ML19112A268: Difference between revisions
StriderTol (talk | contribs) (Created page by program invented by StriderTol) |
StriderTol (talk | contribs) (Created page by program invented by StriderTol) |
||
Line 16: | Line 16: | ||
=Text= | =Text= | ||
{{#Wiki_filter:International Agreement Report LBLOCA Uncertainty Analysis of Maanshan Nuclear Power Plant with RELAP5/SNAP and DAKOTA Prepared by: Chunkuan Shih, Jong-Rong Wang, Jian-Ting Chen, Yuh-Ming Ferng, Shao-Wen Chen, Ting- | {{#Wiki_filter:NUREG/IA-0509 International Agreement Report LBLOCA Uncertainty Analysis of Maanshan Nuclear Power Plant with RELAP5/SNAP and DAKOTA Prepared by: | ||
*Department of Nuclear Safety, Taiwan Power Company 242, Section 3, Roosevelt Rd., Zhongzheng District, Taipei, Taiwan K.Tien, NRC Project Manager | Chunkuan Shih, Jong-Rong Wang, Jian-Ting Chen, Yuh-Ming Ferng, Shao-Wen Chen, Ting-Yi Wang *, and Tzu-Yao Yu* | ||
Institute of Nuclear Engineering and Science, National Tsing Hua University Nuclear and New Energy Education and Research Foundation 101 Section 2, Kuang Fu Rd., HsinChu, Taiwan | |||
* Department of Nuclear Safety, Taiwan Power Company 242, Section 3, Roosevelt Rd., Zhongzheng District, Taipei, Taiwan K. Tien, NRC Project Manager Division of Systems Analysis Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington, DC 20555-0001 Manuscript Completed: June 2018 Date Published: April 2019 Prepared as part of The Agreement on Research Participation and Technical Exchange Under the Thermal-Hydraulic Code Applications and Maintenance Program (CAMP) | |||
Published by U.S. Nuclear Regulatory Commission | |||
AVAILABILITY OF REFERENCE MATERIALS IN NRC PUBLICATIONS NRC Reference Material Non-NRC Reference Material As of November 1999, you may electronically access Documents available from public and special technical NUREG-series publications and other NRC records at libraries include all open literature items, such as books, NRCs Library at www.nrc.gov/reading-rm.html. Publicly journal articles, transactions, Federal Register notices, released records include, to name a few, NUREG-series Federal and State legislation, and congressional reports. | |||
publications; Federal Register notices; applicant, Such documents as theses, dissertations, foreign reports licensee, and vendor documents and correspondence; and translations, and non-NRC conference proceedings NRC correspondence and internal memoranda; bulletins may be purchased from their sponsoring organization. | |||
and information notices; inspection and investigative reports; licensee event reports; and Commission papers Copies of industry codes and standards used in a and their attachments. substantive manner in the NRC regulatory process are maintained at NRC publications in the NUREG series, NRC regulations, The NRC Technical Library and Title 10, Energy, in the Code of Federal Regulations Two White Flint North may also be purchased from one of these two sources. 11545 Rockville Pike Rockville, MD 20852-2738 | |||
: 1. The Superintendent of Documents U.S. Government Publishing Office These standards are available in the library for reference Mail Stop IDCC use by the public. Codes and standards are usually Washington, DC 20402-0001 copyrighted and may be purchased from the originating Internet: bookstore.gpo.gov organization or, if they are American National Standards, Telephone: (202) 512-1800 from Fax: (202) 512-2104 American National Standards Institute 11 West 42nd Street | |||
: 2. The National Technical Information Service New York, NY 10036-8002 5301 Shawnee Road www.ansi.org Alexandria, VA 22312-0002 (212) 642-4900 www.ntis.gov 1-800-553-6847 or, locally, (703) 605-6000 Legally binding regulatory requirements are stated only in A single copy of each NRC draft report for comment is laws; NRC regulations; licenses, including technical available free, to the extent of supply, upon written specifications; or orders, not in NUREG-series publications. | |||
request as follows: The views expressed in contractor prepared publications in this series are not necessarily those of the NRC. | |||
Address: U.S. Nuclear Regulatory Commission The NUREG series comprises (1) technical and adminis-Office of Administration trative reports and books prepared by the staff (NUREG-XXXX) or agency contractors (NUREG/CR-XXXX), (2) | |||
Multimedia, Graphics, and Storage & | |||
proceedings of conferences (NUREG/CP-XXXX), (3) reports Distribution Branch resulting from international agreements (NUREG/IA-XXXX), | |||
Washington, DC 20555-0001 (4) brochures (NUREG/BR-XXXX), and (5) compilations of E-mail: distribution.resource@nrc.gov legal decisions and orders of the Commission and Atomic Facsimile: (301) 415-2289 and Safety Licensing Boards and of Directors decisions under Section 2.206 of NRCs regulations (NUREG-0750). | |||
Some publications in the NUREG series that are posted at NRCs Web site address www.nrc.gov/reading-rm/ DISCLAIMER: This report was prepared under an doc-collections/nuregs are updated periodically and may international cooperative agreement for the exchange of technical information. Neither the U.S. Government nor any differ from the last printed version. Although references to agency thereof, nor any employee, makes any warranty, material found on a Web site bear the date the material expressed or implied, or assumes any legal liability or was accessed, the material available on the date cited responsibility for any third partys use, or the results of such may subsequently be removed from the site. use, of any information, apparatus, product or process disclosed in this publication, or represents that its use by such third party would not infringe privately owned rights. | |||
2019 Prepared as part of The Agreement on Research Participation and Technical Exchange Under the Thermal-Hydraulic Code Applications and Maintenance Program (CAMP) | NUREG/IA-0509 International Agreement Report LBLOCA Uncertainty Analysis of Maanshan Nuclear Power Plant with RELAP5/SNAP and DAKOTA Prepared by: | ||
Chunkuan Shih, Jong-Rong Wang, Jian-Ting Chen, Yuh-Ming Ferng, Shao-Wen Chen, Ting-Yi Wang *, and Tzu-Yao Yu* | |||
Institute of Nuclear Engineering and Science, National Tsing Hua University Nuclear and New Energy Education and Research Foundation 101 Section 2, Kuang Fu Rd., HsinChu, Taiwan | |||
* Department of Nuclear Safety, Taiwan Power Company 242, Section 3, Roosevelt Rd., Zhongzheng District, Taipei, Taiwan K. Tien, NRC Project Manager Division of Systems Analysis Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington, DC 20555-0001 Manuscript Completed: June 2018 Date Published: April 2019 Prepared as part of The Agreement on Research Participation and Technical Exchange Under the Thermal-Hydraulic Code Applications and Maintenance Program (CAMP) | |||
Published by U.S. Nuclear Regulatory Commission | |||
ABSTRACT The objective of this study is to assess the applicability of the RELAP5/MOD3.3 model of Maanshan NPP on LBLOCA transient. Maanshan NPP was the first three-loop PWR in Taiwan constructed by Westinghouse. The total power of the two units of the power plant is 2785 MWt, which consist of 2775 MWt for reactor power and 10 MWt for cooling pumps. For the last few years, the TRACE model of Maanshan NPP was developed and several kinds of transient event including the LOCA, AOO and the URG transient were performed and verified. Recently, the RELAP5/MOD3.3 code is another important development priority for our group. In 2015, the RELAP5/MOD3.3 model of Maanshan NPP was developed with SNAP interface. | |||
An uncertainty analysis methodology of the postulated LBLOCA for Maanshan NPP using RELAP5/MOD3.3, DAKOTA, and SNAP codes are established in this study. The uncertainty parameters are related to ECCS property. The ranges of uncertainty analysis are based on FSAR and Training Manual of Maanshan. The maximum PCT of the uncertainty analysis is 1293 K. The PCT of FSAR is 1383.7 K. Hence, the result of the uncertainty analysis is below the FSAR data. The difference of uncertainty analysis results and FSAR data may be caused by the different calculation procedures, phenomenological modeling, and nodalization, etc. | |||
Additionally, the PCT criteria of 10 CFR 50.46 is 1477.6 K (2200°F). The prediction of the uncertainty analysis is below the criteria of 10 CFR 50.46. This implies that Maanshan NPP in LBLOCA transient is at a safety situation when ECCS is performed. | |||
iii | |||
and | FOREWORD The U.S. NRC has developed a thermal hydraulic analysis code, RELAP5, has been designed to perform best-estimate analysis of loss-of-coolant accidents, operational transients, and other accident scenarios in reactor systems. Models used include multidimensional two-phase flow, non-equilibrium thermo-dynamics, generalized heat transfer, reflood, level tracking, and reactor kinetics. Traditionally, the RELAP5 code analysis model was developed by ASCII file, which was not intelligible for the beginners of computer analysis. Fortunately, and graphic input interface, SNAP is developed by Applied Programming Technology Inc. and conducted by the U.S. NRC, the model development process becomes more convenient. | ||
To obtain the authorization of these codes, Taiwan and the United States have signed an agreement on CAMP (Code Applications and Maintenance Program) which includes the development and maintenance of RELAP5 code. NTHU (National Tsing Hua University) is the organization in Taiwan responsible for the application RELAP5 and SNAP in thermal hydraulic safety analysis. Therefore, the RELAP5/MOD3.3 model of Maanshan nuclear power plant has been developed. To expand the applicability of the RELAP5/MOD3.3 model, an uncertainty analysis methodology of the postulated LBLOCA is established in this study. | |||
v | |||
TABLE OF CONTENTS ABSTRACT ............................................................................................................................... iii FOREWORD ............................................................................................................................... v TABLE OF CONTENTS ........................................................................................................... vii LIST OF FIGURES .................................................................................................................... ix EXECUTIVE | |||
and | ==SUMMARY== | ||
........................................................................................................... xi ABBREVIATIONS AND ACRONYMS ..................................................................................... xiii 1 INTRODUCTION ............................................................................................................. 1 2 MODEL ESTABLISHMENT ............................................................................................ 3 2.1 Hydraulic Components ........................................................................................ 3 2.2 Control Systems ................................................................................................. 8 2.3 Reactor Kinetics................................................................................................. 11 3 METHODOLOGY .......................................................................................................... 19 4 LBLOCA BASE CASE...21 4.1 Initial Condition and Basic Assumption.....................21 4.2 Analysis Results.....22 4.2.1 Critical Heat Flux Phase...22 4.2.2 Upward Core Flow Phase....23 4.2.3 Downward Core Flow Phase.. 24 4.2.4 Refill Phase 24 4.2.5 Reflood Phase26 4.2.6 Long-Term Core Cooling..27 5 LBLOCA UNCERTAINTY ANALYSIS....29 5.1 Uncertainty Parameters....29 5.2 Results........29 6 CONCLUSIONS ............................................................................................................ 33 7 REFERENCES .............................................................................................................. 35 vii | |||
