize and plan a more formal effort devoted to the assessment of the most recent frozen versions of the codes, namely TRAC-PF1/ MOD 1, RELAP5/M002, and TRAC-B01/M001. An International Thermal-Hydraulic Code Assessment Program (ICAP) was organized. A number of countries and organizations are participating in the program, which is coordinated by the The duration is approximately 5 years, from 1986-1991. Its purpose is NRC.

to obtain a considered view of the accuracy and validity of the codes over their range of applicability, while suggesting possible improvements, as needed, to the codes. The ICAP is intended to develop and impreve user guidelines for 1

applying the code. A document has been issued (NUREG-1271) to describe the procedures by which the program is managed.

The ICAP participants perform assessments using data from their experiment facilities and power plants and provide the results to the NRC. The NRC uses this information and the results of domestic assessment and application of each code to arrive at overall conclusions on the accuracy and reliability of the codes for best-estimate predictions of plant behavior. The size of this uncertainty is used in deciding when code improvement should be considered to be complete.

3.1.4.2 Program Description ,

Common guidelines and procedures for conduct of a coordinated International program were prepared during 1985 and a Program Group was constituted. The 7 program is coordinated through this Program Group, which provides peer review of i the codes and their assessment. The principal products of the program are code improvements through correction of code deficiencies and development of user guidelines for nodalization. The ICAP also provides assessment results that are used in the CSAU methodology. The goal is to arrive at a code whose validity and accuracy has been quantified over its range of applicability through use of engineering methods and a well-founded, well-documented assessment data base.

The sharing of effort by the international nuclear community will allow this goal to be reached through an uptimum use of limited resources. The information is being documented in NUREG reports that provide indiv Hual results and summaries of overall results.

The most recent version of tne codes to have been independently assessed were RELAP5/M001 and TRAC-PD2 (TRAC-PF1 received only limited assessment). The in-dependent assessments were concentrated on small-break LOCAs in the case of RELAP5 and on large-break LOCAs for TRAC. Since then, TRAC-PWR was extended 3-10

to small breaks and transients with the addition of secondary side modeling and two-fluid capability. RELAPS was also extensively modified to include secondary side and control system modeling and two-fluid capability. The current assess-ment effort is aimed at refining the code and defining its capabilities. The assessment effort includes a minimum validation matrix for thermal-hydraulic codes formulated through the OECD Nuclear Energy Agency's Committee on the Safety of Nuclear Installations. The ICAP is complemented by domestic analytical programs using TRAC-PWR. RELAP5 assessment is limited to ICAP. The assessment effort is necessary, and the ICAP will provide the NRC with most of what is needed.

. 3.1.4.3 Results, Status, and Plans i

The program at LANL integrates all assessment work being performed with TRAC-PWR. '

An important part of this is the ICAP. The basis of ICAP is a series of bilateral exchange agreements concerning use of TRAC-PWR by France, West Germany, Italy, Japan, Sweden, and the United Kingdom. The integration of assessment results includes domestic use of TRAC-PWR under 2D/3D (see Section 3.1.1), ROSA-IV (Section 3.1.2), and MIST (Section 3.1.3), and other domestic use of the code.

Assessment results are expected to be produced at the rate of about 20-30 cases per year for TRAC-PWR. This information will be summarized periodically through topical reports. The results from this program are also used to discover and correct errors in coding and logic and to make model improvements, where necessary.

The program at INEL integrates all assessment work being performed with RELAP5.

An important part of this is the ICAP. The basis of ICAP is a series of bilateral agreements concerning the use of RELAPS by Belgium, Finland, Italy, Japan, Korea, Netherlands, Spain, Sweden, Switzerland, Taiwan, Yugoslavia, the Joint Research Center Ispra Establishment, and domestic users of RELAPS. Assessment results are expected to be produced at the rate of about 20-30 cases per year for RELAP5. This information will be summarized periodically tl. rough topical reports. The results from this program are also used to discover and correct errors in coding and logic and to make model improvements, where necessary.

Products to be obtained by the NRC from the ICAP efforts are the following:

code assessment cases useful for evaluating scalability, applicability, and uncertainties; experiment data from foreign test facilities; and plant transient data from foreign reactors.

3.1.5 TRAC-PWR This program provides support for maintaining and improving the Transient Reactor .

Analysis Code for Pressurized Water Reactors (TRAC-PWR). In addition, user convenience features are developed so the code will be easier to use and mere portable. The work is performed in conjuriction with both domestic and interna-tional independent assessment programs in which code deficiencies are identified.

These deficiencies are assessed as to the effect on calculational results and addressed on a priority basis.

3.1.5.1 Purpose and Need The objective of this progan is to provide the NRC with a detailed, best-estimate, efficient compute. code for the analysis of postulated accidents and 3-11

l transients in PWR systems. This program focuses on the hardware, thermal-hydraulic, and heat transfer phenomena that are unique to PWR systems and their response in transients. It also provides analytical support to the NRC experimental safety programs. The success of this development is demonstrated in part by the widespread use of and interest in TRAC-PF1/M001. The code has i l been distriDuted to U.S. companies and universities for their use. l 3.1.5.2 Program Description i

The TRAC-PWR code development program has been conducted since 1975. It consists l of the development of versions of the code that can model thermal-hydraulic phenomena important to the NRC. Four major versions of the code have been .

released (TRAC-PIA, TRAC-PD2, TRAC-PF1, and TRAC-PF1/ MOD 1). Each version has first undergone developmental assessment at LANL and then been subjected to independent code assessment. Each version was accompanied by code description, developmental assessment, and independent asressment documents. TRAC-PF1/M001 is currently undergoing independent assessment through the ICAP.

Current planning calls for continued maintenance and error corrections as well as code improvements. A new version of the code, which will contain the two-step numerics technique in the 3D vessel compon(9t as well as other improvements deemed necessary through ICAP, will be released.

The final version of this code, TRAC-PF1/M002, is planned for release in 1989.

This version will contain all the capabilities of TRAC-PF1/M001 but will have an improved interfacial heat and mass transfer package that will allow more accurate prediction of a broader range of transients. In addition, modeling changes identified through independent assessment will be corrected to the extent possible. This work is being carried out as part of a cooperative effort with the United Kingdom and Japan. The development effort is described in a code improvement plan, which was circulated to the ACRS and to ICAP participants.

ICAP participants approved a program of work in December '.987 and agreed to form a technical program group to guide the work. This group will meet periodically to oversee the work until the release of M002.

3.1.5.3 Results, Status, and Plans The TRAC-PF1/M002 code has been under development since 1984 when the development and testing of the 30-2 STEP numerics were first started as an internally funded Los Alamos project. The 3D-2 STEP numerical method was demonstrated to be accurate and fast running. Since the implementation of the 3D-2 STEP required a signif-icant rewrite of the 3D numerics routines in TRAC, it was decided to invert the TRAC 3D data base to aid in the vectorization of the 3D numerics on a CRAY or CRAY-like vector machine. An assessaant to verify that the 3D-2 STEP numerics were not affecting the accuracy of TRAC was completed in 1986.

Over the past 2 years, in an effort to maintain similarity with M001 and to make improvements over the MODI code in selected areas, approximately 260 updates have been implemented. These updates have included error corrections, new user conveniences and capabilities that were originally implemented into the MODI ,

code. Although M001 and M002 are similar, a direct line-by-line comparison  !

between them is impossible because the M002 models have been improved.

l 3-12

The following user conveniences and capabilities have been implemented into both the MO R and M002 computer codes:

1. Multisnur_e connection capability,

2. PWR wif-initialization capability,

3. Steam-separator model, and

4. Countercurrent flow-limit ng radel.

The followir.g features and improved models have been implemented into the M002 code only-l

1. 3D-2 STEP numerics

2. Vectorization of the vessel component

3. Inversion of the vessel data base and introduction of phantom cells to accomnodate vectorization

4. Generalized heat structures '

5. Improved core void fraction / core heat-transfer model

6. Solution to conserve momentum flux

7. Consistent wall shear between 1D and 3D

8. Improved wall-shear model that fixes laminar flow errors, includes surface roughness effect in turbu'ent regime, and improves accuracy of two phase model

9. Improved valve model based on ubserved experiment data for partially closed gate valves

10. Elimination of the Gauss-Siedel method and development o4 the capacitance method for solving the ves Al matrix equations

11. Replacement of the subcooled boiling model with the TRAC-PWR model

12. General orientation of the VESSEL ccmponent to account for gravitational acceleration direction The following code deficiencies hn e been identified and will be addressed in the M002 code: core reflood m del, critical flow model, branching flow at tees, downcomer penetration, condensation in the cold and hut legs, interfacial  ;

shear and heat transfer, pump evrgy sourca term, accumulator modeling, upper plenum modellig, OTSG heat transfer model ka, pd axial cond W on.

3.1.6 TRAC-BWR 3.1.6.1 Purpose and Need The Transient Reactor Analysis Code for Boiiing Water Reactors 'iRAC-BWR) was developed at INEL. The objectiva of thh. program was to provide the NRC with a detailed, best-estimate, efficient computer code for the analysis of postulated accidents and transients in BWR systems. This program was unique among advanced code development projects in that it focused on the hardware, :.M rmal-hydraulic, and heat transfer phenomena that distinguish BWR systems and their response in transients. In addition to pros hing a best-estimate analysis capability for BWR systems, the code can also ld used to address current licensing concerns such as anticipcted transients without scram (ATWS) m ths small break LOCA.

It also provides analytical support to the NRC experimentti safety programs.

The success of this development is attributed in part to the participation of the General Electric Company as a part of the Full Integral System Test (FIST) experimental program funded by General Electric, the NRC, and EPRI. ,

3-13

3.1.6.2 Program Description TRAC-BF1/M001 has been completed. The program will be terminated as a result of budget reductions. The work on TRAC-BF1/M001 included:

1. Remove deficiencies identified by independent assessment;

2. Complete developmental assessment;

3. Provide updated code and user guidelines manual;

4. Complete the development of an input. generator for cross sections and l one-dimensional neutronics; l

l 5. Provide error corrections, user-convenience features, and code improvements l as directed by HRC; and

6. Provide code configuration control, maintenance, and documentation for all tasks related to TRAC-BWR.

3.1.6.3 Results, Status, and Plans l

Previous work has focused on improvements to the hydrodynamic models; completion l of programming changes for increased portability, maintainability, and machine l independence; preparation of a detailed report describing the component models l and constitutive relations; and implementation of the hybrid Courant limit l violating numerics in the three-dimensional VESSEL component.

