ML19126A278

SECY-89-012: Staff Plans for Accident Management Regulatory and Research Programs

ML19126A278
Person / Time
Issue date:	01/18/1989
From:	NRC/EDO
To:
References
NUDOCS 8901270293
Download: ML19126A278 (17)

Text

January 18, 1989 POLICY ISSUE SECY-89-012 (Notation Vote)

For: The Commissioners From: Victor Stello, Jr.

Executive Director for Operations

Subject:

STAFF PLANS FOR ACCIDENT MANAGEMENT REGULATORY AND RESEARCH PROGRAMS

Purpose:

The purposes of this paper are:

1. To describe the major goals, framework, and elements of NRC's accident management program, and the approach of implementing the Accident Management portion of the integration plan for severe accident closure (SECY-88-147), aha

2. To summarize the research efforts planned or now underway to furnish confirmatory information and technical support to the staff in support of Accident Management,

3. To obtain approval of the Accident Management research program through FY90.

Background:

Accident Management encompasses those actions taken during the course of an accident by the plant bpeiatin9 and technical staff to: (1) prevent core damage, (2) terminate the progress of core damage if it begins and retain the core within the reactor vessel, (3) maintain containment integrity as long as possible, and (4) minimize offsite releases. Accident management, in effect, extends the defense-in-depth principle to plant operating staff by extending the operating procedures well beyond the plant design basis into severe fuel damage regimes, with the goal of taking advantage of existing plant equipment and operator skills and creativity to find ways to terminate accidents beyond the design basis or to limit offsite releases.

CONTACT:

L. Shotkin, RES, 49-23530 R. Barrett, NRR, 49-21089

The NRC staff has concluded, based upon PRAs and severe accident analyses, that the risk associated with severe core damage accidents can be further reduced through effective accident management. In this context, effective accident management would ensure that optimal and maximum safety benefits are derived from available, existing systems and plant operating staff through pre-planned strategies. Furthermore, the International Nuclear Safety Advisory Group (INSAG) in its report on Basic Safety Principles for Nuclear Power Plants concluded that accident management and mitigation measures can significantly reduce risk. Accordingly, accident management is considered to be an essential element of the severe accident closure process described in the Integration Plan for Closure of Severe Accident Issues (SECY-88-147) and the Generic Letter on the Individual Plant Examination (Generic Letter 88-20).

In the !PE Generic Letter, the staff deferred the requirement to develop an accident management plan, stating that we are currently developing more specific guidance on this matter and are working with NUMARC to (1) define the scope and content of acceptable accident management programs, and (2) identify a plan of action that will ultimately result in incorporating any plant-specific actions deemed necessary, as a result of the

!PE, into an overall severe accident management program.

Since that time we have made considerable progress towards development of an Accident Management Program that would lead to enhanced accident management capabilities in the nuclear industry. This program will be supported by an Accident Management Research Program, which will be coordinated with and use the results of other programs, including the Containment Performance Improvement Program (CPI), the Severe Accident Research Program (SARP), and the Human Factors Research Program. A summary description of the Accident Management Program and supporting research programs is presented below, and in Attachment 1.

Discussion: The fundamental objective of the proposed Accident Management Program is the following:

Each NRC licensee shall implement for each nuclear plant an "Accident Management Plan" which provides a framework for evaluating information on severe accidents, including that developed through conduct of the Individual Plant Examinations (IPEs), for preparing and implementing severe accident operating procedures, and for training operators and managers in these procedures.

2

The "Accident Management Plan" developed by licensees for each plant will be expected to have four subsidiary objectives which are:

(1) Developing technically sound strategies for maximizing the effectiveness of personnel and equipment in preventing and mitigating potential severe accidents. This includes ensuring that guidance and procedures to implement these strategies are in place at all plants, (2) Assuring that installed instrumentation and equipment called for in the diagnosis and control of accidents beyond the design basis are identified and assessed to determine their availability and capabilities, and the need for incremental improvements to existing systems to assure their availability is assessed, (3) Assuring that nuclear plant staff are trained in the procedures and guidance to follow in the event of an accident beyond the design basis of the plant, and utility management is trained and prepared to deal with severe accidents, and (4) Providing a technical basis for assessing the effectiveness of specific accident management strategies and capabilities.