LIST OF FIGURES Figure 1 Reactor Core Components of Maanshan NPP in SNAP Interface ........................ 4 Figure 2 Components of First Loop of Maanshan NPP in SNAP Interface ......................... 5 Figure 3 RCP Properties of Maanshan NPP in SNAP Interface ......................................... 6 Figure 4 Heat Exchanger Component 250 of Maanshan NPP in SNAP Interface .............. 7 Figure 5 Heat Structure 2500 Properties in SNAP Interface ............................................... 7 Figure 6 Components in Secondary Side of Maanshan NPP in SNAP Interface ................ 8 Figure 7 Feedwater Control System in SNAP Interface ...................................................... 9 Figure 8 Steam Dump Control System in SNAP Interface ................................................ 10 Figure 9 Heater of the Pressurizer Control System in SNAP Interface ............................. 11 Figure 10 Pressurizer Injection Control System in SNAP Interface .................................... 11 Figure 11 Properties in Axial Direction of Heat Structure 1201 in SNAP ............................ 12 Figure 12 Properties in Radial Direction of Heat Structure 1201 in SNAP .......................... 13 Figure 13 Power Ratio for Branch 100 to 116 of Heat Structure 1201 ................................ 13 Figure 14 Doppler Effect Reactivity Feedback Table ......................................................... 14 Figure 15 Doppler Effect Heat Structure Weighting Factor ................................................. 15 Figure 16 Density Effect Reactivity Feedback Table .......................................................... 16 Figure 17 Moderator Density Effect Volume Weighting Factor ........................................... 16 Figure 18 Reactor Scram Reactivity Feedback Properties ................................................. 17 Figure 19 Uncertainty Parameters Setting in DAKOTA ...................................................... 20 Figure 20 RELAP5/SNAP LBLOCA Break Model............................................................... 22 Figure 21 LBLOCA Break Flow Rate ................................................................................. 23 Figure 22 Void Fraction in Vessel ...................................................................................... 24 Figure 23 Cladding Temperature at the Lower Part ........................................................... 25 Figure 24 Core Vapor Velocity ........................................................................................... 25 Figure 25 Core Collapsed Water Level .............................................................................. 26 Figure 26 Core Flow Regime during Reflooding Phase...................................................... 27 Figure 27 Cladding Temperature Axial Profile .................................................................... 28 Figure 28 Core Collapsed Water Level of 59 Cases .......................................................... 30 Figure 29 PCT of 59 Cases................................................................................................ 31 ix | |||
EXECUTIVE | |||
==SUMMARY== | |||
RELAP5/MOD3.3 Patch04 code, which was developed for light water reactor transient analysis at Idaho National Engineering Laboratory for U.S. NRC, is applied in this research. This code is often performed to support rulemaking, licensing audit calculations, evaluation of accident, mitigation strategies, evaluation of operator guidelines, and experiment planning analysis. Same as other thermal hydraulic analysis codes, RELAP5/MOD3.3 is based on nonhomogeneous and non-equilibrium model for the two-phase system. However, calculations in this code will be solved by a fast, partially implicit numerical scheme to permit economical calculation of system transients. It can produce accurate transient analysis results in relatively short time. | |||
SNAP is an interface of NPP analysis codes which developed by U.S. NRC and Applied Programming Technology, Inc. Different from the traditional input deck in ASCII files, the graphical control blocks and thermal hydraulic connections make researches comprehend the whole power plant and control system more easily. Due to these advantages, the RELAP5/MOD3.3 model of Maanshan NPP was developed with SNAP interface. | |||
DAKOTA code can be used in uncertainty analysis. DAKOTA is applied for the sampling of input parameters, the calculation of correlations, and ranking of input parameters. DAKOTA provides Monte Carlo sampling and Latin Hypercube sampling methods combined with various PDFs including normal, lognormal, uniform, logunifrom, hypergeometric, and user-supplied histograms. In addition, the uncertainty analysis of DAKOTA also can run with the SNAP interface. | |||
Maanshan NPP is located on the southern coast of Taiwan. Its nuclear steam supply system is a type of PWR designed and built by Westinghouse for Taiwan Power Company. In this research, a RELAP5/MOD3.3 model of Maanshan NPP for LBLOCA transient was developed. | |||
This research also presents an uncertainty analysis methodology of LBLOCA using RELAP5/MOD3.3, DAKOTA, and SNAP codes. | |||
The results of the RELAP5 analysis for LBLOCA base case indicate that the RELAP5/SNAP model is able to manage a LBLOCA simulation. The phenomenon of the base case is consistent with the LOFT experiment and the analysis results of Westinghouse. According to analysis results, the evaluation of quench front velocity during reflooding phase strongly affects the PCT. | |||
The performance of ECCS may affect reflooding phase. Therefore, an uncertainty analysis was performed in this research. | |||
Most of the uncertainty parameters are related to ECCS property in the uncertainty analysis. | |||
The ranges of uncertainty analysis are based on FSAR and Training Manual of Maanshan NPP. | |||
The maximum PCT of the analysis is 1293 K that includes the uncertainty effect. However, the PCT of Maanshan FSAR is 1383.7 K. The difference of analysis results and FSAR data may be caused by the different calculation procedures, phenomenological modeling, and nodalization, etc. In addition, the PCT criteria of 10 CFR 50.46 is 1477.6 K (2200). Hence, the result of the uncertainty analysis is below the criteria of 10 CFR 50.46. This indicates that Maanshan NPP in LBLOCA transient is at a safety situation when ECCS is performed. | |||
xi | |||
ABBREVIATIONS AND ACRONYMS ACC Accumulator BAF Bottom of Active Fuel CAMP Code Applications and Maintenance Program CLOF Completed Loss of Flow ECCS Emergency Core Cooling System FWPT Feedwater pump(s) Trip FSAR Final Safety Analysis Report HHSI High Head Safety Injection kg kilogram(s) | |||
LBLOCA Large Break Loss Of Coolant Accident LHSI Low Head Safety Injection MPa Megapascal(s) | |||
MSIVC Main Steam Isolation Valves Closure NPP Nuclear Power Plant NRC Nuclear Regulatory Commission NRWL Narrow Range Water Level NSSS Nuclear Steam Supply System NTHU National Tsing Hua University PDF Probability Distribution Function PORVs Power-Operated Relief Valves PWR Pressurized Light Water Reactor PLOF Partial Loss of Flow RCS Reactor Coolant System RCP Reactor Coolant Pump S/G Steam Generator SI Safety Injection SNAP Symbolic Nuclear Analysis Program SRV Safety/Relief Valves TAF Top of Active Fuel TRACE TRAC/RELAP Advanced Computational Engine WRWL Wide Range Water Level xiii | |||
1 INTRODUCTION Maanshan Nuclear Power Plant (NPP) is the third NPP in Taiwan. Also, it is the first Pressurized Water Reactor (PWR) located at the south of Taiwan. There are two units in the Maanshan NPP. The total power of the nuclear steam supply system is 2785 MWt, which consist of 2775 MWt for reactor power and 10 MWt for cooling pumps [1]. For the last few years, our group has developed the models of Taiwan NPPs with TRACE code in SNAP interface [2, 3]. Further, it is necessary to perform the NPP transients with several analysis codes so that the data results could be compared with each other to ensure the consistency. Therefore, the RELAP5/MOD 3.3 Patch04 code was chosen to develop a new Maanshan NPP model. Different from the traditional ASCII input deck, the RELAP5/MOD 3.3 model was developed with SNAP interface. | |||
DAKOTA is an uncertainty analysis code [5] and is applied for the sampling of input parameters, the calculation of correlations, and ranking of input parameters. The uncertainty quantification package of DAKOTA provides Monte Carlo sampling and Latin Hypercube sampling methods combined with various PDFs including normal, lognormal, uniform, logunifrom, hypergeometric, and user-supplied histograms. | |||
SNAP is an interface of NPP analysis codes which developed by U.S. NRC and Applied Programming Technology, Inc. Different from the traditional input deck in ASCII files, the graphical control blocks and thermal hydraulic connections make researches comprehend the whole power plant and control system more easily [6]. Due to these advantages, an uncertainty analysis methodology of LBLOCA using RELAP5/MOD3.3 and DAKOTA with SNAP interface is established in this study. Moreover, due to the SNAP interface, the analysis results could be transferred into animations which were more attractive and more understandable. With the animation, interactions of different components and parameters could be easily observed. | |||
To ensure the applicability of this model, three startup tests cases including feedwater pumps trip (FWPT), turbine trip (PAT50) and main isolation valves closure (MSIVC) had been analyzed in 2015 [7]. With the comparison of RELAP5 results and startup tests data, it shows that the RELAP5/MOD 3.3 model of Maanshan NPP is consistent with the startup tests data. To expand the applicability of this RELAP5 model, the postulated LBLOCA is performed. Two main steps are in this study. First, the establishment of RELAP5/MOD3.3 model for the postulated LBLOCA of Maanshan NPP is performed. An analysis of LBLOCA base case is performed by using this model to confirm the phenomenon of LBLOCA. Second, to confirm the uncertainty of ECCS that may affect the PCT, the uncertainty analysis of RELAP5/MOD 3.3 model combining with DAKOTA is performed in this study. | |||
1 | |||
was | 2 MODEL ESTABLISHMENT As the Maanshan NPP operated in normal conditions, coolant water in primary system will carry the heat generated by the fuel rods to the steam generator. Feedwater in the secondary system then obtain the heat, evaporate and drive turbines to generate electricity. According to the energy conservation principle, internal energy of steam, which had driven turbines, will decrease. This lower internal energy steam will then go through the condenser and be transferred into feedwater and re-injected into the steam generator. However, it is difficult to develop the entire recirculation system with the analysis code. The computational time will be impractically long and the mass balance will be hard to reach. Hence, as developing the RELAP5 model, it is practical to define the feedwater pumps and the turbines as the boundary conditions because the main purpose of this model is to obtain the NSSS reactions during the transient. For the NSSS system of Maanshan NPP, the feedwater pumps, auxiliary feedwater pumps, turbines, safety/relief valves, steam dump valves and PORVs were defined as the boundary conditions and developed by the Time Dependent Volume component in the RELAP5 program [7]. | ||