Improvements to the hydrodynamic models included the implementation of the Bestion correlation for the interfacial friction in rod bundles and implementation of the Megahed correlation for film condensation. Extensive changes to the FORTRAN codien have been made to yield a code that is 99% ANSI Standard FORTAN-77 for ease )f portability and machine independence. These changes included provisions '

for execution on 32 bit machines such as IBM mainframes as well as a minicomputers.

l A detailed r port describing the component models and constitutive relations is i being prepared. This report describes the basis of each of the models, the range l of validity of each of the correlations, and the implementation of the models and correlations in the code. Other work that will not be completed includes l improvements to the conduction solution and associated moving mesh reflood methodolo0y for compatibility with the fast numerics; improved code efficiency; l improvements to the containment capability; and the implementation of the three-dimensional neutron statics module for the generation of one-dimensional neutron cross sections for use in the one-dimensional neutron kinetics model in 1RAC-BF1.

3.1.7 RELAP5 3.1.7.1 Purpose and Need In 1974, the Atomic Energy Commission initiated a project to develop a new generation of computer codes capable of modeling the non-equilibrium and unequal velocities between the liquid and vapor phases that are expected to i occur during a large-break LOCA. In 1976, the NRC sponsored a new advanced i l

3-14

(

(i.e., non-equilibrium and two-fluid velocity) code called RELAPS. It was applied in support of the LOFT and Semiscale programs to calculate not only the large-break LOCA experiments but also small-break LOCAs and operational transients.

RELAPS is a fast running, user-convenient code that is being used by the NRC in the resolution of current safety issues, in the evaluation of plant operator guidelines, and as a tool for auditing safety analyses submitted by licensees.

In addition, the code is in wide use by the nuclear industry both in this country and abroad for design and safety analyses.

l 3.1.7.2 Program Description This program serves to mv -d improve the RELAPS computer code. It pro-vides periodic updates, cou. .mprovements, and user support includir g the NRC support. Furthermore, thD., program encourages the formation of a domestic code user group to I.elp support code maintenance and improvement.

3.1.7.3 Results, Status, and Plans RELAPS is u.ad by NRR and AE0D for safety analysis of PWRs through the use of the nuclear plant analyzer. It is used by NRR to audit license submittals and analyze unresolved safety issues. Recent examples of its application in assisting the regulatory process include analysis of the effect of reactor coolant pump operation during small-break LOCAs and analysis of "feed and bleed" heat removal with lost. of feedwater (TMI restart issue). The Division of Reactor Accident Analysis in RES uses the code (in conjunction with the SCDAP code) to perform severe accident sequance analyses. Other results from regulatory applications include RELAPS demonstration that significant conservatisms exist in the current 10 CFR Part 50 Appendix K emergency core cooling evaluation model requirements.

Since April 1985, the M002 version of REL/P5 has been frozen for production use and assessmcat through ICAP. Frozen means that only error corrections and user conveniences were issueo as updates to the base version (Cycle 36). The main-tenance of the code encompassed resolving user inquiries on exceptional problems, modeling advice, and computer compatibility. Approximately one-half of the average 100 inquiries received each year resulted in correcting code errors.

The balance of inquiries were resolved through user guidelines or were defined as performance deficiencies requiring model improvements. Besides.the mathematical model improvements, other improvements for RELAP5/ MOD 3 include computational speed, portability, and reliability. Improvements in these areas reduce the cost for future analysis from both computation and staff standpoints.

Results coming out of the ICAP are providing most of the guidance on areas where code improvements are needed. By the end of FY 1987, deficiencies that had been identified included modeling of interfacial drag, entrainment, flow regime definition, CHF, and post-CHF heat transfer. Planning for improving the next version of the code (M003) is based largely on these results, but also responds to identified deficiencies in modeling the once-through steam generator used in B&W-designed plants. These plans were documented in a code improvement plan circulated to the ACRS and to ICAP participants. The MOD 3 development will be carried out in cooperation with ICAP members. The MOD 3 development plan was agreed upon by ICAP participants at a meeting in December 1987. A technical program group will meet periodically to oversee the work until M003 is released.

3-15

The precursor to M003 is a developmental version of M002 into which improvements in the above areas are being made. In FY 1988, model improvements include incorporation of a countercurrent flow-limiting model and a revised interfacial drag model for bubbly / slug flow applicable to rod bundles (Bestion). The code was also extensively rewritten so as to provide compatibility.with 32, 60, and 64 bit computers. This means that MOD 3 can be easily adapted to CDC, IBM, and CRAY computers, as well as new super minicomputers. In addition, work began to vectorize the code so as to realize the executional speed gains possible on vector computers.

Current plans call for releasing MOD 3 in June 1989 as the final planned version l of RELAP. Improvements will be made in the areas of modeling injection sprays (i.e. , auxiliary feedwater and pressurizer), cladding metal-water reaction, vertical flow regimes, interfacial heat transfer at high void fraction, vapor I pull-through and liquid entrainment at the opening of a small break, and smoothing  !

transitions in interfacial drag. Work will continue with vectorization, with the expectation that a 50 percent vector fraction is possible. Documentation for the M003 version will be completed, including the users manual and guidelines, developmental assessment report, and quality assurance report. User assistance and configuration control for the MOD 2 version will continue until then. After the release of M003, these same activities will be continued with MOD 3. The MOD 3 code will be assessed by ICAP members during the remainder of the ICAP agreements, until 1991.

Whether or not sn improved version beyond M003 will be needed will depend on the outcome of code uncertainty evaluations, which will quantify the degree of code accuracy for specific plant designs and postulated accidents. Only in those instarces in which the code accuracy (uncertainty) is found to be inadequate for regulatory decisionmaking will improvements be made. Long-term maintenance of the code will diminist as reliability continues to improve; however, the large user community will need an ongoing user support function.

3.1.8 CSAU Implementation 3.1.8.1 Purpose and Need In response to the needs generated by the proposed revision of the ECCS acceptance criteria (discussed in Section 2.2.1), NRC has developed, together with its contractors and consultants, a methodology called Lode Scaling, Applicability and Uacertainties (CSAU) evaluation methodology. It addresses in a unified and systematic manner questions related to:

1. Code scaling capability,

2. Code applicability to scenarios of interest to nuclear power plant (NPP) safety studies, and

3. Total uncertainty associated with calculation of a key parameter (for example, peak cladding temperature) for a specific accident scenario in a specific NPP.

The end product of the CSAU method is an estimate of the total uncertainty which accounts for:

3-16 r7 y

1. Uncertainties due to code and experiments, that is, due to closure rela-tions, models, numerics, code structure, instrumentation, measurement noise, and test conditions,

2. Uncertainty due to nodalization-(user effect),

3. Uncertainty due to scaling,

4. Uncertainty due to code deficiency, and

5. Uncertainty due to plant parameters, which include operating conditions (such as peak linear power, burnup, etc.), plant physical parameters (such ,

as geometries, flow resistances, etc.). and boundary conditions (such as I break size and location, containment pressure, safety injection flow rates). -l l

l 3.1.8.2 Program Description i The CSAU method rests on five elements, that is, on:

1. Complete documentation for a code,

2. Identification of processes important to the specific scenario,

3. Evaluation of code coplicability and assessment,

4. Evaluation of code scaling capability, and

5. Quantification of total uncertainty.

These elements are briefly discussed below. The details are presented in NUREG-1230.

1. Code Documentation The prerequisite for applying the CSAU method is the availability of complete doct'entation, which includes:

a. Ccde manual,

b. Model and correlations quality assurance document,

c. User guide, and

d. Assessment.

The models and correlations document responds-to the need to have detailed information on each closure equation used in the code. Thus for each correla-tion, model, and criterion used in the code, the document provides information on:

a. Its original source,

b. Its data base,

c. Its accuracy, and

d. Its applicability to NPP conditions.

If these closure equations are inadequate or if they are used outside their data base, the document provides (qualitative) information on whether or not the code has the capability to model a process (or processes) important to an accident scenario. Currently, LANL is preparing a document for TRAC-PF1/M001, while INEL is preparing documents for RELAP5/M002 and TRAC-BF1.

3-17

2. Identifhation of Important Processes The focus of the CSAU method is on important thermal-hydraulic processes with detailed attention given only to those processes that contribute significantly to overall uncertainty. Consequently, for each scenario or family of scenarios, the CSAU method requires the development of Process Identification and Ranking Tables (PIRTs), which identify and rank processes important to the scenario and thereby establish its modeling requirements. A PIRT document for a large-break LOCA has been prepared by INEL. Similar documents for small-break LOCA scenarios will be prepared by INEL in FY 1988 and FY 1989.

3. Code Applicability and ly essment The PIRT document and the nodels and correlations document are used in the CSAU method to:

a. Evaluate qualitatively the applicability of the code to calculate the specific scenario, and

b. Establish a code assessment test matrix relevant to the specific scenario or family of scenarios.

Data from separate effect test (SET) facilities are used in the CSAU method to:

a. Evaluate the capability and associated uncertainty of the code to calculate prccesses important to the scenario, and

b. Assess code capability and associated uncertainty to scale up these processes (some to full scale).

INEL is currently preparing code assessment test matrices relevant to large-and small-break LOCA scenarios to be used in the ICAP, 20/3D, MIST, and ROSA-IV cooperative programs. This will ensure that these programs become coordinated and integrated and directed to the needs of the CSAU evaluation methodology.

4. Code Scaleup Capability To address the issue concerned with the scaleup capability of a code, the CSAU evaluation methodology uses the PIRT and the models and correlations documents. l The PIRT is used together with test facility design and scaling criteria to (etermine whether or not important components and/or processes are affected by scale distortions. The models and correlations document is used to evaluate ahether or not closure relations have the capability (adequate data base) to uale up these processes to full scale. If important components and/or processes are not affected by scale distortions present in integral effect test (IET) facil-ities and the relevant closure relations are adequate, then the results of code assessment calculations can be used directly to evaluate code total uncertainty.

However, if these important processes are affected by scale distortions and/or relevant closure relations have inadequate scaleup capability, then experimencs are carried out in large (up to full) scale facilities and used for code assess-ment. For example, a full-scale facility, that is, UPTF, is used to resolve scaling issues resulting from distortions (of plena and downcomers) present :n s'i IET facilities. Test data from ROSA-IV, MIST, ICAP, and other programs are tud to address other questions related to scaling.

3-18

5. Quantification of Total Uncertainty In the CSAU evaluation methodology the total uncertainty consists of the components listed in Section 3.1.8.1, which are quantified in the following manner.

Uncertainties due to code and experiments (which include closure relations, models, numerics, code structure, instrumentation, noise, and test conditions) are quantified in assessment calculations of separate and integral effects tests. The quantification deals only with processes important to the scenario established in the PIRT document. j Uncertainties in scaling up of important processes are quantified by using l separate effect tests which, if needed, are carried out in facilities up tc full scale (for example, in UPTF).

l l

The uncertainty due to nodalization is addressed by using the same nodalization in NPP, SET, and IET calculations. To this end NPP calculations are first performed with a specific nodalization. The adequacy of this nodalization is then tested for key components in the SET and IET assessment calculations.