The NRC Accident Management Program i5 aimed at promoting the most effective use of available utility resources (people and hardware) to prevent and mitigate severe accidents. This would largely be achieved through incrementa*1 improvements in the existing emergency procedures and training programs, and by additional planning for severe accidents that could strengthen the support provided to the plant operating staff in case of a severe accident. Hardware changes or other plant modifications to reduce the frequency of severe accidents are not a central aim of this program, although limited, minor modifications may be identified during the process of developing an Accident Management Plan.

Elements of a Utility Accident Management Plan To varying degrees, accident management capabilities already exist at all U.S. reactors, largely in response to regulatory requirements, implementation guidance, and review criteria, such as that set forth in 10 CFR 50.47 and NUREG-0737 Supplement 1. However, these capabilities are not as strong and comprehensive as they might be 3

through use of a more disciplined approach. The risks associated with severe reactor accidents can be further reduced through implementation of utility accident management plans which incorporate improvements to current utility capabilities in five general areas:

0 Accident Management Procedures On the basis of existing PRAs the staff has identified several generic accident management strategies that can greatly enhance a licensee's ability to cope with the accident scenarios that tend to dominate risk in PRAs. This information will be prcvided to each licenser ith the request that they evaluate its benefits for their plants.

0 Training In Severe Accidents Operators, technical support staff, and managers responsible fur responding in the event of an accident should be generally aware of the progression of severe accidents (i.e., their symptoms and timing) and should be proficient in potential response strategies. Licensees' accident management plans are expected to be aimed at upgrading existing training programs as needed to ensure that training for these personnel includes an appropriate treatment of severe accident management. This will be done with as little impact as possible on the current training curricula for licensed plant operators.

0 Accident Management Guidance Each licensee will be expected to make available for the technical support staff and managers a set of guidance for diagnosing the progress of severe accidents and planning the appropriate response. The NRC staff will work with industry (e.g., NUMARC/EPRI) towards the development of generic guidance to licensees in this area.

0 Instrumentation Licensees will be expected to review instrumentation changes that might be needed at their plants in order to implement their accident management procedures.

Currently, the NRC Office of Nuclear Regulatory Research and EPRI are independently assessing the need for and availability of instrumentation during various accident scenarios. When comparing the 4

results of this assessment to Regulatory Guide 1.97 which describes dcceptable standards for post accident monitoring instrumentation, it is expected that the impact on existing instrumentation will be minimal.

0 Oecisionmdking Responsibilities Each licensee's "Accident Management Plan 11 will be expected to include a review, and modification if necessary, of the plant's current decisionmaking authority for accident management strategies, to assure:

well-established, clear lines of authority and communications for severe accident conditions, assigned responsibilities for specific key decisions and established authority ano criteria for procedural overrides and ad-hoc equipment/procedure modifications.

Approach For Accident Management Implementation The staff intends to work with industry to define the scope and attributes of a utility accident management plan which meets the four major plan objectives, and to develop guidelines which describe the plant-specific implementation of such a plan. The principal interaction to date with industry on accident management has been through the Nuclear Management and Resources Council (NUMARC). A working group has been established by NUMARC to address the matter ot severe accidents, with accident management being a high priority. In addition, we expect NUMARC to provide industry's perspective and bring about the necessary industry-supported initiatives on accident management. The staff also plans to interact through NUMARC with the owners' groups for each reactor vendor since the prospective accident management procedures and equipment improvements are closely related to the emergency procedures guidelines that have already been developed by the owners*

groups.

The regulatory mechanism for obtaining improvements in industry accident management capabilities wi11 be through issuance of a generic letter. A draft of the generic letter will be circulated to utilities through NUMARC and to the public for comment. Additionally, the contents of the generic letter will be discussed at workshops associated with the IPE generic letter. An outine of the generic letter is attached to this Commission Paper (Attachment 2).