2.1 Hydraulic Components As mentioned in section 2, there are three recirculation loops in the Maanshan NPP. In each loop, there is a RCP and S/G. On the hot leg of second loop, a pressurizer which can adjust the pressure of RCS with the spray valves and electronic heater was developed. In this analysis model, there are several Branch components developed to simulate the reactor vessel. | |||
According to the core arrangement, Branch components from number 140 to 156 were connected together as the average fuel channel. Branch components from number 120 to 136 were connected together as the hottest fuel channel. Branch components from number 100 to 116 were connected together as the bypass flow channel, as shown in Figure 1. Also, these channels will be connected to the heat structure components to obtain the heat and do the reactor kinetic analysis. | |||
For those three recirculation loops in primary side, they were developed by Pipe, Valve, Branch, Jump and Single Volume, as shown in Figure 2. For these three-digit components, the first digit stands for the loop number (2 for first loop, 3 for second loop and 4 for third loop). Further, the other digits of these components represents to the component types. For instance, component 280 is the recirculation pumps in first loop and component 380 is the recirculation pumps in second loop. Though the Pump component in RELAP5 code has been developed with the pump parameters from Westinghouse, pump characters of the RCPs in this model was input according to Taiwan Power Company NPP training materials and past research models which were calculated by RELAP5-3D and TRACE codes, as shown in Figure 3. | |||
In addition to RCPs, another important thermal hydraulic component in the primary side is heat exchanger. Pipe 250, which was developed for heat exchanger in first loop, was divided into eight nodes. According to the geometry of heat exchanger, junction between fourth cell and fifth cell was 180 degree as shown in Figure 4. Further, the heat structure component can be view as structural component once the both side of the heat structure were connected to thermal hydraulic components. Hence, the Pipe 250 was connected to the left boundary of heat structure 2500, as shown in Figure 5. Likewise, the Pipe 350 of secondary loop was connected to the left boundary of heat structure 3500 and Pipe 450 of third loop was connected to the left boundary of heat structure 4500. | |||
3 | |||
Figure 1 Reactor Core Components of Maanshan NPP in SNAP Interface 4 | |||
Figure 2 Components of First Loop of Maanshan NPP in SNAP Interface Similar to primary loops, the secondary loops of Maanshan NPP were developed with Pipe, Valve, Branch, Pump and Single Volume. Specially, to simulate the feedwater, auxiliary feedwater and steam dump systems, which flow rate was determined by system feedback, the Time Dependent Junction was used. With the same rules of primary loops, the components number in secondary loops was numbered in three digits. The first digit stands for the loops number and the other two digits stand for the component types. For instance, the component 520 were heat exchanger in first loop because the first digit 5 represents the first loop and the latter two digits 20 represents the heat exchanger. Component 520 was connected to the right boundary of heat structure 2500, which allows the heat transfer from component 250 in primary side to component 520 in secondary side. Due the heat from primary side, the water in component 520 will evaporate and go through the next component 522. Component 522 was a separator which can increase the quality up to 99.7%. This dried steam will then leave the separator and go through the main steam line isolation valve (component 543), turbine control valve (component 774), turbine stop valve (component 775) and drive turbine, as shown in Figure 6. | |||
As mentioned in section 2, the steam dump system was composed by 10 steam dump valves, six turbine bypass valves and several controlling equipment. To save the computational time, this RELAP5 model merged 10 steam dump valves into four groups. Each group was developed by a Time Dependent Junction component which the total steam flow rate was consistent to the operating conditions. Likewise, six turbine bypass valves were developed by two Time Dependent Junction components. | |||
5 | |||
Figure 3 RCP Properties of Maanshan NPP in SNAP Interface 6 | |||
Figure 4 Heat Exchanger Component 250 of Maanshan NPP in SNAP Interface Figure 5 Heat Structure 2500 Properties in SNAP Interface To simplify the feedwater control system, the feedwater pumps and valves were developed by Time Dependent Volume and Time Dependent Junction respectively. For the Time Dependent Volume components, the fluid boundary conditions were referred to the thermal hydraulic properties of feedwater during operation. Therefore, the control system need only concern the effect of NRWL, steam flow rate and feedwater flow rate to determine the feedwater flow rate. | |||
Once the flow rate was determined, Time Dependent Junctions which related to the control 7 | |||
blocks in the feedwater control system will inject the adequate feedwater into recirculation loops. | |||
Details of the feedwater control system will be discussed in the following section. | |||
Figure 6 Components in Secondary Side of Maanshan NPP in SNAP Interface 2.2 Control Systems In operation, the purpose of water level/feedwater control system is to ensure that water in the steam generator can cover the heat exchanger. For Maanshan NPP, the feedwater flow rate was determined by three units including NRWL in steam generator, steam flow rate and feedwater flow rate. As the water level deviate the setting values, the control system will adjust the injection of the feedwater flow rate to maintain the water level of the steam generator. | |||
Further, two water level measuring systems including the NRWL and WRWL calculate water level with pressure difference. Different from the TRACE model of Maanshan NPP which our group had developed before, there is no water level sensor signal component in the RELAP5 code. As a result, the measurements of water level were developed and composed with density, pressure and volume signal components which was shown in Figure 7. | |||
8 | |||
Figure 7 Feedwater Control System in SNAP Interface In addition to the feedwater control, the steam dump system was also an important response mechanism. As mentioned above, the steam dump system of Maanshan NPP can be divided into two types including the pressure control mode and the Tave mode. The pressure mode was initiated as core power was in range from 0% to 10%, which will not be discussed and applied in this research. Hence, the setting of the steam dump system was only referred to the response of Tave mode. As shown in Figure 8, there are three control blocks with sum function calculated the average core temperature values of loop 1 to loop 3 respectively. Then, the control block 308 with max function will compare the maximum of average core temperature (Tave) in loop 1 to loop 3 with No Load Temperature (Tno load, 564K in Maanshan NPP). | |||
Referring this comparison, the control blocks 318 can convert the difference of Tave and Tno load into steam dump flow rate with Table 15. As the temperature difference exceeded 0% | |||
(0), the first group of dump valves was opened. As the temperature difference exceeded 16% | |||
(15.8), the first groups of dump valves were fully opened and the second group of dump valves started to open and so on. | |||
The pressure and water level control system of pressurizer includes the heater and the spray valve. There two types of heater including control heater and backup heater. The control heater and spray valves were applied for adjusting the pressure inside the pressurizer. From Figure 9, the pressure of pressurizer will be compared to rated pressure in control block 120. With the comparison of these two pressure values, the difference can be transferred into the open of the spray valve (control blocks 123) and the power of heater (control block 121 and heat structure 1212 and 1222). However, the control heater is also related to the lower water level of 9 | |||
pressurizer (control block 121). As the water level has been lower than 14%, the power of control heater will be zero (control block 124), which means the heater trip. In addition, if the trip setting of control block 121 was assigned to other trip signals, then the control heater can be tripped manually. | |||
Figure 8 Steam Dump Control System in SNAP Interface The backup heater is related to the charging control system of pressurizer. As shown in Figure 10, the maximum core temperature will be transferred into program water level through control block 130. Then, the water level will be subtracted from the actual water level. If the difference of these two water levels is larger than 5%, the backup heater will be initiated (control block 132). Further, the water level will be transferred into charging flow rate (control block 136) to adjust the water level inside the pressurizer. However, as the safety injection signal is initiated, the charging flow rate will be forced to zero. | |||
10 | |||
Figure 9 Heater of the Pressurizer Control System in SNAP Interface Figure 10 Pressurizer Injection Control System in SNAP Interface 2.3 Reactor Kinetics In this RELAP5 model of Maanshan NPP, there are two sets of heat structures, which component numbers are 1201 and 1601, developed to simulate the hot fuel channel and average fuel channel. These heat structures were divided into 16 nodes (shown in Figure 11) in axial and 7 nodes in radial (shown in Figure 12). For the axial nodes, they were connected to the Brach components of the reactor core respectively. For radial nodes, the first 4 nodes stand for fuel pellets; the fifth node is filled helium inside the fuel rod and the sixth and seventh nodes 11 | |||
are fuel cladding. The materials of each node can be defined manually. In this model, thermal properties (thermal conductivity and thermal capacity) of material 1 for the first 4 nodes were referred to that of Uranium dioxide. The material 2 for node 5 was referred to the helium thermal properties and the material 3 for node 6 and 7 was referred to that of zircalloy. | |||
Figure 11 Properties in Axial Direction of Heat Structure 1201 in SNAP Heat source of the heat structure can be set with the total reactor power or power table. In this model, at the beginning of the model assessment the heat source was set with power table which was referred to the startup test data results of Maanshan NPP to ensure the applicability of the thermal hydraulic components. After that, heat source of the heat structure would be set with total reactor power to ensure the point kinetic feedback calculations. For those heat structure components which were developed as the fuel bundles, the left boundaries were set as symmetry and the right boundaries were connected to the Branch components. For these connections of Brach components and heat structures, the power ratio should be defined (as shown in Figure 13) respectively to calculate the correct heat transfer. The power ratio setting of this RELAP5 model was referred to the TRACE models which were fully developed and assessed before. | |||
12 | |||
Figure 12 Properties in Radial Direction of Heat Structure 1201 in SNAP Figure 13 Power Ratio for Branch 100 to 116 of Heat Structure 1201 13 | |||