This approach, that is, performing NPP, IET, and SET calculations with the same nodalization, provides a uniform basis for their assessment and comparison. This eliminates a spurious "user" effect from an uncertainty evaluation.

An estimate of uncertainty due to code deficiency is introduced to account for:

a. Key components that do not have an adequate SET data base for code assessment.

b. Code assessment calculati:ns of a key component that are in poor agree-ment with SET data,

c. Key components / processes modeled by inadequate closure equations identified in a diagnostic analysis.

"ncertainties in plant parameters and state are concerned with:

a. Plant conditions, such as peak linear heat rate, peak power location, hot l assembly rod power and burnup, and core average fluid temperature,

b. Plant physical parameters, such as flow areas, volumes, loop resistances, i I

and pump resistance and degradation.

l

c. Boundary conditions, such as break size and location, containment pressure, 1 and safety injection flow rate. i l

The total uncertainty is obtained by combining uncertainties due to:

a. Code and experiment,

b. Scaling,

c. Nodalization,

d. Code deficiencies, and

e. Plant parameters and state.

i 3-19

3.1.8.3 Results, Status, and Plans l l

The CSAU program was impleinented in May 1987 and is being carried out at INEL, LANL, and Brookhaven National Laboratory. As a first application, the CSAU i evaluation methodology is used to quantify the uncertainty of the calculated l peak cladding temperature for a large-break LOCA in a four-loop W plant using i the TRAC-PF1/M001 code. This first application is to be completed in FY 1988.

It will be followed by quantifying uncertainties for RELAPS/ MOD 2 and TRAC-BF1 to compute the peak cladding temperature in a large-break LOCA in FY 1989.

The CSAU evaluation methodology will then be applied to quantify uncertainties to compute various small-break LOCA scenarios using TRAC-PF1/ MOD 1, RELAP5/ MOD 2, and TRAC-BF1. This work is expected to be completed in FY 1990.

3.2 Ongoing Base Programs l

The following sections describe four programs with a perceived long-term need and thus without a defined endpoint. The programs provide technical support, nuclear plant analyzer improvement, data bank support, and basic studies.

3.2.1 Thermal-Hydraulic Technical Support 3.2.1.1 Purpose and Need The objective of this research element (technical support) is to ensure sufficient ongoing thermal-hydraulic work such that the necessary staffing level and scope of expertise is available at the Technical Support Center (TSC) to support the NRC staff in resolution of priority issues as they arise. The need for a technical support center and its basic function were defined in NRC's thermal-hydraulic integration plan (NUREG-1244). The need to synthesize research and operational results from a wide spectrum of sources to assist in resolution of the issues for which the research programs were initially formulated was presented in an NRR user need letter dated March 9, 1984.

The products of research programs have normally consisted of numerous individual contractor reports providing results of individual portions of the overall program. These reports are aimed at documenting research results as opposed to presenting information in a form that can be readily incorporated into the regulatory process or be used to address the resolution of the particular issue that originally motivated the research. There is a need, therefore, for summary reports that integrate and assess the significance of individual reports and answer the original concerns. Such reports should lead to staff positions on issues.

Thermal-hydraulic research has generated a large amount of information. The need to synthesize and implement research results is apparent in view of the resource expenditures being allocated to thermal hydraulics, currently about $60M per year internationally and $1400M total to date. Most of this information resides in large numbers of programmatic reports. Almost all currently planned research will be completed by about 1990, while nuclear power is expected to continue much longer, even if new plants are not built. The existing research informa-tion must be documented in a form that can be readily retained and used in the long term or else much of it will be effectively lost.

3-20

3.2.1.2 Program Description The TSC provides three basic services: (1) priority issue studies, (2) code applicability studies, and (3) program planning assistance. Priority issue studies will be implemented as needed to meet NRC needs, e.g., to evaluate the regulatory significance of an operating reactor transient related to plant design and operation. Code applicability studies will evaluate the ability of NRC codes to predict the response of each operating reactor type to transients i

of safety significance. Program management and planning will assist NRC in planning, implementing, and monitoring thermal-hydraulic research; will manage priority issue resolution tasks at the TSC, and will synthesize applicable data for issue closure. As an example, RES will review the advanced LWR concepts when they are sufficiently developed for detailed review. At that time the l applicability of existing best-estimate codes will be established, and the need for additional code validation using industry data will be evaluated. In addi-tion, the need for code improvements and for additional experimental support will be established.

The TSC at INEL assists the NRC in providing integration in thermal hydraulics by carrying out activities at the direction of the NRC in the following areas:

1. Identifying existing information available and formulating research programs and schedules;

2. Providing syntheses of completed research to assist in incorporating research results into the regulatory process;

3. Maintaining cognizance of research being performed or having been performed by other NRC contractors, industry, and foreign countries;

4. Maintaining computer codes, quantifying the code accuracy with respect to regulatory issues and objectives, assessing the code scaling capability, and assessing the effects of any code shortcomings and identifying any compensating errors (i.e., getting the right answer for the wrong reason);

5. Providing rapid technical response to urgent regulatory issues raised as a result of transients in operating power plants.

6. Providing support for special studies and projects such as the ECCS rule revision and improvement of the capabilities of NRC Black Fox training simulator;

7. Providing options and recommendations for future research requirements with respect to technical basis for NRC audit of industry requests concerning improving plant capacity factors, improving technical specifications, and improving plant design and safety sysi. ems; evaluating effects of design differences between nonstandardized plant designs; and achieving better j operator training, better operator procedures, and better diagnostic aids; -

8. Summarizing and integrating results from different programs or tasks, discussing their significance, and providing recommendations regarding their applicability to the regulatory program; and l

l 3-21

9. Demonstrating the effectiveness of a multidisciplinary research approach to issue resolution.

The main focus of the TSC is the establishment of validated analytical tools used by the NRC for the evaluation of LWR safety.

A second goal is to put in place mechanisms to permit the rapid deployment of personnel and resources to address safety issues related to thermal hydraulics as they arise. A guiding principle is rendering the products of research (i.e.,

codes, data, methods) more usable in the regulatory framework, that is, synthesizin~g research results so that they are directly applicable in resolving safety issues, i While the computer codes and the supporting experimental data base have matured, quantification of the relative accuracy of these tools over the wide range of accident conditions and plant designs has yet to be achieved. This evaluation of code applicability must be focused on the events identified as significant in terms of risk as well as those that actually occur in reactors but that may have escaped the attention of current risk methodology. Code applicability must be focused on these transients, and the need for additional data and code refinement should be defined in that context.

Dealing effectively with new technical issues has historically been troublesome from several standpoints. It is generally the case that reactor safety issues encompass several technical disciplines requiring a multidisciplinary approach.

However, the expertise base is scattered for these disciplines. In addition, transitory issues are difficult to manage administratively, causing disruption of ongoing work and requiring diversion of existing. resources, and are often incompatible with existing organizational alignments. These difficulties can be overcome by providing more contractor flexibility to respond to NRC's needs.

Finally, thermal-hydraulic safety issues in the-past have been resolved through a combined analytical and experimental approach. With the shutdown of existing experiment facilities, careful planning must be performed to provide a continuing '

experimental capability as part of the TSC.

The TSC encompasses four major task areas: management and planning, code applicability evaluation, continuing program support, and priority studies.

The code applicability and continuing program support areas represent relatively stable, baseload activities, whereas priority studies are relatively unpredict-able. The philosophy is to coordinate the diversion of resources (i.e., manpower and material) from stable areas to the priority studies as necessary to accom-modate new issues. Consolidation of work in the continuing program support area will facilitate this process. Details regarding these areas are provided in Appendix B.

3.2.1.3 Results, Status, and Plans The operational performance study described in Section 2.3 is an example of a priority study completed during FY 1987. Annually updated TSC plans are examples of planning activities. A summary of all major results dealing with the effectiveness of primary system feed and bleed cooldown and depressuriza-tion is an example of a synthesis task.

3-22

(

t Future efforts at the TSC will involve the following tasks: j 1

1. Complete initial applicability assessment of computer codes RELAPS, TRAC-PWR, and TRAC-BWR for transients of concern to NRC for each LWR vendor type reactor. This will provide the NRC staff with an independent ability to evaluate the safety of transient responses of present and proposed reactors.

2. Resolve priority issues as they arise and as recommended by the thermal-hydraulic Regulatory Research Review Group. This will provide the NRC staff with the information necessary to take corrective actions, if necessary, to ensure adequate safety in' nuclear plants.

3. Prepare planning reports as appropriate. These plans become the basis for research programs essential to closure of safety issues of concern to the NRC.

4. Prepare summary reports of research results applicable to important safety issues or other major topics of interest. These reports provide the NRC staff with a basis to evaluate the adequacy of current regulatory practices or the need for alternative actions related to these safety issues or topics.

5. Recommend improvements to plant safety that could be made through improved operator performance shaping factors. Among these are use of plant-specific simulators for operator training, for procedure evaluation and preparation, and for feedback concerning plant design and modification.

The work in this program could be extended in the following areas to NRC's benefit:

1. Perform parametric studies to evaluate the impact of B&W Owners Group plant design and operations recommendations on operator response and plant safety.

2. Develop methods to support guidelines for rating procedures and training by the measured impact they have on crew performance.

3. Establish a technical basis for NRC assessment of programs of the Institute of Nuclear Power Operations for evaluations of training.

4. Develop a technical basis to support guidelines and criteria for simulator training in the following areas:

a. Fidelity of the thermal-hydraulic representation,

b. Control room similitude,

c. Appropriate scenarios for training,

d. Instructor requirements, and

e. Licensing examiner requirements.

5. Develop methods for prediction of crew response to infrequent and complex events.

3-23

3.2.2 Nuclear Plant Analyzer 3.2.2.1 Purpose and Need For more than a decade the NRC developed three major thermal-hydraulic computer codes for the analysis of nuclear power plant response to accident conditions.

For most of this time, the assimilation of computed output from the RELAP5, TRAC-PWR, and TRAC-BWR codes has been based on printed and plotted display of the computed results. This process is clearly time consuming for skilled analysts to interpret the mass of information generated in analyses.

3.2.2.2 Program Description l

l The nuclear plant analyzer (NPA) was developed primarily to provide a more direct, rapid means of displaying several system variables simultaneously in a compact, comprehensive format that could be readily assimilated by analysts to improve )

interpretive efficiency and accuracy. The NPA provides color graphics displays  !

of computed output on a high resolution monitor that can be driven interactively while a computation is in progress, or in a replay mode from a file generated by an earlier calculation.