5

It is the staff's view that many elements of accident management are sufficiently well understood and separate from the IPE analysis, that improvements in accident management capabilities can be realized prior to completion of an IPE. Examples of such improvements include near term implementation of certain accident management procedures (to be included in the Generic Letter on accident management) and the revision of training curricula for emergency response personnel to include current insights on severe accident progression and phenomena.

In this regard, industry will be encouraged to implement accident management improvements as soon as practical.

The staff intends to be flexible with regard to implementation of aspects of accident management requiring plant-specific information to be learned from the IPE, recognizing that some licensees are well advanced in PRA and severe accident studies for their plants, whereas other licensees are only beginning to consider severe accident analysis in response to the IPE Generic Letter.

Indeed, some licensees have already made significant advances in several of the general areas of accident management mentioned previously. Thus, the details of the development and implementation of plant-specific accident 111anagement plans will be pursued with each licensee on an individual basis.

Implementation of the Accident Management Program is expected to be accomplished through an extension of existing industry programs, and evaluated through existing regulatory mechanisms. For instance, new training requirements could be integrated with the existing INPO training program. Generic accident management procedures could be incorporated into emergency procedure guidelines by the vendor owners groups and assessed by NRC according to existing channels of review. Other aspects of the program, such as hardware modifications and implementation of computational tools, could be implemented through internal utility mechanisms such as their 10 CFR 50.59 process ana would be subject to NRC audit and inspection in the usual way.

NRC Accident Management Research Program The NRC research program has an important role to play in contributing to the Accident Management Program. Our understanding of the physical progression of severe core damage accidents is incomplete, and the NRC will rely on the research program to supply needed information and provide insights for accident management, particularly in the area of limiting 6

potential radioactive releases and stabilizing conditions should the reactor vessel be breached. Research information will be developed in this severe accident regime for many years, and as a result we can expect that licensees' accident management plans will be further upgraded as new research information is evaluated and a consensus reached that revisions to severe accident strategies can reduce risks further.

Research activities will center on assessing the feasibility of various strategies that might be implemented by utilities to prevent or mitigate severe accidents, and on identifying those which should be considered for inclusion in utility accident management plans. This will include an investigation of specific accident management strategies applicable to the period before the core would penetrate the reactor vessel (in-vessel accident management) and those applicable following postulated reactor vessel penetration (containment and release management). In all cases, the design and operational requirements for strategy execution will be evaluated, but emphasis will also be given to examining potential circumstances under which certain operator actions could worsen accident consequences or adversely impact the ability to achieve a long-term, stable state. Much of the information needed for accident management research will be drawn from several existing NRC programs (e.g., the CPI and Severe Accident Research Programs), as wll as existing programs in other countries.

Research activities are divided into a short-term effort and a long-term program. The short-term effort will support two principal elements in the closure plan: (1) definition of example severe accident management strategies (primarily preventative) which can be formulated from existing insights reports, PRA studies, and completed research; and (2) development of a framework which defines the necessary components of a functioning utility severe accident management plan. The products will be documents to be included as appendices to the generic letter on accident management.

The long-term research program includes those activities of a confirmatory nature, which are not required for closure. The short-term program to achieve closure is proceeding effectively with the present knowledge base.

The long-term program can be considered as an augmentation to the existing Severe Accident Research Program (SARP) and will draw heavily from the results of SARP. In fact, SARP will provide improved knowledge and phenomenological modeling for such complex processes as steam explosions, 7

degraded core coolability, crust formation, hydrogen behavior, vessel depressurization, direct containment heating, etc., which will then be used to evaluate both the benefits and the adverse effects of candidate strategies. The research program has been formulated to provide the NRC staff with an organized, comprehensive basis for evaluation of generic accident management strategies. The program will define supporting studies needed from related programs (e.g.,

SARP, Human Factors research, PRA studies). It will also integrate all appropriate results from these programs, and others related to accident management (e.g., CPI, IPE and international research), into a practical assessment for generic accident management strategies. It is anticipated that these strategies will be primarily mitigative in nature. Upon completion, it will provide a reasonably detailed evaluation of both the instrumentation and training capabilities needed to support these strategies.