For the point kinetic model, in addition to defining the power ratio of each node of heat structures, the ratio and position of reactivity feedback should also be defined. The reactivity feedback is dominated by Doppler Effect and Moderator density effect. The previous one is related to the temperature of fuel rods; hence, the check list of fuel temperature and reactivity should be added into the Power components. With this table (as shown in Figure 14), the RELAP5 code can calculate the corresponding reactivity feedback due to fuel temperature. | |||
Further, the fuel temperature feedback ratio should also be defined manually in the Heat Weighting settlement of Power component (shown in Figure 15). Similarity, to calculate the Moderator feedback, the checklist of coolant density and reactivity should be defined (shown in Figure 16). Then, with the volume weighting list (shown in Figure 17), the Branch components which were developed as the reactor core would be connected to the point kinetic calculation. | |||
With these settings, the RELAP5 code could calculate the power variation due to temperature and density changes inside the reactor core during transient events. | |||
Figure 14 Doppler Effect Reactivity Feedback Table As mentioned above, the startup assessment transient events were calculated with power table first to ensure the applicability of thermal hydraulic components. Then, the point kinetic model would be applied to do the whole assessment of Maanshan RELAP5 model. As performed with power table, the RELAP5 model needs no control system to simulate reactor scram. However, when performed with point kinetic model, the reactor scram control is required. For instance, Table 100 is the scram reactivity feedback table which will start to dominate the power variation once the trip logic/variable gate is initiated as shown in Figure 18. From this figure, it is obvious that the table could cause a large negative reactivity feedback in few seconds to simulate the control rods insertion. | |||
14 | |||
Figure 15 Doppler Effect Heat Structure Weighting Factor 15 | |||
Figure 16 Density Effect Reactivity Feedback Table Figure 17 Moderator Density Effect Volume Weighting Factor 16 | |||
Figure 18 Reactor Scram Reactivity Feedback Properties 17 | |||
3 METHODOLOGY This methodology has four steps which are (1) develop the basic analysis model using RELAP5. | |||
(2) ranking of plant status parameters. (3) ranging of plant status uncertainty parameters (4) generate the input files by random sampling. | |||
(1) Develop the basic analysis model using RELAP5: | |||
RELAP5/MOD3.3 code with SNAP interface is used to do the calculations in this research. The RELAP5 Maanshan NPP analysis model had been developed few years ago. Several verifications and validations had been done, including main steam line isolation valve closure (MSIVC), feedwater pumps trip (FWPT), turbine trip (PAT50), | |||
partial loss of flow (PLOF), complete loss of flow (CLOF) [7]. | |||
(2) Ranking of plant status parameters: | |||
Before the uncertainty analysis, it is necessary to do the preliminary analysis or study to determine which uncertainty parameters are important. In this research, a LBLOCA base simulation was done to identify the uncertainty parameters. Initial conditions were according to the real operation conditions. Initial power was set to be 2822MWt and assumed all the systems were operated in the normal status. The uncertainty parameters were evaluated based on the results of the base case, including pressurizer pressure, accumulator (ACC) pressure, ACC volume, ACC water temperature, safety injection (SI) water temperature, feedwater temperature. Detail of the LBLOCA base analysis is indicated in next chapter. | |||
(3) Ranging of plant status uncertainty parameters: | |||
To define the uncertainty parameters of plant status, the uncertainty range and the distribution function are both needed. The uncertainty range in this research are according to Maanshan FSAR and Maanshan Training Manual. The distribution function is set to be uniform distribution in this research. Detail of the uncertainty parameters are listed in chapter. | |||
(4) Generate the input files by random sampling: | |||
DAKOTA code is used to do the sampling in this research. RELAP5 and DAKOTA code are coupled on the SNAP interface. The input uncertainty parameters were sampled and generated the RELAP5 input files. Figure 19 is an example of parameters input, uncertainty range and distribution function setting in DAKOTA. | |||
Wilks formula was adopted to determine the minimum number of runs. In nuclear safety analysis, only upper bond of PCT is the most concern. For the one-sided statistical tolerance limits, the minimum number of runs is given by Wilks formula: | |||
1 is the percentage of tolerance limit, is the confidence level, n is the minimum number of runs. For example, the analysis results in this research achieve / = 95/95 criterion, it means there is 95% confidence level that the results are below the 95% of tolerance limit. To achieve the criterion, the minimum number of runs n = 59. | |||
18 | |||
Figure 19 Uncertainty Parameters Setting in DAKOTA 19 | |||
4 LBLOCA BASE CASE 4.1 Initial Condition and Basic Assumption The initial conditions and the ECCS properties are listed in Table 1. The initial power is 2822 MWt. The LBLOCA base case was performed in this study. It was assumed that the break occurred at the primary side, at first loop cold-leg near the vessel. The break type was assumed to be double-ended and the break model is shown in Figure 20. Before the pipe rupture, 400 sec steady-state calculation had been done to ensure all the system operated correctly. In the LBLOCA transient, High Head Safety Injection (HHSI), ACC injection, and Low Head Safety Injection (LHSI) are included. It was assumed that all the safety injection systems could operate normally. The safety injection signal setpoint is 11.8 MPa and delay time is 27 sec. The ACC injection set point is 4.24 MPa. The minimum operating pressure of Low Head Safety Injection (LHSI) is 1.15 MPa. | |||
Table 1 Initial Condition and ECCS properties Parameters Values Initial power (MWt) 2822 Core flow rate (kg/s) 14176 Tavg* (K) 582.97 RCS* pressure (MPa) 15.862 Scram setpoint (MPa) 12.8 SI* signal setpoint (MPa) 11.8 SI signal delay (s) 27 ACC liquid volume (m3/tank) 27.9 ACC injection setpoint (MPa) 4.24 ACC liquid temperature (K) 311 LHSI operating pressure(MPa) 1.15 Break size (m2) 0.38 21 | |||
Figure 20 RELAP5/SNAP LBLOCA Break Model 4.2 Analysis Results The LBLOCA transient can be divided into six stages, which are critical heat flux phase, upward core flow phase, downward core flow phase, refill phase, reflood phase, and long term cooling phase. The exact postulated event sequence of Maanshan NPP is listed in Table 2. | |||
Table 2 Postulated LBLOCA Events Sequence of Maanshan NPP Action Time (sec) | |||
Pipe rupture 0 Scram 1.8 RCP trip 1.8 SI signal 2.8 ACC injection 5.5 HHSI 29.8 LHSI 29.8 ACC injection ended 69.5 End of event 700 4.2.1 Critical Heat Flux Phase Immediately following the cold-leg break, the fluid discharged out of the pipe with very high flow rates, which can be observed in Figure 21. Due to the extremely high break flow rate, the system pressure drop rapidly. This depressurized phenomenon resulted in water flashing in the 22 | |||
reactor vessel. The void fraction of vessel is shown in Figure 22, which indicates that the void fraction went up to one in very few seconds. Because of the flashing phenomenon in reactor core, the fuel cladding became fully uncovered. As shown in Figure 23, the cladding temperature increasing dramatically to the first peak in the very early period. | |||
Figure 21 LBLOCA Break Flow Rate 4.2.2 Upward Core Flow Phase The coolant in loop system still had great inertia at the very beginning (0 ~ 5 sec), and flashing phenomenon in the reactor vessel also generated a great pressure drop. Therefore, there was some two-phase mixture could be pulled into the core, which made the cladding temperature decrease. Because of the bottom-up mixture cooling effect, the temperature drop was more distinct at the lower part. This phenomenon can be seen in Figure 23, there was a temperature drop just after the first peak. | |||
23 | |||
Figure 22 Void Fraction in Vessel 4.2.3 Downward Core Flow Phase As pump head continues to degrade, and the coolant slow down. The pressure drop at break started to fully dominate and pull flow down through the core, up through the downcomer to the break. Figure 24 shows the vapor velocity in reactor core, which indicate a velocity increasing downward at about 6 ~ 10 sec. High speed vapor flow also provided good core cooling. This phenomenon can be seen in Figure 23, which shows a second temperature drop at about 10 sec. | |||
4.2.4 Refill Phase Following the blowdown phase, the water level was below the BAF (see Figure 25). Therefore, the cladding temperature increased (see Figure 23). This period is featured of rapid cladding temperature increasing because of lacking water and steam cooling. | |||
24 | |||
Figure 23 Cladding Temperature at the Lower Part Figure 24 Core Vapor Velocity 25 | |||
4.2.5 Reflood Phase Due to the ACC injection, the water level continued to increase. Figure 25 shows the core collapsed water level. At about 19 sec, the coolant reached the BAF and lower core region began to quench. During this period, core cooling might increase because of upward vapor flow and liquid entrainment. Therefore, the cladding temperature could decrease before the coolant recovered the fuel cladding. | |||
In general, if the income core flow velocity exceeded certain value, it was assumed that the flow regime near the quench front was Inverted-Slug flow. The references [8~10] indicate that the minimum core coolant inlet velocity should be 0.15m/s to form an Inverted-Slug flow regime. | |||
These studies also indicate that the quenching temperature depends strongly on the core inlet coolant velocity. | |||
The flow regime during reflooding phase was shown in Figure 26. Under the quench front, fuel cladding was covered by liquid and cladding temperature dropped to saturation temperature. At inverted-annular and droplet region, the two phase mixture still provided good cooling effect that made the cladding temperature go down. Therefore, this period is the most important in LBLOCA analysis. Figure 27 shows the quench and rewet points for the nodes. | |||
Figure 25 Core Collapsed Water Level 26 | |||
4.2.6 Long-Term Core Cooling Whole fuel cladding was recovered by coolant at about 500 sec, which can be seen in Figure | |||
: 27. With the pumped safety injection operating continuously, the water will never below the TAF again. | |||
Figure 26 Core Flow Regime during Reflooding phase 27 | |||
Figure 27 Cladding Temperature Axial Profile 28 | |||
5 LBLOCA UNCERTAINTY ANALYSIS 5.1 Uncertainty Parameters According to the previous Chapter, the evaluation of quench front progressing during the reflooding phase strongly affects the cladding temperature. The quench front velocity is a relation between TQ and h [11, 12]. TQ is the quenching temperature, and h is the heat transfer coefficient of the coolant under the quench front. TQ and h strongly depend on system pressure and subcooling of the coolant. In LBLOCA case, the performance of ECCS may affect the PCT, system pressure, and subcooling of the coolant. Therefore, the uncertainty analysis by using RELAP5, DAKOTA, and SNAP codes was performed in this study. The uncertainty parameters which are mainly the properties of ECCS are presented in Table 3. These parameters include pressurizer pressure, accumulator (ACC) pressure, ACC volume, ACC water temperature, safety injection (SI) water temperature, and feedwater temperature. The uncertainty ranges are based on FSAR and Training Manual of Maanshan NPP. The distribution functions are all set to be uniform distribution. According to the Wilks formula, 59 of runs should be conducted to achieve the 95/95 criterion. These uncertainty parameters would be randomly sampled and generated 59 RELAP5 input files. | |||