The interactive mode of NPA application was developed to allow the analyst to simulate plant operator response to ongoing accident transients such as closing or opening valves, starting or stopping pumps, manipulating control rods, etc.

This feature allows study of proposed operating procedures under accident conditions, as well as the analysis of uncertainties in such procedures. The replay mode was developed to permit rerun of completed calculations at various display rates to allow detailed study of plant response.

3.2.2.3 Results, Status, and Plans The NPA has been operational on the INEL CDC CYBER-176 mainframe since 1985 and was also implemented on the LANL CRAY-15 mainfrant in 1987. The NPA has also been used in the NRC Operations Center to assist in emergency drills for the Center staff by serving as a simulator for a plant under9cing an emergency.

Interest in using the NPA as a teaching aid in the NRC's Technical Training Center has been indicated by the Center staff, and the means for implementation of the system are under review (see Section 3.3.3).

Planned enhancements for the NPA include modifications to allow color graphics display on PC color monitors in addition to the high resolution monitors in current use. This will allow a broader audience for NPA use than is now possible because of costs for high resolution monitors. The NPA software will also be expanded to allow user generation of the components (map, driver, and mask). At present these components are generated by INEL programmers excluaively, thereby severely limiting the flexibility of obtaining new or modified graphics displays at other sites.

A continuing effort is in place to expand a library of plant NPAs for each plant vendor and vendor reactor type in operation. Work in FY 1988 will add a fully detailed NPA for the Dresden-3 BWR plant and reduced detail NPAs for the Browns Ferry BWR and Grand Gulf BWR plants, as well as five new replay files to be generated from existing NPAs.

3-24

3.2.3 Nuclear Plant Data Bank 3.2.3.1 Purpose and Need The generation of plant models for analysis of NPP accidents and transients has been performed by a staff of skilled analysts in the Department of Energy's national laboratories, namely: LANL for TRAC-P and INEL for TRAC-B and RELAP5.

The large quantity of plant design and operational data required to generate plant models for these codes has been garnered from a variety of sources and documents. Records of sources and checks to confirm the accuracy of the data used have generally not been documented in a formal process, so that the quality of data currantly in use in most of the models is unknown. In addition to these uncertainties, the conversion of these data into plant models is also a source of uncertainty in the final accuracy of models.

3.2.3.2 Program Description l To minimize or eliminate model uncertainties and produce plant models consistent with the rigor contained in the code phenomenological models, the development of the Nuclear Plant Data Bank (NPDB) was bagd on providing quality-assured input data for a data repository and a fully prescribed automated process for the generation of plant models from these data. These two basic functions of the NPDB are performed by:

1. The Plant Data Management System (PDMS), which includes PC software used by the analyst for file data entry from screen prompts and graphics.

Plant design and operating data entered are based on a Minimum Data Set ~

requirement for the range of accidents to be analyzed and the code to be used for analyses. Raw data from all sources are converted in the PDMS for entry to the NPDB data repository on the mainframe computer (CRAY-1S at LANL). A written record of all sources of data and confirmation of the quality assurance process is also included in a Plant Data Manual.

2. The Plant Model Generator System consisting of mainframe computer software used by the analyst to generate a TRAC-PF1 input model from quality-assured data retrieved from the NPDB data repository.

3.2.3.3 Results, Status, and Plans The NPDB (under joint development by LANL and Scientech) is complete for a 4-loop Westinghouse type plant using the TRAC-PF1 code. This version of the NPDB models the nuclear steam supply system portion of the plant. This program has been terminated as a result of budget reductions.

3.2.4 Basic Studies 3.2.4.1 Purpose and Need The ultimate product of thermal-hydraulic research is accurate computer codes l used to predict the behavior of plants during various accidents and transients, j During the development of these codes (TRAC-PWR, TRAC-BWR, RELAPS, COBRA, and l

RAMONA), a strong program of small expariments and modeling of key phenomena I

(heat transfer, fluid flow regimes, etc.) was required. Basic studies have also included instrumentation development efforts to obtain measurements necessary 3-25 l

1

l 1

to compare the results of computer code predictions with experiments. Now that the computer codes are in an advanced state of development, these basic studies will be focused on resolving remaining shortcomings in the code modeling.

Potential experiments or studies will be identified under the ICAP (see Sec-tion 3.1.4) and the CSAU program (see Section 3.1.8).

3.2.4.2 Program Description Much of the planned future work has not yet been defined. While many code deficiencies are known, the importance of these deficiencies, as well as identification of other needed improvements, will come from the ICAP and CSAU programs. This will allow prioritization of code improvement efforts and identification of additional basic studies, if any, required to make the code improvements. However, some areas of code improvement likely to be required can currently be identified, and some ongoing efforts are discussed below.

The issue of boron mixing in BWRs during an ATWS was discussed in Section 2.8.

A research program has been initiated at the University of California at Santa Barbara to investigate this problem and improve code modeling of boron mixing.

Studies of various flow regimes and scaling of these flow regimes is under way at Argonne National Laboratory (ANL). The program is currently concentrating on modeling of the flow in the hot leg U-bend of a B&W OTSG (see Sections 2.4, 2.5, and 3.3.1).

Other ongoing basic studies include research on water hammer (see Section 2.11) and reactivity transients. Future anticipated work includes basic studies to improve the modeling of auxiliary feedwater spray and heat transfer in a B&W OTSG (see Sections 2.4, 2.5, and 3.3.1) and other code improvements identified by ICAP or CSAU. New initiatives such as containment and balance of plant (Section 3.3.5) and accident management (Section 3.3.6) may also require basic studies to improve code models used under these programs.

3.2.4.3 Results, Status, and Plans The initial effort on boron mixing should be completed in FY 1988. If it is shown that good mixing will occur, no further research would be required since the perfect mixing assumption in existing computer codes would be an adequate representation of the process. However, if it is shown that settling of boron is likely to occur, more sophisticated models of boron mixing would have to be developed. The modeling that would be used would probably be analogous to models used to calculate thermal mixing and stratification developed under the pressurized thermal shock program (see Section 2.10). This would require additional research extending into FY 1990 to develop models, with implementa-tion of the models in the computer codes in FY 1991.

6 The program at ANL will play a key role in determining uncertainty in computer code predictions due to scaling of flow regimes and other thermal-hydraulic processes (see CSAU program in Sections 2.2 and 3.1.8). Support for design and scaling of any required future thermal-hydraulic experiment facilities (see Section 3.3.2) will also be provided.

3-26

3.3 New Programmatic Initiatives The following sections describe six new programs which are being developed.

The programs cover 0TSG testing, integral experiment facility review, NPA utilization, water-hammer guideline development, containment-related tt.armal-hydraulic studies, and accident management-related thermal-hydraulic studies.

3.3.1 B&W OTSG Testing 3.3.1.1 Purpose and Need When the current IST program is corepleted, the data base will consist primarily of integral system test data for small-break LOCA, feed and bleed, steam generator tube rupture, some scaling transients and risk-dominant accident sequences (a.g. ,

station blackout, small-break LOCA without high pressure injection). However,

) experiment data are also needed to assess predictive ct3 abilities for transients such as steam line break, main feedline break, loss of ~4 n feedwater, and steam generator overfill. Most of these transients are initiated from full power. The 19-tube steam generators that are a part of the MIST facility could be used for investigation of OTSG behavior for these transients if a sufficient power supply were available.

Issues regarding OTSG characteristics deal with the ability of the auxiliary fcedwater (AFW) to remove decay heat during off-normal conditions. Identifica-tion and understanding of performance characteristics in the OTSG are, as noted above, important since the secondary side mass inventory is much smaller in the OTSG than in the U-tube steam generator.

Decay heat removal capability is provided by injection of AFW into the boiler region at the periphery and near the top of the tube bundle and results in a complex counterflow situation within the boiler. Radial three-dimensional penetration of the AFW into the tube bundle is restricted by the tube array, and downward penetration is restricted by the tuce support plates (TSPs) and the upward flow of steam.

Calculation of 0TSG heat removal capability during AFW injection requires modeling of several phenomena, including feedwater penetration, wetting of the tube walls, flooding at the TSPs, wall-to-fluid heat transfer, direct contact condensation on the secondary side, and condensation inside the tubes (boiler condenser mode). Many of these phenomena are encountered at different times during the course of any transient with AFW operation.

The code modeling of AFW behavior has been inferred from findings of steam water tests performed at a B&W plant at zero power and air-water AFW penetration testing under a limited range of expected steam generator thermal-hydraulic conditions.

The validity of extrapolating the current methodology of steam generator modeling to a full-scale OTSG under a wide range of thermal-hydraulic conditions encountered during abnormal transients can be improved by the additional separate-effect experiment data planned for herein.

3-27

3.3.1.2 Status and Plans Under the current NRC Thermal-Hydraulic Research Plan for Babcock and Wilcox Plants (NUREG-1236), the procurement of a facility to obtain data on thermal-hydraulic behavior in OTSG associated with steam line break, feedline break, loss of feedwater transient, and steam generator overfill will be initiated in FY 1989 with testing and analysis scheduled for completion in the following year, FY 1990. In FY 1989, preliminary scoping design of the experiment facility simulating a large segment of an OTSG will be initiated. The design will be completed in FY 1990, with construction scheduled for completion in the same year. ' Testing at this large segment OTSG to investigate auxiliary feed-water effect on OTSG thermal-hydraulic performance will be completed in FY 1991.

Continued efforts are being made for cooperative research between the industry and NRC in the OTSG separate-effect testing program.

3.3.2 Continuing Experimental Capability 3.3.2.1 Purpose and Need The NRC has characteristically used both experiment data and best-estimate '

thermal-hydraulic analyses to resolve priority safety issues that, for example, arise periodically as a result of operating reactor transients. Experiment data are often needed to investigate system responses for which data do not exist and which may be key elements in determining plant response. These may be either separate-effect experiments or integral experiments. Small, separate- i effect experiments are not difficult to implement; however, large separate-effect experiments and integral experiments can be very expensive, require long lead times, and require expert staff to design, operate, and evaluate the results.

With the impending closure of the MIST integral facility, which models a B&W plant, there will be no operating integral facility in the U.S. to meet these possible future data needs. Thus, the NRC must rely on foreign facilities (e.g., ROSA-IV), attempt to reactivate an existing U.S. facility (e.g., Semi-scale), wait for a new facility to be built, or accept the uncertainty in response implicit in the existing data base.

One of the system characteristics for which little or no integral data exist is multidimensional behavior; this results from the common design practice of building small-scale, full-height facilities that are characteristically tall and thin. An integral facility scaling study performed as part of this program (NUREG/CR-4824) identified the benefits of an approximately photographic reduction of a power plant to a small scale.