The long-term program will also evaluate the potential adverse effects that could occur if and when these strategies are applied. Examples are strategies such as water addition to a degraded core and use of special emergency equipment. Each must be analyzed carefully tc fully understand the consequences of these strategies.

Some of the specific issues to be addressed include:

instrumentation needs for proper diagnosis of the course of events and their observable symptoms; the capability of existing equipment to bring the reactor to a long-term stable state; the effect of timing of human action on the success of candidate strategies; the consequences of adding water to a degraded core; and the uncertainties in phenomenological knowledge and its consequences on strategy development.

Even though this research will concentrate on all consequences of potential actions, it is currently expected that the ultimate guidance to the operating staff is to always add water during the course of a severe accident.

The NRC resources associated with reaching closure on accident management will be approximately 5 FTE of staff effort and about $500K of contractor assistance over the next twelve months. Resources for the long-term research program are $1.4M in FY89 and $4M in FY90 (See Attachment 1). Resource needs beyond FY90 are expected to be similar, and will be submitted for Commission approval as port of the normal budgetary process.

8

Recommendation: That the Commission note the approach proposed herein for implementing the accident management portion of the Integration Plan for Closure of Severe Accident Issues (SECY-88-147), and approve the accident management research projects for FY89 and FY90.

Scheduling: This paper is scheduled to be considered at an open meeting on January 23, 1989.

ctor Stel; Jr Executive Director for Operations Attachments:

1. NRC A/M Research Projects for FY89/90

2. Outline of Generic Letter 9

ATTACHMENT 1 HRC A/M RESEARCH PROJECTS FOR FY 89/90 A. Ongoing Projects

1. Depressurization to minimize direct containment heating.

An analytic study of station blackout (TMLB') at the Surry plant, using depressurization through the PORV. Results indicate that the pressurizer surge line could overheat and rupture, leading to reasonably early depressurization.

2. LWR recriticality.

An analytic study of injecting unborated water, as well as water with increasing levels of boron concentration, into an LWR core:

(1) when the control blades have melted, but the fuel rods are intact; and (2) when the core geometry has degraded.

3. Instrumentation requirements.

Development and demonstration of a structured methodology to determine adequacy of existing plant instrumentation to monitor severe accidents.

4. Report on candidate A/M strategies; Appendix I to the Generic Letter.

5. Report on framework of an A/M program; Appendix II to the Generic Letter.

B. New Projects (FY 89/90)

NRC has a longer-term program to conduct research directly related to A/M.

This longer-term research would focus on assuring integration of completed ongoing resarch (from CPI, SARP, PRA, and international cooperative programs) to effectively evaluate A/M guidance, while simultaneously evaluating potential disadvantages of seemingly advantageous accident management actions. Our longer-term research program has been thought out in detail and discussed with the various contractors scheduled to conduct the research. The following tasks will be covered by this program in FY 89/90.

In-vessel A/M candidate procedures Ex-vessel A/M candidate procedures Information needs for A/M A/M framework implementation Bounds of coolability of a degraded core Uncertainties affecting A/M BWR A/M insights (MARK-I, III)

Strategies related to OCH Diagnostic Computational Aids Guidelines for industry audit Support for Plant Exercises The total funding required to complete these tasks is $1370K in FY 89 and

$3935K in FY 9CJ.

Attachment 2 Outline of Generic Letter

SUBJECT:

Accident Management (Supplement to Generic Letter 88-20, "Individual Plant Examination.")

1. Objective Licensees shall certify to the NRC that they have put in place an "Accident Management Plan" providing a framework for evaluating information on severe accidents, for preparing and implementing severe accident procedures, and for training operators and managers in those procedures.

2. Elements of a utility Accident Management Plan should include the following:

0 Procedures: Each licensee should, in the context of performing their IPE, evaluate and implement severe accident procedures. The staff has identified general strategies which should be included and has compiled a list of "example" procedures (see Appendix I) which should be considered for early implementation (e.g., prior to completion of the IPE where the licensee determines this to be of benefit to his plant protection and to public safety).

0 Training in severe accidents should be provided for operators, technical staff and managers responsible for responding in the event of a severe accident.