Table 3 Uncertainty Parameters Parameter Deviation PDFs Reference 15.86 MPa 1 PZR pressure FSAR | |||
(+/-0.21MPa) 2 ACC pressure 4.14 - 4.55 MPa Training manual 27.9 m3 (+/-0.425 3 ACC volume Training manual m3) Uniform distribution 4 ACC temperature 310.93 - 338.71 K Training manual 5 SI temperature 300 K (+19.4 K) FSAR Feedwater 6 500 K (+/-1.22 K) Training manual temperature 5.2 Results The core collapsed water level can be seen in Figure 28, which shows the time of end-of-blowdown is estimated at about 19 sec. Reflooding begin at about 37 sec, the time which the coolant reaches the BAF. PCT of 59 cases was shown in Figure 29. During blowdown and refilled phases, the temperature results of 59 cases are very similar. The time of the first temperature peak of 59 cases all occur at about 7 sec and the value of all cases are also almost identical (1293 K). In the beginning of the pipe break, a severe depressurization causes liquid in reactor core suddenly transfer to vapor phase. The heat transfer coefficient of vapor is much lower than liquid makes the cladding temperature reach to the first peak at about 7 sec. It is obvious that the first peak is caused by the heat transfer coefficient difference between liquid and vapor. Therefore, these six uncertainty parameters, which are related to ECCS, have no effect on the first peak. After the coolant reaches BAF, both PCT and water level curves start to have distinct deviations. During this period, the coolant started to attach the lower part of fuel 29 | |||
cladding. Some liquid droplets would be entrained by vapor from lower part to upper part that makes the cladding temperature do not increase dramatically like the first peak. Therefore, the deviations of ECCS properties made the cooling effect different and makes different PCT. The red curve in Figure 29 shows the highest PCT of all cases, which is about 1293 K. The PCT of LBLOCA in Maanshan FSAR is 1383.7 K. The difference of analysis results and FSAR data may be caused by the different calculation procedures, phenomenological modeling, and nodalization, etc. In addition, the PCT criteria of 10 CFR 50.46 is 1477.6 K (2200). Comparing the criteria and the results of uncertainty analysis, the results of uncertainty analysis are below the criteria. | |||
Figure 28 Core Collapsed Water Level of 59 Cases 30 | |||
Figure 29 PCT of 59 Cases 31 | |||
31 | |||
6 CONCLUSIONS This study presents an uncertainty analysis methodology of a postulated LBLOCA of Maanshan NPP using RELAP5/MOD3.3, DAKOTA, and SNAP codes. The results in base case analysis shows that the model of Maanshan NPP is able to manage a LBLOCA simulation. The phenomenon of the base case is consistent with the LOFT experiment and the analysis results of Westinghouse [13, 14]. The evaluation of quench front velocity during reflooding phase strongly affects the PCT. The performance of ECCS may affect reflooding phase. Therefore, an uncertainty analysis was performed in this study. | |||
For the uncertainty analysis, most of the uncertainty parameters are related to ECCS property. | |||
The ranges of uncertainty analysis are based on FSAR and Training Manual of Maanshan NPP. | |||
The maximum PCT of the analysis is 1293 K. This result includes the uncertainty effect. The PCT of Maanshan FSAR is 1383.7 K. Therefore, the prediction of the uncertainty analysis is below the FSAR data. The difference of analysis results and FSAR data may be caused by the different calculation procedures, phenomenological modeling, and nodalization, etc. In addition, the PCT criteria of 10 CFR 50.46 is 1477.6 K (2200). Hence, the result of the uncertainty analysis is also below the criteria of 10 CFR 50.46. This indicates that Maanshan NPP in LBLOCA transient is at a safety situation when ECCS is performed. | |||
33 | |||
7 REFERENCES | |||
: 1. Taiwan Power Company, Final Safety Analysis Report for Maanshan Nuclear Power Station Units 1&2 (FSAR), Taiwan Power Company, Republic of China (Taiwan), 1983. | |||
: 2. Jong-Rong Wang, Che-Hao Chen, Hao-Tzu Lin, Chunkuan Shih, Assessment of LONF ATWS for Maanshan PWR Using TRACE Code, NUREG/IA-0436, 2014. | |||
: 3. Hao-Tzu Lin, Jong-Rong Wang, Kai-Chun Huang, Chunkuan Shih, Show-Chyuan Chiang, Chia-Chuan Liu, Station Blackout Mitigation Strategies Analysis for Maanshan PWR Plant using TRACE, Annals of Nuclear Energy, 89, 2016. | |||
: 4. Information Systems Laboratories, Inc., RELAP5/MOD3.3 Code Manual Volume I: Code Structure, System Models, and Solution Methods, Information Systems Laboratories, Inc., Rockville Maryland, 2010. | |||
: 5. Adams, B. M., et al., DAKOTA, a Multilevel Parallel Object-oriented Framework for Design Optimization, Parameter Estimation, Uncertainty Quantification, and Sensitivity Analysis, Sandia National Laboratories, 20-21, 66-67, 99, 2011. | |||
: 6. Applied Programming Technology, Symbolic Nuclear Analysis Package (SNAP) User's Manual, Applied Programming Technology Inc., Bloomsburg, 2007. | |||
: 7. Chunkuan Shih, Jong-Rong Wang, Shao-Wen Chen, Hao-Chun Chang, Show-Chyuan Chiang, and Tzu-Yao Yu, RELAP5/MOD3.3 Model Assessment of Maanshan Nuclear Power Plant with SNAP Interface, NUREG/IA-0472, 2017. | |||
: 8. A.D. Siegel, Rewetting Temperature during Top and Bottom Reflood, M.S. Thesis, University of Illinois, 1979. | |||
: 9. A.D. Siegel and J.J. Carbajo, Trans. ANS 35, pp. 327-328, 19. | |||
: 10. Kevin D. Kimball, Quench Front Propagation during Bottom Reflooding of a Heated Annular Channel, Nuclear Engineering and Design 76, pp. 79-88, 1983. | |||
: 11. Duffey, R.B., Porthouse, D.T.C., The Physics of Rewetting in Water Reactor Emergency Core Cooling. Nuclear Engineering and Design 25, pp. 379~394, 1973. | |||
: 12. B.D.G. Piggott and D.T.C. Porthouse, A Correlation of Rewetting Data. Nuclear Engineering and Design 32, pp. 171~181, 1975. | |||
: 13. S. M. Modro, S. M. Modro, V. T. Berta, A. B. Wahba, Review of LOFT Large Break Experiments, NUREG/IA-0028, 1989. | |||
: 14. Westinghouse Company, Best-Estimate Analysis of the Large-Break Loss-of-Coolant Accident for Maanshan Units 1 and 2 Nuclear Power Plant Using the ASTRUM Methodology, WCAP-17054-P, 2009. | |||
35 | |||
NUREG/IA-0509 LBLOCA Uncertainty Analysis of Maanshan Nuclear Power Plant with RELAP5/SNAP and DAKOTA April 2019 Chunkuan Shih, Jong-Rong Wang, Jiang-Ting Chen, Yuh-Ming Ferng, Technical Shao-Wen Chen, Ting-Yi Wang*, and Tzu-Yao Yu* | |||
Institute of Nuclear Engineering and Science | |||
* Department of Nuclear Safety, Taiwan Power Co. | |||
National Tsing Hua University 242, Section 3, Roosevelt Rd., Zhongzheng Dist. | |||
Nuclear and new Energy Education and Research Foundation Taipei, Taiwan 101 Section 2, Kuang Fu Rd., HsinChu, Taiwan Division of Systems Analysis Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington, D.C. 20555-0001 K. Tien, NRC Project Manager The objective of this study is to assess the applicability of the RELAP5/MOD3.3 model of Maanshan NPP on LBLOCA transient. Maanshan NPP was the first three-loop PWR in Taiwan constructed by Westinghouse. In 2015, the RELAP5/MOD3.3 model of Maanshan NPP was developed with SNAP interface. An uncertainty analysis methodology of the postulated LBLOCA for Maanshan NPP using RELAP5/MOD3.3, DAKOTA, and SNAP codes are established in this study. The uncertainty parameters are related to ECCS property. The ranges of uncertainty analysis are based on FSAR and Training Manual of Maanshan. The maximum PCT of the uncertainty analysis is 1293 K. The PCT of FSAR is 1383.7 K. Hence, the result of the uncertainty analysis is below the FSAR data. The difference of uncertainty analysis results and FSAR data may be caused by the different calculation procedures, phenomenological modeling, and nodalization, etc. | |||
Additionally, the PCT criteria of 10 CFR 50.46 is 1477.6 K (2200). The prediction of the uncertainty analysis is below the criteria of 10 CFR 50.46. This implies that Maanshan NPP in LBLOCA transient is at a safety situation when ECCS is performed. | |||
Probability Distribution Function (PDF) | |||
Feedwater Pump(s) Trip (FWPT) | |||
Complete Loss of Flow (CLOF) | |||
High Head Safety Injection (HHSI) | |||
Chunkuan Shih, Jong-Rong Wang, Jiang-Ting Chen, Yuh-Ming Ferng, Shao-Wen Chen, Ting-Yi Wang*, Tzu-Yao Yu* Department of Nuclear Safety, Taiwan Power Co. | |||
Maanshan NPP was the first three-loop PWR in Taiwan constructed by Westinghouse. | |||
In 2015, the RELAP5/MOD3.3 model of Maanshan NPP was developed with SNAP interface. An uncertainty analysis methodology of the postulated LBLOCA for Maanshan NPP using RELAP5/MOD3.3, DAKOTA, and SNAP codes are established in this study. The uncertainty parameters are related to ECCS property. The ranges of uncertainty analysis are based on FSAR and Training Manual of Maanshan. The maximum PCT of the uncertainty analysis is 1293 K. The PCT of FSAR is 1383.7 K.Hence, the result of the uncertainty analysis is below the FSAR data. The difference of uncertainty analysis results and FSAR data may be caused by the different calculation procedures, phenomenological modeling, and nodalization, etc. Additionally, the PCT criteria of 10 CFR 50.46 is 1477.6 K (2200). The prediction of the uncertainty analysis is below the criteria of 10 CFR 50.46. This implies that Maanshan NPP in LBLOCA transient is at a safety situation when ECCS is performed.Probability Distribution Function (PDF) Feedwater Pump(s) | |||
Top of Active Fuel (TAF) | Top of Active Fuel (TAF) | ||
Maanshan Nuclear | Maanshan Nuclear Power Plant (PWR) | ||
International Agreement Report LBLOCA Uncertainty Analysis of Maanshan Nuclear Power Plant with RELAP5/SNAP and DAKOTA Prepared by: Chunkuan Shih, Jong-Rong Wang, Jian-Ting Chen, Yuh-Ming Ferng, Shao-Wen Chen, Ting- | NUREG/IA-0509 International Agreement Report LBLOCA Uncertainty Analysis of Maanshan Nuclear Power Plant with RELAP5/SNAP and DAKOTA Prepared by: | ||
*Department of Nuclear Safety, Taiwan Power Company 242, Section 3, Roosevelt Rd., Zhongzheng District, Taipei, Taiwan K.Tien, NRC Project Manager | Chunkuan Shih, Jong-Rong Wang, Jian-Ting Chen, Yuh-Ming Ferng, Shao-Wen Chen, Ting-Yi Wang *, and Tzu-Yao Yu* | ||
Institute of Nuclear Engineering and Science, National Tsing Hua University Nuclear and New Energy Education and Research Foundation 101 Section 2, Kuang Fu Rd., HsinChu, Taiwan | |||
* Department of Nuclear Safety, Taiwan Power Company 242, Section 3, Roosevelt Rd., Zhongzheng District, Taipei, Taiwan K. Tien, NRC Project Manager Division of Systems Analysis Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington, DC 20555-0001 Manuscript Completed: June 2018 Date Published: April 2019 Prepared as part of The Agreement on Research Participation and Technical Exchange Under the Thermal-Hydraulic Code Applications and Maintenance Program (CAMP) | |||
Published by U.S. Nuclear Regulatory Commission | |||
AVAILABILITY OF REFERENCE MATERIALS IN NRC PUBLICATIONS NRC Reference Material Non-NRC Reference Material As of November 1999, you may electronically access Documents available from public and special technical NUREG-series publications and other NRC records at libraries include all open literature items, such as books, NRCs Library at www.nrc.gov/reading-rm.html. Publicly journal articles, transactions, Federal Register notices, released records include, to name a few, NUREG-series Federal and State legislation, and congressional reports. | |||