3.3.2.2 Status and Plans A concept for an integral facility that is an approximately photographic reduc-tion of a B&W plant has been developed and analyzed. The system would cost approximately $15M. Its response has been analyzed using both hand calculation techniques and using our most sophisticated thermal-hydraulic code (TRAC-PWR).

In view of the cost of this concept, and the fact that no important safety issue has been defined to require data from such a facility, the program was not pursued. It could be reactivated at a later time if a clear need for data of this type was shown (for example, to provide data on transient response of one of the advanced LWR concepts under development by DOE, EPRI, and industry).

3-28

The current program was closed out in 1988 with documentation of important results of the study.

l 3.3.3 Support to NRC Operations and Technical Training Centers j 3.3.3.1 Purpose and Need One area of support to the Operations Center and the Technical Training Center is the application of the plant analyzers. The nuclear plant analyzer (NPA)

(see Section 3.2.2) has been used in emergency drills for the NRC's Operations Center teams in Bethesda since 1986. The NPA generates information that would normally be transmitted to the staff by phone from plant instrument readings in an actual emergency. The NPA is used in the replay mode and runs at real time for these exercises.

The NRC's Technical Training Center staff in Chattanooga, Tennessee, expressed interest in using the NPA as a training aid for their advanced level reactor courses. It is believed that the NPA type display of plant thermodynamic response during accident transients would help fill the gap in the understanding of such accidents. The current classroom training uses piant schematics and computed time-histories of selected plant variables and subsequent training on full-scale plant simulators. The NPA will serve to provide trainees with an insight to plant responte that cannot be obtained from the current classroom presentations and the simulator experience.

3.3.3.2 Status and Plans To expand the use of the NPA in Operations Center drills, the potential for interactive use of the NPA will be investigated. Such an application requires real-time, online calculations to allow response to simulated drill team deci-sions. Attaining the needed computational speed from the complex codes used with the NPA requires relatively simplified models or faster computers. Reduced detail models have been and continue to be developed by INEL.

A longer-term task is to provide state-of-the-art transient thermal-hydraulic response predictive capability for use in simulators and for procedure valida-tion as well as to demonstrate the importance of having both high fidelity and proper utilization of the capabilities in training, procedure evaluation, etc.

l 3.3.4 Support to Regional Inspectors 3.3.4.1 Purpose and Need One area of support to regional inspectors is the need to prepare a guidebook on condensation-induced water hammer (WH) in PWRs and BWRs to be used by NRC for conducting:

1. Licensing reviews

2. Diagnostic evaluation of a VH event

3. Post-WH event analysis

4. NRR operation procedures evaluation 3-29 l

3.3.4.2 Status and Plans The following tasks hav2 been identified:

1. Review, synthesize, and classify available information and data on WH phenomena in a nuclear power plant. The classification should be based on, and take into account, factors such as plant geometry, operating condition, initiating event, dominant mechanism, etc.

J. Review,' discuss, and evaluate models and analytical methods (including codes) currently available to analyse WH events in nuclear power plants

and calculate resulting loads. Quantify the conservatism of these models and/or methods.

3. Provide methods to perform bounding calculations (and an estimate of accuracy) for each class of WH processes identified in item 1 above.

Whenever possible, present results in graphic form.

I

4. Provide a general trouble-shooting procedure of collecting data on an event, hypothesizing alternative WH phenomena that might have caused the event, then eliminating spurious alternatives by use of the calculational methods provided in items 2 and 3 above.

5. Provide a casebook that uses the above analytic procedures to diagnose several actual nuclear power plant water-hammer events.

3.3.5 Containment and Balance of Plant 3.3.5.1 Purpose and Need There are several thermal-hydraulic issues: (1) NRR needs to understand the range of applicability and limitation of the COBRA-NC code. (This code is being used by NRR to analyze equipment qualification for a LOCA or steam line break accident.) (2) NRR needs test data, specifically heat transfer data, under superheated steam blowdown conditions. (After obtaining the data, new improved heat transfer correlations may be needed.). (3) During a severe accident in a BWR Mark I containment, water in the suppression pool may be entrained into venting paths during attempts to vent. (This would lead to degradation cf venting effectiveness.)

3.3.5.2 Status and Plans The above issues are being discussed with NRR for formulation of a user need letter.

3.3.6 "Front-end" Support for Accident Management 3.3.6.1 Purpose and Need The purpose of this program is to investigate operator actions that can prevent occurrence of direct containment heating (DCH) through depressurization of the reactor coolant system. Specifically, the station blackout without auxiliary feedwater transient will be analyzed and uncertainties in the front-end will be determined. The front-end refers to that portion of the accident nrior to 3-30

core melt. The NRC accident management program contributc:. to the resolution of this issue through activities to develop the methods and data base that will determine the feasibility of depressurization in PWRs as an accident management strategy (as a means to eliminate DCH risk). The specific thermal-hydraulic needs are: (1) performing plant parametric calculations for a specific station blackout sequence to determine pressure, latest time to initiate auxiliary feedwater (to avoid cladding temperature excursion), and time to open relief valves; (2) performing plant parametric calculations to determine the impact of erroneous operator actions; and (3) defining uncertainties and their treatment.

Results will be integrated with other research activities in human factors, severe accidents, and probabilistic risk assessment on DCH. Based on the applicability of the results, the Individual Plant Examination effort for PWRs will be supported. The actisity involves short-term high priority research I (6/12 months) to perform calculations for a selected plant and long-term l research (24 months) to quantify uncertainties related to the calculations. I These calculations may determine the feasibility of depressurization as an accident management strategy for eliminating the DCH risk from high pressure accident scenarios such as the station blackout sequence.

3.3.6.2 Status and Plans The research has been divided into categories that can be summarized as follows:

Plant Calcolations - Perform calculations with the SCOAP/RELAPS code far the Surry plant to study severai options to show that the plant can be depressur-ized and to establish the necessary operator actions. The SCDAP/RELAPS code and the MELPROG/ TRAC code have the capability to predict the entire accident sequence, including metal-water reaction, radiation heat transfer, relocation and slumping of the core, and failure of the vessel (bottom or nozzles) or sp tem piping (including steam generator tubes). A complete transient sequence for a base case will be calculated and sensitivity studies will be performed as needed.

Uncertainties - Assess the capability of the code to calculate depressurization and other important parameters, such as the collapsed liquid level, which occurred during the depressurization process in Semiscale test S-PB-02 and ROSA-IV test SB-PR-01. NRC is currently negotiating with our foreign partners to obtain test data for cyclic accumulator injection. Data can be obtained from SPES (Italy), ROSA-IV (Japan), PKL (Germany), and Bethsy (France) facilities.

These data will be used to assess the capability of the SCDAP/RELAP5 code to calculate depressurization with cyclic accumulator injection.

1 3-31

APPENDIX A RECENT NRR USER NEED REQUESTS AND RES RESPONSES

1. Memo, Denton (NRR) to Minogue (RES), "NRR User Needs for the PWR Versions of TRAC and RELAP5," 6/13/84.

User Need: Continue to maintain both RELAP5 and TRAC-PWR.

Since 1975, RES has been supporting the development of two PWR codes, TRAC and RELAP. This was prompted by the strong need for tools for the analysis of l transients and breaks. Each code had capabilities and limitations that pre-cluded concentrating strictly on one or the other. TRAC and RELAP have now reached the level of being capable, versatile codes, and no major new cap-abilities are needed. Since both codes are operational and each is valuable in its own right, RES will continue to maintain the two codes. Maintenance is comprised of several separate tasks, includirg adapting the code to changes in computers and compilers; correcting software arrors; correcting e fors in modeling logic; improving user conveniences; improving input / output; improving user guidelines and code documentation; and providing consultation and trouble shooting to users. RELAP in particular has a large number of users outside INEL.

User Need: No further major code development or improvement required.

The code development effort has been marked in the past by continual release of new versions. This was largely unavoidable because of the limitations each version had, the rapid progress being made, and the strong need that existed for better codes. The codes have now, however, achieved a level of relative maturity. RES froze the codes in December 1984 for approximately 2h years while they are being assessed; the frozen versions being TRAC-P71/M001 Version 12 and RELAP5/M002 Version 36. Improvements in physical modeling will, in general, not be incorporated during the period in which the coces are frozen.

Only error corrections, user conveniences, and input / output improvements will be '

permitted.

While the codes are mature, they may not be adequate ir, all cases for the pur-poses for which they are intended to be used. Major code development is com-plete, but that is not to say that no further improvement is needed. Use of the codes, particularly through the assessment effort, will highlight many ar-eas where improvements should be made, and these will have to be prioritized according to NRR needs.

User Need: Reduce code assessment effort to that required for regulatory process.

The assessment effort has the objective of quantitatively defining the accuracy of the codes. This means the ability to provide best-estimate calculations with 95 percent probability levels, consistent with SECY-83-472. The most recent versions of the codes that have been independently assessed were A-1

RELAP5/M001 and TRAC-PD2 (TRAC-PF1 received only limited assessment). The independent assessments were concentrated on small-break LOCAs in the case of RELAP and on large-break LOCAs for TRAC. Subsequently, significant changes have been made to both codes. RES, therefore, organized an international code assessment and applications program (ICAP) for TRAC and RELAP. The assessment effort includes a minimum validation matrix formulated by the OECD Nuclear Energy Agency's Committee on the Safety of Nuclear Installations. The ICAP is complemented by use of TRAC in connection with MIST, 20/3D, and ROSA-IV. A l limited independent assessment program on TRAC is also conducted at Sandia.

l RELAP assessment is limited to ICAP. The assessment of TRAC-PF1/M001 and

' RELAPS/ MOD 2 is necessary, and the ICAP will provide the NRC with most of what is needed. Domestic assessment efforts are at a very low level, complement ,

ICAP, and could not be further decreased without being eliminated entirely. l User Need: Form a RES-NRR Regulatory Research Review Group on the codes to review the need for new developn.ent efforts and code assessment.

i According to the plan (NUREG-1244) for technical integration of agency activities approved by the Executive Director for Operations, a Regulatory Research Review Group will be maintained for thermal hydraulics between RES and NRR. This group provides oversight of code development and code assessment. Most recently this group met and provided recommendations for RES priorities concerning implementa-tion of 3D-2 STEP numerics in TRAC-PWR.

User Need: Continue to maintain technical expertise at LANL and INEL.

It is RES policy to maintain technical expertise at more than one laboratory.

INEL has been designated as the thermal-hydraulic technical support center.

LANL has responsibility for TRAC-PWR. Programs at LANL include TRAC-PWR code development, use of TRAC-PWR under ICAP, 2D/3D, and ROSA-IV and MIST analysis using TRAC-PWR. Plans for continuing this work presently extend to 1990 and beyond.