0 Guidance and computational aids for diagnosing and responding to accidents should be provided to technical support staff and managers.

0 An assessment of the need for and availability of instrumentation should be performed.

0 A review of utility decisionmaking processes for severe accident response should be conducted.

3. NRC/NUMARC Interaction NRC staff will work closely with industry groups (e.g., owner's roups and EPRI) through the NUMARC Severe Accident Working Group (SAWGJ to better define the attributes of an Accident Management Plan and to develop guidelines that define a generic framework for a utility to evaluate and organize its resources to prepare for and respond to severe accidents (to be supplied as Appendix II). These guidelines, to be developed as part of an industry initiative, should address each of the key attributes of a utility Accident Management Plan and provide a basis for responding to the generic letter.

4. Utility Response 0

Within 90 days of receipt of this letter, licensees should submit a response certifying their conmitment to implement an Accident Management Plan that provides a framework for evaluating and implementing severe accident procedures.

0 In each case, a best estimate schedule should be included for implementing their Accident Management Plan.

5. Regulatory Basis 10 CFR 50.54(F)

6. Appendices:

I. Candidate Procedures for Near Tenn Implementation (draft attached)

II. Framework of an Accident Management Program ( under development)

APPENDIX 1 Each utility should systematically seek to identify and implement effective procedures (and associated hardware) to optimize the plant's accident management resources. These procedures should at least address three global strategies derived from operational experience and PRA insights which have significant potential for reducing plant risk:

0 Procedures for conserving and/or replenishing limited utility resources during the course of an accident. These resources would include, for example battery capacity, borated water, and compressed air.

0 Procedures for using plant systems and components for innovative applications during an accident. This would include enabling crossties of support systems or the use of fire systems, or CRO pumps (in the case of a BWR), for decay heat removal. In addition, this category includes procedures to connect alternate electrical power sources to meet

\ 1 critical safety needs during an accident scenario.

Procedures for defeating interlocks and overriding component protective trips in emergency situations. An example of this strategy would be the ability to reopen MSIV's in a BWR ATWS event.

These three strategies, and others as appropriate, should be applied to each of the applicable major safety functions of the plant: reactivity control, coolant inventory control, heat removal and containment perfonnance; as well as to the prt'nciple support functions: RCS depressurization, electric power, equipment cooling, and air systems. In the context of performing the IPE, each utility shoul-d be alert to identify and implement effective procedures associated with these strategies. Your evaluation should not be limited to the strategies given in Table 1. For instance, procedures to maintain containment function and delay or prevent possible early containment failure should also be assessed.

Table 1 contains a list of "example" procedures for each global strategy.

These procedures are categorized according to the safety functions they relate to and the types of plants they apply to. These procedures are to be evaluated on a plant-specific basis and considered for near term implementation prior to completion of the IPE (absent sound arguments to the contrary).

This letter assumes implementation of existing EPGs addressed in the ATWS and Station Blackout rules, in addition to feed-and-bleed and BWR contairvnent venting provisions of the EPGs. There may be some areas of overlap between these existing emergency operating procedures and the procedures described in Table 1, however, the NRC believes the benefits from additional utility review and strengthening of existing procedures would include both reduction of risk magnitude and uncertainty from severe accidents and 11K>reover would outweigh the costs of any duplication of effort.

TAl3Lt Gener*i c Accident Managment Procedures For Near Term Evlation and Implementation

/l.t fee ted Plant cabilitt Prine iSafe.!l_ Objec! iv1:

lobl Strtegt 1 £le Procedure £..y11ct ion --BWRPWR Prevention Mitigat1on I. Lonservi11g and Replenbhing Limited Resour1.es Procedure to refill RWST with Coolant Inventory X X X X borated water, or CST with condensate. Contain11,e11 t Performo nee Assure adequate supply of boron on site.

0 Maintain lCCS suction to condensate Coolant Inventory X X systems to avoid pump failure due to high su,,pression pool temperatu rt!.

Procedures fur throttlin Coolant Inventory X X X containment. sprays to conserve water for cure injection.