publications; Federal Register notices; applicant, Such documents as theses, dissertations, foreign reports licensee, and vendor documents and correspondence; and translations, and non-NRC conference proceedings NRC correspondence and internal memoranda; bulletins may be purchased from their sponsoring organization. | |||
and information notices; inspection and investigative reports; licensee event reports; and Commission papers Copies of industry codes and standards used in a and their attachments. substantive manner in the NRC regulatory process are maintained at NRC publications in the NUREG series, NRC regulations, The NRC Technical Library and Title 10, Energy, in the Code of Federal Regulations Two White Flint North may also be purchased from one of these two sources. 11545 Rockville Pike Rockville, MD 20852-2738 | |||
: 1. The Superintendent of Documents U.S. Government Publishing Office These standards are available in the library for reference Mail Stop IDCC use by the public. Codes and standards are usually Washington, DC 20402-0001 copyrighted and may be purchased from the originating Internet: bookstore.gpo.gov organization or, if they are American National Standards, Telephone: (202) 512-1800 from Fax: (202) 512-2104 American National Standards Institute 11 West 42nd Street | |||
: 2. The National Technical Information Service New York, NY 10036-8002 5301 Shawnee Road www.ansi.org Alexandria, VA 22312-0002 (212) 642-4900 www.ntis.gov 1-800-553-6847 or, locally, (703) 605-6000 Legally binding regulatory requirements are stated only in A single copy of each NRC draft report for comment is laws; NRC regulations; licenses, including technical available free, to the extent of supply, upon written specifications; or orders, not in NUREG-series publications. | |||
request as follows: The views expressed in contractor prepared publications in this series are not necessarily those of the NRC. | |||
Address: U.S. Nuclear Regulatory Commission The NUREG series comprises (1) technical and adminis-Office of Administration trative reports and books prepared by the staff (NUREG-XXXX) or agency contractors (NUREG/CR-XXXX), (2) | |||
Multimedia, Graphics, and Storage & | |||
proceedings of conferences (NUREG/CP-XXXX), (3) reports Distribution Branch resulting from international agreements (NUREG/IA-XXXX), | |||
Washington, DC 20555-0001 (4) brochures (NUREG/BR-XXXX), and (5) compilations of E-mail: distribution.resource@nrc.gov legal decisions and orders of the Commission and Atomic Facsimile: (301) 415-2289 and Safety Licensing Boards and of Directors decisions under Section 2.206 of NRCs regulations (NUREG-0750). | |||
Some publications in the NUREG series that are posted at NRCs Web site address www.nrc.gov/reading-rm/ DISCLAIMER: This report was prepared under an doc-collections/nuregs are updated periodically and may international cooperative agreement for the exchange of technical information. Neither the U.S. Government nor any differ from the last printed version. Although references to agency thereof, nor any employee, makes any warranty, material found on a Web site bear the date the material expressed or implied, or assumes any legal liability or was accessed, the material available on the date cited responsibility for any third partys use, or the results of such may subsequently be removed from the site. use, of any information, apparatus, product or process disclosed in this publication, or represents that its use by such third party would not infringe privately owned rights. | |||
2019 Prepared as part of The Agreement on Research Participation and Technical Exchange Under the Thermal-Hydraulic Code Applications and Maintenance Program (CAMP) | NUREG/IA-0509 International Agreement Report LBLOCA Uncertainty Analysis of Maanshan Nuclear Power Plant with RELAP5/SNAP and DAKOTA Prepared by: | ||
Chunkuan Shih, Jong-Rong Wang, Jian-Ting Chen, Yuh-Ming Ferng, Shao-Wen Chen, Ting-Yi Wang *, and Tzu-Yao Yu* | |||
Institute of Nuclear Engineering and Science, National Tsing Hua University Nuclear and New Energy Education and Research Foundation 101 Section 2, Kuang Fu Rd., HsinChu, Taiwan | |||
* Department of Nuclear Safety, Taiwan Power Company 242, Section 3, Roosevelt Rd., Zhongzheng District, Taipei, Taiwan K. Tien, NRC Project Manager Division of Systems Analysis Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington, DC 20555-0001 Manuscript Completed: June 2018 Date Published: April 2019 Prepared as part of The Agreement on Research Participation and Technical Exchange Under the Thermal-Hydraulic Code Applications and Maintenance Program (CAMP) | |||
Published by U.S. Nuclear Regulatory Commission | |||
ABSTRACT The objective of this study is to assess the applicability of the RELAP5/MOD3.3 model of Maanshan NPP on LBLOCA transient. Maanshan NPP was the first three-loop PWR in Taiwan constructed by Westinghouse. The total power of the two units of the power plant is 2785 MWt, which consist of 2775 MWt for reactor power and 10 MWt for cooling pumps. For the last few years, the TRACE model of Maanshan NPP was developed and several kinds of transient event including the LOCA, AOO and the URG transient were performed and verified. Recently, the RELAP5/MOD3.3 code is another important development priority for our group. In 2015, the RELAP5/MOD3.3 model of Maanshan NPP was developed with SNAP interface. | |||
An uncertainty analysis methodology of the postulated LBLOCA for Maanshan NPP using RELAP5/MOD3.3, DAKOTA, and SNAP codes are established in this study. The uncertainty parameters are related to ECCS property. The ranges of uncertainty analysis are based on FSAR and Training Manual of Maanshan. The maximum PCT of the uncertainty analysis is 1293 K. The PCT of FSAR is 1383.7 K. Hence, the result of the uncertainty analysis is below the FSAR data. The difference of uncertainty analysis results and FSAR data may be caused by the different calculation procedures, phenomenological modeling, and nodalization, etc. | |||
Additionally, the PCT criteria of 10 CFR 50.46 is 1477.6 K (2200°F). The prediction of the uncertainty analysis is below the criteria of 10 CFR 50.46. This implies that Maanshan NPP in LBLOCA transient is at a safety situation when ECCS is performed. | |||
iii | |||
and | FOREWORD The U.S. NRC has developed a thermal hydraulic analysis code, RELAP5, has been designed to perform best-estimate analysis of loss-of-coolant accidents, operational transients, and other accident scenarios in reactor systems. Models used include multidimensional two-phase flow, non-equilibrium thermo-dynamics, generalized heat transfer, reflood, level tracking, and reactor kinetics. Traditionally, the RELAP5 code analysis model was developed by ASCII file, which was not intelligible for the beginners of computer analysis. Fortunately, and graphic input interface, SNAP is developed by Applied Programming Technology Inc. and conducted by the U.S. NRC, the model development process becomes more convenient. | ||
To obtain the authorization of these codes, Taiwan and the United States have signed an agreement on CAMP (Code Applications and Maintenance Program) which includes the development and maintenance of RELAP5 code. NTHU (National Tsing Hua University) is the organization in Taiwan responsible for the application RELAP5 and SNAP in thermal hydraulic safety analysis. Therefore, the RELAP5/MOD3.3 model of Maanshan nuclear power plant has been developed. To expand the applicability of the RELAP5/MOD3.3 model, an uncertainty analysis methodology of the postulated LBLOCA is established in this study. | |||
v | |||
TABLE OF CONTENTS ABSTRACT ............................................................................................................................... iii FOREWORD ............................................................................................................................... v TABLE OF CONTENTS ........................................................................................................... vii LIST OF FIGURES .................................................................................................................... ix EXECUTIVE | |||
and | ==SUMMARY== | ||
........................................................................................................... xi ABBREVIATIONS AND ACRONYMS ..................................................................................... xiii 1 INTRODUCTION ............................................................................................................. 1 2 MODEL ESTABLISHMENT ............................................................................................ 3 2.1 Hydraulic Components ........................................................................................ 3 2.2 Control Systems ................................................................................................. 8 2.3 Reactor Kinetics................................................................................................. 11 3 METHODOLOGY .......................................................................................................... 19 4 LBLOCA BASE CASE...21 4.1 Initial Condition and Basic Assumption.....................21 4.2 Analysis Results.....22 4.2.1 Critical Heat Flux Phase...22 4.2.2 Upward Core Flow Phase....23 4.2.3 Downward Core Flow Phase.. 24 4.2.4 Refill Phase 24 4.2.5 Reflood Phase26 4.2.6 Long-Term Core Cooling..27 5 LBLOCA UNCERTAINTY ANALYSIS....29 5.1 Uncertainty Parameters....29 5.2 Results........29 6 CONCLUSIONS ............................................................................................................ 33 7 REFERENCES .............................................................................................................. 35 vii | |||
LIST OF FIGURES Figure 1 Reactor Core Components of Maanshan NPP in SNAP Interface ........................ 4 Figure 2 Components of First Loop of Maanshan NPP in SNAP Interface ......................... 5 Figure 3 RCP Properties of Maanshan NPP in SNAP Interface ......................................... 6 Figure 4 Heat Exchanger Component 250 of Maanshan NPP in SNAP Interface .............. 7 Figure 5 Heat Structure 2500 Properties in SNAP Interface ............................................... 7 Figure 6 Components in Secondary Side of Maanshan NPP in SNAP Interface ................ 8 Figure 7 Feedwater Control System in SNAP Interface ...................................................... 9 Figure 8 Steam Dump Control System in SNAP Interface ................................................ 10 Figure 9 Heater of the Pressurizer Control System in SNAP Interface ............................. 11 Figure 10 Pressurizer Injection Control System in SNAP Interface .................................... 11 Figure 11 Properties in Axial Direction of Heat Structure 1201 in SNAP ............................ 12 Figure 12 Properties in Radial Direction of Heat Structure 1201 in SNAP .......................... 13 Figure 13 Power Ratio for Branch 100 to 116 of Heat Structure 1201 ................................ 13 Figure 14 Doppler Effect Reactivity Feedback Table ......................................................... 14 Figure 15 Doppler Effect Heat Structure Weighting Factor ................................................. 15 Figure 16 Density Effect Reactivity Feedback Table .......................................................... 16 Figure 17 Moderator Density Effect Volume Weighting Factor ........................................... 16 Figure 18 Reactor Scram Reactivity Feedback Properties ................................................. 17 Figure 19 Uncertainty Parameters Setting in DAKOTA ...................................................... 20 Figure 20 RELAP5/SNAP LBLOCA Break Model............................................................... 22 Figure 21 LBLOCA Break Flow Rate ................................................................................. 23 Figure 22 Void Fraction in Vessel ...................................................................................... 24 Figure 23 Cladding Temperature at the Lower Part ........................................................... 25 Figure 24 Core Vapor Velocity ........................................................................................... 25 Figure 25 Core Collapsed Water Level .............................................................................. 26 Figure 26 Core Flow Regime during Reflooding Phase...................................................... 27 Figure 27 Cladding Temperature Axial Profile .................................................................... 28 Figure 28 Core Collapsed Water Level of 59 Cases .......................................................... 30 Figure 29 PCT of 59 Cases................................................................................................ 31 ix | |||