User Need: Faster codes should be economically justified.

Work on improved code numerics has an economic basis and is pursued within the limits of available budget and the expected cost / benefit. Running times fer TRAC-PWR calculations are sometimes excessive due to the three-dimensional formulation. There are instances where sensitivity studies are limited due to computer costs. The 3D-2 STEP numerics, included in TRAC-PF1/M002, will improve i running time. This code version will also include limited vectorization.

However, there are no plans to extensively vectorize the code or for parallel processing due to budget limitations.

Some vectorization has been performed on RELAPS/ MOD 2. There are no plans to extensively vectorize the code or for parallel processing.

User Need: Enter data for plants into the Nuclear Plant Data Bank (NPDB).

Plant input decks currently available to RES are being entered into the NPDB.

The task of entry of other plants will be assumed by NRR.

A-2

2. Memo Denton (NRR) to Minogue (RES), "NRR Thermal-Hydraulic Research Needs," 9/12/84.

User Need: Data on feed line and steam line breaks; data on reactor coolant pump rertart in a voided loop.

During FY 1985, Semiscale performed experiments on steam line and feedwater line breaks. Two experiments were also performed to evaluate steam generator liquid holdup similar to S-UT-3. During FY 1986 a series of experiments on small-break LOCAs without high pressure injection were performed; the effect of reactor coolant pump restart in a voided loop was included in this test series.

Related tests in ROSA-IV are planned, and data are expected to be obtained through the anticipat d ICAP agreements on BETHSY, SPES, and LOBI-II. The forthcoming data should satisfy the deficiency that has existed to date with respect to data on secondary side transients and breaks in Westinghouse and Combustion Engineering geometry.

l User Need: Applicability and adequacy of the codes to Westinghouse plants I with upper head injection.

An INEL evaluation of the use of Semiscale for UHI/UPI testing concluded that the facility would be less than ideal due to its atypicalities, in particular, its pronounced one-dima.nsionality. UHI testing was, therefore, deleted from the Semiscale program. However, the desired information will be obtained in the ROSA-IV program, which includes alternative ECCS tests. RES will soon complete its evaluation of CCTF data obtained for the study of UPI. RES is also negotiating with its 20/3D partners for inclusion of UPI testing in UPTF.

The adequacy of the data and the status of assessment of the codes for UHI/UPI will be reviewed periodically.

User Need: Applicability and adequacy of the codes to BWR-2 plants without jet pumps.

RES has discussed with General Electric their plans for assessment of BWR/2 designs without jet pumps but has not yet determined whether an additional NRC assessment program will be required. RES will continue discussions with General Electric regarding their plans and will provide NRR with recommenda-tions concerning the ability of the General Electric program to meet NRC needs.

User Need: MIST follow-on program to study transients.

RES has completed contract negotiations for the MIST follow on program through FY 1988 to study a spectrum of transients. RES is working with NRR and LANL to define the details of the follow-on tests.

User Need: Develop acceptance criteria for code accuracy.

The question of how good is good enough has been present since code development began and has been much debated without the development of a satisfactory con-sensus to date. The end uses of the codes and the required accuracy that these applications imply determine when code development can be considered to be fully completed. Any acceptance criteria that are ultimately defined must be based on the needs of NRR users and must, therefore, be jointly determined between NRR and RES. As we develop yearly estimates of code accuracy, it may turn out A-3

i I

that an asymptotic value is reached, which would define the limit of code accuracy.

User Need: Quantify the code accuracy to allow best estimate predictions with I 95% probability level.

The assessment effort has the objective of quantitatively defiaing the accuracy of the codes. This means the ability to provide best-estimate calculations with 95 percent probability levels, consistent with SECY-83-472. RES organized an international code assessment and applications program (ICAP) for TRAC and RELAP5. The assessment effort includes a minimum validation matrix formulated by the OECD Nuclear Energy Agency's Committee on the Safety of Nuclear Installa-tions. The . JAP is complemented by use of TRAC in connection with MIST, 2D/3D,

.and ROSA-IV. A limited independent assessment program on TRAC is also conducted at Sandia. RELAPS assessment is limited to ICAP. The assessment of TRAC-PF1/

M001 and RELAPS/ MOD 2 is necessary, and the ICAP will provide the NRC with most of what is need'd.

User Need: Maintain integral experimental capability to respond to new issues as they arise and to information needs for advanced reactor de-signs; consider additional facilities similar to the University of Maryland loop.

Maintenance of knowledgeable contractor staff and experimental capability is RES policy to address unforeseen future thermal-hydraulic issues as they arise or to address future plant designs. We have been evaluating different options to select the most cost-effective one that will meet NRC's needs. INEL has performed an evaluation of different scaling opproaches applicable to small, cost-effective test facilities, e.g. , the U.11versity of Maryland facility. The INEL study also included the cost of retaining a testing capability in existing facilities (i.e., FIST, MIST, and Semiscale) either at their current location i

or consolidated at INEL. l l '

User Need: An advanced simulator is preferred to a thermal-hydraulic facility for human factors research.

Semiscale and MIST tests generally have included the use of applicable operator guidelines for system recovery. Further, the pressurized thermal shock studies included the results of operator actions. These have often yielded valuable insights into the appropriateness of the operator guidelines for accomplishing the intended plant response. This practice will be continued throughout the remainder of these programs. There are, however, no plans for developing inte-gral thermal-hydraulic experimental facilities for the purpose of conducting human factors studies, for which an advanced simulator is more appropriate.

RES has no current plans or budgeted provisions for development of an advanced simulator. There are plans to evaluate the Black Fox BWR simulator at the Chattanooga training center for its applicability to off-normal transients.

User Need: No presently identified need for a thermal-hydraulic facility for equipment qualification.

RES has not identified any equipment qualification needs that would greatly benefit from an integral thermal-hydraulic test facility.

A-4

User Need: Effect of sump blockage, cavitation, and grit on RHR and containment spray pumps.

RES believes the issues of pump and spray performance under a variety of condi-l tions, including strainer blockage, pump cavitation, and degradation of seals, l

have been adequately addressed in work done for Unresolved Safety Issue A-43, I.

"Containment Emergency Sump Performance." In any event, RES agrees that such I issues are better addressed through separate-effect tests than through an in-tegral test facility.

User Need: Steam generator level swell and magnitude of steam superheat during steam line breaks.

The M8-2 program has provided data addressing the issue of steam generator lev-el swell and steam superheat during a main steam line break. In addition, the ICAP agreement with Italy includes model development and assessment with the GEST-GEN and GEST-SEP steam generator facilities.

3. Memo, Sheron (NRR) to Shotkin (RES), "Self-Initialization for TRAC and RELAP5," 11/5/84.

User Need: Implement self-initialization in the codes.

RES has directed LANL and INEL to promptly implement self-initialization during FY 1986 for TRAC-PWR and RELAP5, respectively. The next released version of the codes will contain this feature.

4a. Memo, Denton (NRR) to Hinogue (RES), "NRR User Needs for the Nuclear Plant Data Bank, 5/17/85.

4b. Memo, Denton (NRR) to Minogue (RES), "Long-Range Plans for NPA/NPDB,"

6/30/86.

User Need: Fill the NPDB at the rate of 8 to 10 plants per year.

There are three options for obtaining input decks for the NPDB. The first is to obtain decks already prepared for use of the codes TRAC-PWR, RELAP5, TRAC-BWR, and RETRAN. The existing decks were prepared for different purposes and contain various levels of detail. Some decks were prepared with simple noding that would preclude certain calculations such as large-break LOCAs. The NPDB allows the choice of selecting coarser noding schemes based on more detailed information but does not permit extrapolation to fine noding where the detailed information necessary is not stored. In addition, these decks often lack thorough quality assurance.

The second option is to obtain qualified plant data from utilities as part of a cooperative effort whereby the utilities could access the NPDB in return for expending the effort to produce the information required to prepare input decks. A survey conducted by NRR indicated that utilities were not interested in voluntarily providing qualified plant data.

The third option is to obtain data on a case-by-case basis through special co-operative studies such as the recent pressurized thermal shock program or through regulatory actions following plant transients.

A-5

1 RES responsibility extends as far as preparing data input notebooks that I describe the data required for generating TRAC-PWR, RELAP5, and TRAC-BWR input decks and developing the NPDB software. Actual gathering of plant descriptions l i

and entry of information to the NPDB is not part of the NPA and NPDB program. l Such work is expected to be funded by NRR. RES will, however, enter its exist-

)

ing plant decks beginning in 1987 at the rate of about three per year. Emphasis will be placed on obtaining at least one of each major class of plant design.

User Need: Summary of the available decks in the NPDB and their capabilities.

At present, there is only prototypical data from the Zion plant that is bair.c l used to check out the software for W 4-loop plants.

User Need: Data input books for the spectrum of plant designs in the U.S.

The data input book for the W 4-loop plants will be ready in 1986. Because of FY 1987 funding reductions, work on other data books will not begin until FY 1988.

User Need: Include RETRAN decks as part of NPDB.

RES does not have access to RETRAN decks since they are normally proprietary.

User Need: Capability to enter or review plant data should be available to the NRR staff and NRC contractors.

The development of the NPDB requires that the data are entered and maintained correctly. The prime responsibility for doing so lies with the contractors, LANL and INEL. The NRC and other contractors may, however, require the ability to edit the NPDB directly. This must be done carefully to avoid indiscriminate or ill-conceived changes. Only a limited set of individuals will, therefore, be allowed editing capability, and they will be provided with appropriate pass- l word access. Procedures are being established to ensure that any changes made 1 are done with the knowledge and acquiescence of LANL and INEL.

User Need: NPA and NPDB should run on DOE mainframe.

The NPA and NPDB software are developed for use on DOE mainframe computers at LANL and INEL and the Tektronix 4115B terminal. Additional hardware is not envisaged.

User Need: Complete the NPAs and develop plant model input decks.

See Sections 3.2.2 and 3.2.3 of this plan.

User Need: The integrated NPA/NPDB should be demonstrated as operable from an NRR workstation.

Demonstrations were provided for W 4-loop plants in October 1987, including plant data access, creation of input deck, interactive execution, and output display.

A-6

5. Memo, Denton (NRR) to Minogue (RES), "Request for Follow-on Program in the B&W Integral System Test Facility (MIST)," 10/31/84.

User Need: Upgrade MIST to full power and perform a series of about 25 experiments including secondary system breaks and transiants.