"Procedures to conserve battery El Htric Power X X X X capacity by shedding non-essential loads.

u Procedures for use of portable Electric Power X X X X battl:!ry chargers or other powe1*

sources tu recharge batteries.

This may require re-examination to defeat existing interlocks

I ABLE _!_J CONTINll[_!U Affected r 1a,1 t A plicob il.!_!r Principal SafetL_Qbjective G 1 oba I __ St r l xame 1 e Procedure fe!.l_Function - -------.BW PWR -Prevent1on-Mit1gdt}on-

" Procedures tu enc1ble emergency Ai r/N,; X X X X reple11ishment 0f gas supply, or otherwise assure operability of air operated compo11tnts.

Procedures to enalile early detection, Coolant J 11ventory X X X X isolation, or otherwise mitigate the Con tai rnnent Perf.

effects of an interfacing LOCA.

I I. llse of Systt::111:,/Compo11e11ts In Innovative Applications "Pi-ucedurt:, tu enable emergency use of avo ilable pumps to accomp 1 isli safety functions

- Use ld diesel fire systems for Cuol;irit Inventory X X X X injerLion to BWR core, PWR Heat Removal stet1111 genera tors, or contu *j nmen l Containment Perf.

sprays.

- Use c,f CRD purnps (llWR) or Coolant lnvuitory X X X char'.)ing pumps (PWR) for c.ore injection.

- Use l1f ulternate i11Jectio11 (e.g., Coola11t inventory X X X hydro test i-,u111p) wh1::11 RCP sea 1 cooli11g is lust. 2

- Procedures (and associated hardware) lleat Removal X X X to enable emergency crosstie of Coolant Inventory service water and CCW to RHR ( BIJR) or Feedwater (PWR) .

Risk signfic.a11ce of seal failure is strongly dependent on the seal design.

TABLE I (CONTINUED)

Affected Pl on t A{l_Ql__i_ cabili_!t Principal Safet Objective G lobd I Strcttegx I xample_Prucedure Sdtett Function -----nwP1lR -fireventTori-

_H:Igat}on-

- Usr:: c,t condensote, or startup pumps lieut kemovdl X X for f f::t:dwater injection.

• • Procedures ( u11d hardware) to er,able Ileat Removal X X X X emergency u,nnection of available AC power sou1*c.es to meet critical safety needs.

- Use of diesel generdtor or gas Coo la11t Inventory X X turbirn! generator to drive CRD pu111ps for core injectiori.

- Proced1ires to enable emergency Electric Power X X X X crosslie of AC power between two units or to onsite gos turbine generutor.

'* Procedures l c., enable emf::rgency conr,ection of injectiun systems to alternate water sources

- Proced1ires tu assure dppropriate Coolant I 11ve11tory X X X recirculation switchover arid to Ilea t Rer11ova l cope with the failure to switchover in LOU\,

- Proceduns to er1able emergency ltedt removal X X X c.onnection of service water or E4:juipment Cooli 11g X X X X feedwuter systems to rivers, reservoirs or municipal water systtms.

Proced111*es for Reactivity Control

TABLE I (rnr:1 INUEU)

Affected P 1 an l_EE..!l.!:!!....!...!J...!1. Princi Sate.!l_ULjt.:ctive liloLo l_tra_!:gy xamrle Procedure Sa fe_!:t_ Fu JlC tiOri BWR PWR f'revention l'litition

- l'r-oced1ire to initiate SLC) i11 reactivity Control X X X case of potential core damage and tu guard agaist boron dilution when core injection is res to red.

- Ensur1: dbundd11t supply of Redctivity Control X X X X bur,ited makeup for long-terrn accident control.

III. l,deating lrilerlocks and Component Protective Triµs ira lmergencii*s.

u Procedures to reopen MSIVs and lie at Removo 1 X X X lurliine Bypass Valves to regain the conde11:..tr as a heat sink.

0 Procedure tu extend RCIC Coolant Inventory X X availability by either raisir,g the turbine exhaust pressure trip set point, or overriding the trip function.

' hccedure:, tu e1odlile emergency All X X bypass of protective trips for diesel generdlur:, ar,d injection pumps.