EXECUTIVE | |||
==SUMMARY== | |||
RELAP5/MOD3.3 Patch04 code, which was developed for light water reactor transient analysis at Idaho National Engineering Laboratory for U.S. NRC, is applied in this research. This code is often performed to support rulemaking, licensing audit calculations, evaluation of accident, mitigation strategies, evaluation of operator guidelines, and experiment planning analysis. Same as other thermal hydraulic analysis codes, RELAP5/MOD3.3 is based on nonhomogeneous and non-equilibrium model for the two-phase system. However, calculations in this code will be solved by a fast, partially implicit numerical scheme to permit economical calculation of system transients. It can produce accurate transient analysis results in relatively short time. | |||
SNAP is an interface of NPP analysis codes which developed by U.S. NRC and Applied Programming Technology, Inc. Different from the traditional input deck in ASCII files, the graphical control blocks and thermal hydraulic connections make researches comprehend the whole power plant and control system more easily. Due to these advantages, the RELAP5/MOD3.3 model of Maanshan NPP was developed with SNAP interface. | |||
DAKOTA code can be used in uncertainty analysis. DAKOTA is applied for the sampling of input parameters, the calculation of correlations, and ranking of input parameters. DAKOTA provides Monte Carlo sampling and Latin Hypercube sampling methods combined with various PDFs including normal, lognormal, uniform, logunifrom, hypergeometric, and user-supplied histograms. In addition, the uncertainty analysis of DAKOTA also can run with the SNAP interface. | |||
Maanshan NPP is located on the southern coast of Taiwan. Its nuclear steam supply system is a type of PWR designed and built by Westinghouse for Taiwan Power Company. In this research, a RELAP5/MOD3.3 model of Maanshan NPP for LBLOCA transient was developed. | |||
This research also presents an uncertainty analysis methodology of LBLOCA using RELAP5/MOD3.3, DAKOTA, and SNAP codes. | |||
The results of the RELAP5 analysis for LBLOCA base case indicate that the RELAP5/SNAP model is able to manage a LBLOCA simulation. The phenomenon of the base case is consistent with the LOFT experiment and the analysis results of Westinghouse. According to analysis results, the evaluation of quench front velocity during reflooding phase strongly affects the PCT. | |||
The performance of ECCS may affect reflooding phase. Therefore, an uncertainty analysis was performed in this research. | |||
Most of the uncertainty parameters are related to ECCS property in the uncertainty analysis. | |||
The ranges of uncertainty analysis are based on FSAR and Training Manual of Maanshan NPP. | |||
The maximum PCT of the analysis is 1293 K that includes the uncertainty effect. However, the PCT of Maanshan FSAR is 1383.7 K. The difference of analysis results and FSAR data may be caused by the different calculation procedures, phenomenological modeling, and nodalization, etc. In addition, the PCT criteria of 10 CFR 50.46 is 1477.6 K (2200). Hence, the result of the uncertainty analysis is below the criteria of 10 CFR 50.46. This indicates that Maanshan NPP in LBLOCA transient is at a safety situation when ECCS is performed. | |||
xi | |||
ABBREVIATIONS AND ACRONYMS ACC Accumulator BAF Bottom of Active Fuel CAMP Code Applications and Maintenance Program CLOF Completed Loss of Flow ECCS Emergency Core Cooling System FWPT Feedwater pump(s) Trip FSAR Final Safety Analysis Report HHSI High Head Safety Injection kg kilogram(s) | |||
LBLOCA Large Break Loss Of Coolant Accident LHSI Low Head Safety Injection MPa Megapascal(s) | |||
MSIVC Main Steam Isolation Valves Closure NPP Nuclear Power Plant NRC Nuclear Regulatory Commission NRWL Narrow Range Water Level NSSS Nuclear Steam Supply System NTHU National Tsing Hua University PDF Probability Distribution Function PORVs Power-Operated Relief Valves PWR Pressurized Light Water Reactor PLOF Partial Loss of Flow RCS Reactor Coolant System RCP Reactor Coolant Pump S/G Steam Generator SI Safety Injection SNAP Symbolic Nuclear Analysis Program SRV Safety/Relief Valves TAF Top of Active Fuel TRACE TRAC/RELAP Advanced Computational Engine WRWL Wide Range Water Level xiii | |||
1 INTRODUCTION Maanshan Nuclear Power Plant (NPP) is the third NPP in Taiwan. Also, it is the first Pressurized Water Reactor (PWR) located at the south of Taiwan. There are two units in the Maanshan NPP. The total power of the nuclear steam supply system is 2785 MWt, which consist of 2775 MWt for reactor power and 10 MWt for cooling pumps [1]. For the last few years, our group has developed the models of Taiwan NPPs with TRACE code in SNAP interface [2, 3]. Further, it is necessary to perform the NPP transients with several analysis codes so that the data results could be compared with each other to ensure the consistency. Therefore, the RELAP5/MOD 3.3 Patch04 code was chosen to develop a new Maanshan NPP model. Different from the traditional ASCII input deck, the RELAP5/MOD 3.3 model was developed with SNAP interface. | |||
DAKOTA is an uncertainty analysis code [5] and is applied for the sampling of input parameters, the calculation of correlations, and ranking of input parameters. The uncertainty quantification package of DAKOTA provides Monte Carlo sampling and Latin Hypercube sampling methods combined with various PDFs including normal, lognormal, uniform, logunifrom, hypergeometric, and user-supplied histograms. | |||
SNAP is an interface of NPP analysis codes which developed by U.S. NRC and Applied Programming Technology, Inc. Different from the traditional input deck in ASCII files, the graphical control blocks and thermal hydraulic connections make researches comprehend the whole power plant and control system more easily [6]. Due to these advantages, an uncertainty analysis methodology of LBLOCA using RELAP5/MOD3.3 and DAKOTA with SNAP interface is established in this study. Moreover, due to the SNAP interface, the analysis results could be transferred into animations which were more attractive and more understandable. With the animation, interactions of different components and parameters could be easily observed. | |||
To ensure the applicability of this model, three startup tests cases including feedwater pumps trip (FWPT), turbine trip (PAT50) and main isolation valves closure (MSIVC) had been analyzed in 2015 [7]. With the comparison of RELAP5 results and startup tests data, it shows that the RELAP5/MOD 3.3 model of Maanshan NPP is consistent with the startup tests data. To expand the applicability of this RELAP5 model, the postulated LBLOCA is performed. Two main steps are in this study. First, the establishment of RELAP5/MOD3.3 model for the postulated LBLOCA of Maanshan NPP is performed. An analysis of LBLOCA base case is performed by using this model to confirm the phenomenon of LBLOCA. Second, to confirm the uncertainty of ECCS that may affect the PCT, the uncertainty analysis of RELAP5/MOD 3.3 model combining with DAKOTA is performed in this study. | |||
1 | |||
2 MODEL ESTABLISHMENT As the Maanshan NPP operated in normal conditions, coolant water in primary system will carry the heat generated by the fuel rods to the steam generator. Feedwater in the secondary system then obtain the heat, evaporate and drive turbines to generate electricity. According to the energy conservation principle, internal energy of steam, which had driven turbines, will decrease. This lower internal energy steam will then go through the condenser and be transferred into feedwater and re-injected into the steam generator. However, it is difficult to develop the entire recirculation system with the analysis code. The computational time will be impractically long and the mass balance will be hard to reach. Hence, as developing the RELAP5 model, it is practical to define the feedwater pumps and the turbines as the boundary conditions because the main purpose of this model is to obtain the NSSS reactions during the transient. For the NSSS system of Maanshan NPP, the feedwater pumps, auxiliary feedwater pumps, turbines, safety/relief valves, steam dump valves and PORVs were defined as the boundary conditions and developed by the Time Dependent Volume component in the RELAP5 program [7]. | |||
2.1 Hydraulic Components As mentioned in section 2, there are three recirculation loops in the Maanshan NPP. In each loop, there is a RCP and S/G. On the hot leg of second loop, a pressurizer which can adjust the pressure of RCS with the spray valves and electronic heater was developed. In this analysis model, there are several Branch components developed to simulate the reactor vessel. | |||
According to the core arrangement, Branch components from number 140 to 156 were connected together as the average fuel channel. Branch components from number 120 to 136 were connected together as the hottest fuel channel. Branch components from number 100 to 116 were connected together as the bypass flow channel, as shown in Figure 1. Also, these channels will be connected to the heat structure components to obtain the heat and do the reactor kinetic analysis. | |||
For those three recirculation loops in primary side, they were developed by Pipe, Valve, Branch, Jump and Single Volume, as shown in Figure 2. For these three-digit components, the first digit stands for the loop number (2 for first loop, 3 for second loop and 4 for third loop). Further, the other digits of these components represents to the component types. For instance, component 280 is the recirculation pumps in first loop and component 380 is the recirculation pumps in second loop. Though the Pump component in RELAP5 code has been developed with the pump parameters from Westinghouse, pump characters of the RCPs in this model was input according to Taiwan Power Company NPP training materials and past research models which were calculated by RELAP5-3D and TRACE codes, as shown in Figure 3. | |||
In addition to RCPs, another important thermal hydraulic component in the primary side is heat exchanger. Pipe 250, which was developed for heat exchanger in first loop, was divided into eight nodes. According to the geometry of heat exchanger, junction between fourth cell and fifth cell was 180 degree as shown in Figure 4. Further, the heat structure component can be view as structural component once the both side of the heat structure were connected to thermal hydraulic components. Hence, the Pipe 250 was connected to the left boundary of heat structure 2500, as shown in Figure 5. Likewise, the Pipe 350 of secondary loop was connected to the left boundary of heat structure 3500 and Pipe 450 of third loop was connected to the left boundary of heat structure 4500. | |||