A proposed follow-on program would provide data for non-LOCA events applicable to B&W plants. These tests include feedline and steam line brea<s, steam generator overfill, steam generator tuba rupture, loss of feedwater, station l blackout, anticipated transient without scram, and risk-dominant transients l such as small-break LOCA without high pressure injection, similar to that l

obtained for Westinghouse and Combustion Engineering plants from Semiscale and LOFT. Industry has not committed to support such a program. RES has determined l that a full power upgrade to MIST would not be cost effective under the circum-stances. The ACRS provided similar recommendations. The cost of moving MIST to INEL as part of the technical support center was evaluated and determined not to be feasible. Another possibility would be to initiate a separate effects ,

program similar to MB-2 to inwstigate steam generator heat transfer since many of the most pressing data needs are associated with steam generator phenomena.

This, however, would complement rather than replace the MIST follow-on program.

Various scaling ccmpromises introduced by the MIST design may not allow complete evaluation of the identified issues. Break sizes are limited to small breaks by the currently installed leak locations and the catch tanks. Power is limited to decay heat levels by the available power distribution system. The design of the facility is one-dimensional (tall and thin) and thus multidimensional phenomena such as auxiliary feedwater spray distribution cannot be simulated.

RES has committed to further testing in SaW geo.netry and options are discussed in NUREG-1236.

6. Memo, Houston (NRR) to Bassett (RES), "Request for Assistance in Resolving TMI Action Item II.K.3.30 (SBLOCAs)," 7/19/83.

User Needs: Investigate the acceptability of the inventory distribution calculated by TRAC-PWR and RELAP5 during small-break LOCAs; Determine the sensitivity of code predictions of inventory distribution to steam generator nodalization; Investigate the mechanisms that determine which loop seal initially clears and its impact on core liquid inventory and peak cladding temperature; and Assess the scalability of results to full-scale reactors.

Special tests were performed in Semiscale (LH-series) and in ROSA-IV, and both TRAC-PWR and RELAPS are being used to analyze these data. Several interim reports on these results have been issued; and a final report will be generated after the ROSA-IV analyses are completed.

l l

A-7 l

7. Memo, Sheron (NRR) to Shotkin (RES), "Computer Code Assessment for Both TRAC and RELAP," 11/9/84.

User Need: TRAC-PWR and RELAP5 should be benchmarked against the Ginna steam generator tube rupture event.

The Ginna SGTR has been calculated using TRAC-PF1/M001. Current plans do not call for a parallel calculation with RELAPS.

8. Memo, Denton (NRR) to Minogue (RES), "Potential Need for Additional Experimental Data," 8/12/85.

User Need: Perform experiments related to the Davis-Besse loss of feedwater in MIST and possibly University of Maryland loop; also, possible counterpart test in Semiscale. Determine the adequacy of RELAP5 to calculate the scenario.

A separate response to this request has been forwarded to NRR.

l User Need: Determine adequacy of RELAP4/ MOD 7 calculations of instrument tube failures in Westinghousa plants. Determine the need for new analyses and experimental data.

See Section 3.1.2 of this plan and NUREG/CR-4672.

9. Memo, Denton (NRR) to Minogue (RES), "NRR Comments on Extension of 2D/3D Agreement Beyond April 1985," 5/10/84.

User Need: 2D/3D data on steam binding, ECC bypass, core blockage effects, multidimensional flow effects during refill and reflood, and countercurrent flow limitation.

These data are being provided by the 20/3D program (through 9/30/88) and will be reported to NRR as they become available.

10. Memo, Denton (NRR) to Ross (RES), "Code Maintenance," 9/23/86.

User Need: List of computer codes that should be maintained. TRAC-PF1 and RELAPS will be maintained. TRAC-BF1, COBRA-TRAC, and RAMONA-38 will receive minimal maintenance.

11. Memo, Denton (NRR) to Beckjord (RES), "Clarificatien of User Needs for i Division of BWR Licensing," 10/23/86. i l

User Needs: Complete the independent assessment of the TRAC-BF1 code. ,

Prepare input decks for TRAC-BF1. i Document procedures for generating 1-D cross sections for TRAC-BF1. j Develop a baron mixing model for TRAC-BF1. 1 Update the code documentation and user manual for RAMONA-38. '

See Sections 3.1.6 and 3.2.4 of this plan.

A-8

APPENDIX B THERMAL-HYDRAULIC TECHNICAL SUPPORT CENTER

1. INTRODUCTION The NRC has undertaken an initiative to better integrate and synthesize the results of safety research into the regulatory process. As a part of that initiative, the NRC has established a Technical Support Center (TSC) for thermal-hydraulic research at the Idaho National Engineering Laboratory. The purpose of this appendix is to identify specific functions of the TSC.

The primary emphasis is on rapid response to original issues and demonstrating l the effectiveness of multidisciplinary R&D to issue reevaluation.

l NUREG-1244 spells out NRC's plan to enhance the integration of research results in the area of thern'al hydraulics, and te render the products of research more usable in regulatory decisionmaking. New instructions for planning, coordinating, conducting, and incorporating into the regulatory process and reporting research were delineated. Those germane to the TSC are:

1. Establishment of a joint NRR/RES Regulatory Research Review Group (RRRG) to assist in defining and prioritizing issues requiring resolution and to review research progress to ensure products usable to NRR.

2. Development and publication of a thermal-hydraulic research program plan to provide the detail needed by the various offices to better evaluate planned research.

3. Establishment of a contractor-operator Technical Support Center (TSC) for thermal hydraulics to assist the NRC in the planning, compilation, and synthesis of thermal-hydraulic research. l i

4. Provision for topical summaries of research results and associated code assessment that integrate individual research program results into a form that facilitates utilization in the regulatory process.

This appendix seeks to outline the objectives and specific functions of the TSC (Item 3), including summary reports (Item 4), consistent with the guidance provided in NUREG-1244. Section 2 describes the major objectives of the TSC and Section 3 outlines the methods for conducting TSC functions and relates these to the overall research process directed by NRC.

2. OBJECTIVES The key objective is to support NRC in resolving priority safety issues in all areas related to thermal hydraulics.

B-1

Therefore, the principal purpose of the TSC is to assist the NRC in fulfiilment of the following objectives:

1. Provide the capability to rapidly respond to urgent reguiatory issues raised as a result of actual incidents occurring at operating reactors, oti.er postulated events cr other areas of regulatory concern. An example of a prior incident requiring analysis for regulatory significance was the loss-of-feedwater avent at Davis-Besse. The RRRG will define and pricritize icsues; the TSC will develop a research plan and implement it.following NRC approval. Funding ".ic sta'f may be diverted from other tasks to carry out priority issue work. Impacted programs will be reprogrammed accordingly.

2. Demonstrate the effectiveness of a multidisciplinary research approach to

. issue resolution. Safety issues often i wolve several disciplinary areas l (e.g., thermal hydraulics, structures. human factors, risk assessment).

l Integrating the research in the a;,plicable disciplines promotos efficiency and accuracy of results. When a safety issue assigned to the TSC requires close coupling, the TSC should bring in the necessary expertise to provide an integrated approach. A recent example is the support provided to the NRR on their reassessment of B&W safety.

3. Provide support for special studies and projects. Examples of such studies were the compendium of receerch to support the ECCS rule revision and the review of the potential for upgrading NRC's Black Fox simulator.

4. Identify existing inform. tion and assist the NRC in formulating research programs and schedules. Issues requirint resolution will be identified by the RRRG. The TSC will parform literature reviews and scoping analyses as appropriate anu assist the IRC in planning research that will be conducted by the TSC and/or other appropriate contractors possessing the requisite expertise.

5. Provide synthesis reports cocumenting completed research to assist in incorporating research results into the regulatory process. This includes summarizing and integrating r>sults frem various research programs, discursing their significance, and providing recommendations regarding their applicability to tha regulatory otogram. Topics for synthesis will be identified and prioritized by the h G. Information sources used should include the results of research conducted both domestically and by foreign organizations and encompass experiments, analyses, and reactor operating experience. Incorporation of non-thermal-hydraulic research information should be included'if appropriate.

6. Maintain cognizance of thermal-hydraulic rasearch being performM or having.

Deen performed by othor NRC contractors industry, and foreign cvuntries.

This responsibility of the TSC should be considered a natural part of specifically assigned research tasks. The TSC should serve as a repository for thermal-hydraulic research results mJe available tc it and sheuld recommend to the NRC those data it believes should be retained in the NRC data bank.

B-2

7. Maintain computer codes, quantify their accurae.y with respect to regulatory issues and objectives, assess code scaling capability, and assess the effects of compensating errors. This objective is being partially satis-fied through the conduct of existing programs at INEL. However, the TSC needs to assist the NRC in developing a plan for assessing the applicabil-ity and uncertainty of all best-estimate NRC codes for all applicable reactor types.

8. Provide options and recommendations for future research requirements. Such research would be aimed at: developing the technical basis for NRC audit of industry requests to improve plant capacity factors; improving technical specifications; evaluating effects of design differences between nonstandard plant designs; and achieving better operator training, operator procedures, and improved diagnostic aids.

The Technical Support Center specifically supports NRC in the integration and coordination of thermal-hydraulic research by performing selected priority studies at the specific request and agreement of NRC. Safety issues will be addressed that are of two specific types: (a) priority short-term issue closure (Objectives 2, 3, and 4) and (b) long-term issue closure to ensure regulatory utilization of research results (Objectives 5, 6, 7, and 8). .The functioning organization and work content of the TSC directly reflects the above objectives.

3. OPERATION AND STRUCTURE The TSC responds to direction from the NRC project manager in response to agreements and actions taken by the RRRG. Normally the TSC is tasked and funded to address current and long-term issues. When the RRRG defines a change, the TSC manager will recommend appropriate resource needs, and this will be approved by the appropriate RES management.

The TSC work is divided into two principal thrusts corresponding to the objectives as previously stated:

Priority Issue Resolution: ECCS Rule B&W Safety Evaluation Accident Management Long-Term Cooling Safety Issue Closure CSAU (Code Scaling, Applicability and Uncertain'.y)

Regulatory Utilization: Plant and Transient Analysis Data Evaluation and Use Research and Integration In pursuit and execution of issue closure, the technical approach ensures that input from many sources and talents is brought to bear, and a technical consensus is obtained whenever necessary. A TSC issue manager is designated by the TSC manager and reports directly to his NRC counterpart on the status and results of issue closure. Reporting can be on a formal written or verbal basis. Use of ad hoc groups and review panels is with the consent of NRC. An annual program plan will be prepared. Publication of scientific results in refereed journals will be a goal of the work performed.

B-3

l r

I APPENDIX C ACRONYMS AC Alternating Current ACRG Advanced Code Review Group ACRS Advisory Committee on Reactor Safeguards ADS Automatic Depressurization System ADV Atmuspheric Dump Valve AEC Atomic Energy Commission AE00 (Office for) Analysis and Evaluation of Operational Data (USNRC)

AFW Auxiliary Feedwater ANL Argonne National Laboratory AN0 Arkansas Nuclear One (reactor)

ANS American Nuclear Society ANSI American National Standards Institute A0T Abnormal Operating Transient APS American Physical Society ASME American Society of Mechanical Engineers ATWS Anticipated Transient Without Scram BCL Battelle Columbus Laboratories BCM Boiler-Condenser Mode (of cooling) 80HT Blowdown Heat Transfer BE Best Estimate BMFT Bundesministerium fur Forschung und Technologie (FRG)

BNL Brookhaven National Laboratory B0P Balance of Plant BT Boiling Transition BWOG Babcock and Wilcox Owners Group BWR Boiling Water Reactor B&W Babcock and Wilcox CCFL Countercurrent Flow Limit CCTF Cylindrical Core Test Facility .

CE Combustion Engineering {

CEA Commissariat a l'Energie Atomique (France)  !

CEC Continuing Experimental Capability CFR Code of Federal Regulations CHF Critical Heat Flux CISE Centro Italiano Studi Esperienze, spa (Italy)

CIT Condensation Induced Transient CL Cold leg (pipe)

CLI Cold Leg Injection CLS Cold Leg Suction CNEN Comitato Nazionale Energia Nucleare (Italy)

COBRA Coolant Boiling in Rod Arrays (code)

CPU Central Processing Unit CRDM Control Rod Drive Mechanism CS Core Spray CSAU Code Scali.), Applicability, and Uncertainty C-1

CSNI Committee on the Safety of Nuclear Installations CVCG Chemical and Volume Control System DBA Design Basis Accident DBLOCA Design Basis Loss-of-Coolant Accident DCH Direct Containment Heating DECLG Double-Ended Cold Leg Guillotine (break)

DNB Departure from Nucleate Boiling 00E- Department of Energy ECC Emergency Core Cooling ECCS Emergency Core Cooling System EFS- Emergency Feedwater System EG&G INEL Contractor EM Evaluation Model EPRI Electric Power Research Institute ERDA Energy Research and Development Administration ESF Engineered Safety Feature FIST Full Integral Simulation Test FLECHT Full-length Emergency Cooling Heat Transfer FR Federal Register FRAPCON Fuel Rod Analysis Package - Gap Conductivity (code)

FRAP-S Fuel Rod Analysis Package - Steady State (code)

FRAP-T Fuel Rod Analysis Package - Transient (code)

FRG Federal Republic of Germany FSAR Final Safety Analysis Report FW Feedwater GE General Electric Company HOR Heiss Dampf Reaktor HEM Homogeneous Equilibrium Model HL Hot Leg (pipe)

HLUB Hot leg U-Bend HP Hign Pressure HPCI High Pressure Coolant Injection HPCS High Pressure Core Spray HPI High Pressure Injection HPIS High Pressure Injection System HPV Hign Point Vent HT Heat Transfer H1C Heat Transfer Coefficient IBM International Business Machines ICAP International Code Assessment (and Applications) Program h IET Integral Effect Test INEL Idaho National Engineering Laboratory ISP International Standard Problem IST Integral System Test (program)

JAERI Japan Atomic Energy Research Institute C-2

KfK Kernforschungszentrum Karlsruhe RWU Kraftwerk Union (FRG)

LANL Los Alamos National Laboratory LBLOCA large-Break Loss-of-Coolant Accident LC0 Limiting Condition for Operation LHGR Linear Heat Generation Rate LOBI Loop Blowdown Investigation (facility)

LOCA Loss-of-Coolant Accident LOFT Loss-of-Fluid Test (facility)

LOFW Loss of Feedwater LP Lower Plenum l LPCI Low Pressure Coolant Injection l l

LPCS Low Pressure Core Spray I LPF Lower Plenum Flashing 1.PI Low Pressure Injection LPIS Low Pressure Injection System LSTF Large Scale Test Facility (ROSA-IV)

LWR Light Water Reactor L/0 Length to Diameter MB-2 Model Boiler-2 (facility)

MIST Multiloop Integral System Test (facility)

MIT Massachusetts Institute of Technology MSFB Minimum Stable Film Boiling MSIV Main Steam Isolation Valve MSLB Main Steam Line Break MW(e) Megawatts (electrical)

MW(t) Megawatts (thermal)

NEA Nuclear Energy Agency NESC National Energy Software Center NOS Network Operating System NPA Nuclear Plant Analyzer NPDB Nuclear Plant Data Bank NPP Nuclear Power Plant NRC Nuclear Regulatory Commission NRR (Office of) Nuclear Reactor Regulation (USNRC)

NSSS Nuclear Steam Supply System OECD Organization for Economic Cocperation and Development ORNL Oak Ridge National Laboratory OTIS Once-Through Integral System (facility)

OTSG Once-Through Steam Generator PCT Peak Cladding Temperature PDMS Plant Data Management System PIRT Process Identification and Ranking Table PKL Primarkreislaufe PNL (Battelle) Pacific Northwest Laboratories PORV Power (or Pilot)-Operated Relief Valve PRA Probabilistic Risk Assessment PTS Pressurized Thermal Shock C-3

PV Pressure Vessel PWR Pressurized Water Reactor PZR Pressurizer QA Quality Assurance RCIC Reactor Core Isolation Cooling RCP Reactor Coolant Pump RCS Reactor Coolant System RELAP Reactor Leak and Analysis Program (code) l RES (Office of) Nuclear Regulatory Research (USNRC) l RHR Residual Heat Removal (system)

( RIL Research Information Letter RNB Return to Nucleate Boiling ROSA Rig of Safety Assessment (facility)

RPI Rensselaer Polytechnic Institute RRRG Regulatory Research Review Group RSS Reactor Safety Study RV Reactor Vessel RVVV Reactor Vessel Vent Valve SAI Science Applications, Inc.

SASA Severe Accident Sequence Analysis SBLOCA Small-Break Loss-of-Coolant Accident SCTF Slab Core Test Facility SEASET System Effects and Separate Effects Test SECY Office of the Secretary (USNRC)

SET Separate Effect Test SETS Stability-Enhancing Two-Step (nt.erics)

SG Steam Generator SGTR Steam Generator Tube Rupture SI Safety Injection SLB Steam Line Break SNL Sandia National Laboratory SOLA Soluticn Algorithm for Transient Fluid Flow (code)

SPES Test Facility (Italy)

SRI Standard Research Institute SRV Safety Relief Valve SSTF Steam Sector Test Facility TAG Test Advisory Group THTF Thermal Hydraulic lest Facility TLTA Two-Loop Test Apparatus TMI Three Mile Island (reactor)

TRAC Transient Reactor Analysis Code TSC Technical Support Center TSP Tube Support Plate T/H Thermal Hydraulic UCLA University of California at Los Angeles UCSP Upper Core Support Plate UHI Upper Head Injection UK United Kingdom (England)

C-4

UM University of Maryland UMCP University of Maryland at College Park UP Upper Plenum UPI Upper Plenum Injection UPTF Upper Plenum Test Facility USI Unresolved Safety Issue USNRC U.S. Nuclear Regulatory Commirsion W Westinghouse WH Water Hammer WREM Water Reactor Evaluation Model l

I C-5 l l

,,en .uv.,a.a.~,

g in u t =ve6 .usovuto.1c i.. o.1, u w "T.',"$ BIBLIOGRAPHIC DATA SHEET us .csi ve,,o i o~ , i .s vi.,, NUREG-1252 3 Teigt .ho lutisf LS 3 Lt.vt g L.mn Nuclear Power Plant Themal-Hydraulic Performance Research Program Plan _

a oan amat w Pa'a

.o . ....

. tu i . .., June 1988 6 D.f f atPoAT .1avlo a.o41m ,8.*

July 1988

, . ... o.s...: .i .o ~ ... ... .. ,s .. . oo. . u , ,. c , . , on cia..woa. w o ..n a Division of Reactor and Plant Systems

'"a a 'a **' a *"" "

Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington, DC 20555

i. ... .s o .. . . . . . . .o . .. . ~ o . . 6,~ . . .o . i u ,-- ,, c , ii.v,. o.. .o.i Research Plan

! Same as 7. above ,,,,,;,,,,,,,,,,_,,,

1988 - 1992 13 SWPPht.841..V h0Til 13 331..c7 (D owe, .* ess/

This report provides a descriptior, of the thermal-hydraulic research to be carried out by the Office of Nuclear Regulatory Research. The subject of this plan is plant performance, and the emphasis is on prevention of severe accidents. The plan defines the major issues, relates the proposed research to these issues, defines needed products, provides a historical perspective, and defines the major interoffice and interdisciplinary interfaces. The plan provides a technical basis for planning and conducting current and future thermal-hydraulic research. The plan is consistent with assumptions and guidance in the NRC Five-Year Plan published in March 1988.

,. occ... .a . . .n . . . ..m o c u m ., o . 1, thermal hydraulics plant transients loss of coolant accident computer codes Unlimited regulatory needs TRAC code ,,,,,,,,,,,,,,,,,,,,,,

rasearch programs RELAP code ,,,,,,,,

.ie ~,.... .. :io , ... Unclassified separate effect testing nuclear plant analyzer " " ~ *

• integral effect testing code assessment Unclassified i n %..e s a c.* p.sa s it p ict i

eu.5.tuvtn.s ht .414fth Cff!CE t1998402 292st3tas l

- . . . _ -,- - - - n- ---n- . .-...--n-~~n. . _ ~ ~ - - - - - ~ ~ + , ,,._n- - - - -

.-man- -. -.- as,+ >a- wa- u+.m m.xawn a .

S.

0 1

v  ;-

I' l

r i

l i

t 4

t d

1 4

h f,

k i

i 4

%1 4

i i

d l'

I i

{

i 1

e i

i e

T 1

is l

i I

i

'l I

i l

i i

l i

l i

.. -, --ve-w-r---mm.-, .--m.-,-------a..--,,-e=---e __m.--mm---m-wa__ ,-- _ w

x

. , ;- +

,, , ., , ' E 1 * * '

, {,n i

f3

1. -

UNITED STATES! .

Sycat ,ouam ctass aan NUCLEAR REGULATORY COMMISSION . - Post * * "<5 mo ' +. 2. m

.-WASHINGTON, D.C. 20555 '

. j,, ,,. :y OFFICIAL BUSINESS

~': '.:

- PENALTY FOR PRIVATE USE.' $300 >

120S55078877 1 1ANIR41xA19L US NRC-0 ARM-ADM '

OIV 0Fy PUR SVCS PO ^' <.

v. '

_ PUR MGT HR-PDR NUREG WASHINGTON OC 20555 s

i 4

.=

m

'O z

-n m

3=

?

3g m.

O I

7

' 33 1 1

-E m

z "y ;

t l

N .

e ,

a