3 | |||
Figure 1 Reactor Core Components of Maanshan NPP in SNAP Interface 4 | |||
Figure 2 Components of First Loop of Maanshan NPP in SNAP Interface Similar to primary loops, the secondary loops of Maanshan NPP were developed with Pipe, Valve, Branch, Pump and Single Volume. Specially, to simulate the feedwater, auxiliary feedwater and steam dump systems, which flow rate was determined by system feedback, the Time Dependent Junction was used. With the same rules of primary loops, the components number in secondary loops was numbered in three digits. The first digit stands for the loops number and the other two digits stand for the component types. For instance, the component 520 were heat exchanger in first loop because the first digit 5 represents the first loop and the latter two digits 20 represents the heat exchanger. Component 520 was connected to the right boundary of heat structure 2500, which allows the heat transfer from component 250 in primary side to component 520 in secondary side. Due the heat from primary side, the water in component 520 will evaporate and go through the next component 522. Component 522 was a separator which can increase the quality up to 99.7%. This dried steam will then leave the separator and go through the main steam line isolation valve (component 543), turbine control valve (component 774), turbine stop valve (component 775) and drive turbine, as shown in Figure 6. | |||
As mentioned in section 2, the steam dump system was composed by 10 steam dump valves, six turbine bypass valves and several controlling equipment. To save the computational time, this RELAP5 model merged 10 steam dump valves into four groups. Each group was developed by a Time Dependent Junction component which the total steam flow rate was consistent to the operating conditions. Likewise, six turbine bypass valves were developed by two Time Dependent Junction components. | |||
5 | |||
Figure 3 RCP Properties of Maanshan NPP in SNAP Interface 6 | |||
and | Figure 4 Heat Exchanger Component 250 of Maanshan NPP in SNAP Interface Figure 5 Heat Structure 2500 Properties in SNAP Interface To simplify the feedwater control system, the feedwater pumps and valves were developed by Time Dependent Volume and Time Dependent Junction respectively. For the Time Dependent Volume components, the fluid boundary conditions were referred to the thermal hydraulic properties of feedwater during operation. Therefore, the control system need only concern the effect of NRWL, steam flow rate and feedwater flow rate to determine the feedwater flow rate. | ||
Once the flow rate was determined, Time Dependent Junctions which related to the control 7 | |||
blocks in the feedwater control system will inject the adequate feedwater into recirculation loops. | |||
Details of the feedwater control system will be discussed in the following section. | |||
Figure 6 Components in Secondary Side of Maanshan NPP in SNAP Interface 2.2 Control Systems In operation, the purpose of water level/feedwater control system is to ensure that water in the steam generator can cover the heat exchanger. For Maanshan NPP, the feedwater flow rate was determined by three units including NRWL in steam generator, steam flow rate and feedwater flow rate. As the water level deviate the setting values, the control system will adjust the injection of the feedwater flow rate to maintain the water level of the steam generator. | |||
Further, two water level measuring systems including the NRWL and WRWL calculate water level with pressure difference. Different from the TRACE model of Maanshan NPP which our group had developed before, there is no water level sensor signal component in the RELAP5 code. As a result, the measurements of water level were developed and composed with density, pressure and volume signal components which was shown in Figure 7. | |||
8 | |||
Figure 7 Feedwater Control System in SNAP Interface In addition to the feedwater control, the steam dump system was also an important response mechanism. As mentioned above, the steam dump system of Maanshan NPP can be divided into two types including the pressure control mode and the Tave mode. The pressure mode was initiated as core power was in range from 0% to 10%, which will not be discussed and applied in this research. Hence, the setting of the steam dump system was only referred to the response of Tave mode. As shown in Figure 8, there are three control blocks with sum function calculated the average core temperature values of loop 1 to loop 3 respectively. Then, the control block 308 with max function will compare the maximum of average core temperature (Tave) in loop 1 to loop 3 with No Load Temperature (Tno load, 564K in Maanshan NPP). | |||
Referring this comparison, the control blocks 318 can convert the difference of Tave and Tno load into steam dump flow rate with Table 15. As the temperature difference exceeded 0% | |||
(0), the first group of dump valves was opened. As the temperature difference exceeded 16% | |||
(15.8), the first groups of dump valves were fully opened and the second group of dump valves started to open and so on. | |||
The pressure and water level control system of pressurizer includes the heater and the spray valve. There two types of heater including control heater and backup heater. The control heater and spray valves were applied for adjusting the pressure inside the pressurizer. From Figure 9, the pressure of pressurizer will be compared to rated pressure in control block 120. With the comparison of these two pressure values, the difference can be transferred into the open of the spray valve (control blocks 123) and the power of heater (control block 121 and heat structure 1212 and 1222). However, the control heater is also related to the lower water level of 9 | |||
pressurizer (control block 121). As the water level has been lower than 14%, the power of control heater will be zero (control block 124), which means the heater trip. In addition, if the trip setting of control block 121 was assigned to other trip signals, then the control heater can be tripped manually. | |||
Figure 8 Steam Dump Control System in SNAP Interface The backup heater is related to the charging control system of pressurizer. As shown in Figure 10, the maximum core temperature will be transferred into program water level through control block 130. Then, the water level will be subtracted from the actual water level. If the difference of these two water levels is larger than 5%, the backup heater will be initiated (control block 132). Further, the water level will be transferred into charging flow rate (control block 136) to adjust the water level inside the pressurizer. However, as the safety injection signal is initiated, the charging flow rate will be forced to zero. | |||
10 | |||
Figure 9 Heater of the Pressurizer Control System in SNAP Interface Figure 10 Pressurizer Injection Control System in SNAP Interface 2.3 Reactor Kinetics In this RELAP5 model of Maanshan NPP, there are two sets of heat structures, which component numbers are 1201 and 1601, developed to simulate the hot fuel channel and average fuel channel. These heat structures were divided into 16 nodes (shown in Figure 11) in axial and 7 nodes in radial (shown in Figure 12). For the axial nodes, they were connected to the Brach components of the reactor core respectively. For radial nodes, the first 4 nodes stand for fuel pellets; the fifth node is filled helium inside the fuel rod and the sixth and seventh nodes 11 | |||
are fuel cladding. The materials of each node can be defined manually. In this model, thermal properties (thermal conductivity and thermal capacity) of material 1 for the first 4 nodes were referred to that of Uranium dioxide. The material 2 for node 5 was referred to the helium thermal properties and the material 3 for node 6 and 7 was referred to that of zircalloy. | |||
Figure 11 Properties in Axial Direction of Heat Structure 1201 in SNAP Heat source of the heat structure can be set with the total reactor power or power table. In this model, at the beginning of the model assessment the heat source was set with power table which was referred to the startup test data results of Maanshan NPP to ensure the applicability of the thermal hydraulic components. After that, heat source of the heat structure would be set with total reactor power to ensure the point kinetic feedback calculations. For those heat structure components which were developed as the fuel bundles, the left boundaries were set as symmetry and the right boundaries were connected to the Branch components. For these connections of Brach components and heat structures, the power ratio should be defined (as shown in Figure 13) respectively to calculate the correct heat transfer. The power ratio setting of this RELAP5 model was referred to the TRACE models which were fully developed and assessed before. | |||
12 | |||
Figure 12 Properties in Radial Direction of Heat Structure 1201 in SNAP Figure 13 Power Ratio for Branch 100 to 116 of Heat Structure 1201 13 | |||
this | For the point kinetic model, in addition to defining the power ratio of each node of heat structures, the ratio and position of reactivity feedback should also be defined. The reactivity feedback is dominated by Doppler Effect and Moderator density effect. The previous one is related to the temperature of fuel rods; hence, the check list of fuel temperature and reactivity should be added into the Power components. With this table (as shown in Figure 14), the RELAP5 code can calculate the corresponding reactivity feedback due to fuel temperature. | ||
Further, the fuel temperature feedback ratio should also be defined manually in the Heat Weighting settlement of Power component (shown in Figure 15). Similarity, to calculate the Moderator feedback, the checklist of coolant density and reactivity should be defined (shown in Figure 16). Then, with the volume weighting list (shown in Figure 17), the Branch components which were developed as the reactor core would be connected to the point kinetic calculation. | |||
With these settings, the RELAP5 code could calculate the power variation due to temperature and density changes inside the reactor core during transient events. | |||
Figure 14 Doppler Effect Reactivity Feedback Table As mentioned above, the startup assessment transient events were calculated with power table first to ensure the applicability of thermal hydraulic components. Then, the point kinetic model would be applied to do the whole assessment of Maanshan RELAP5 model. As performed with power table, the RELAP5 model needs no control system to simulate reactor scram. However, when performed with point kinetic model, the reactor scram control is required. For instance, Table 100 is the scram reactivity feedback table which will start to dominate the power variation once the trip logic/variable gate is initiated as shown in Figure 18. From this figure, it is obvious that the table could cause a large negative reactivity feedback in few seconds to simulate the control rods insertion. | |||
14 | |||
Figure 15 Doppler Effect Heat Structure Weighting Factor 15 | |||
Figure 16 Density Effect Reactivity Feedback Table Figure 17 Moderator Density Effect Volume Weighting Factor 16 | |||
Figure 18 Reactor Scram Reactivity Feedback Properties 17 | |||
3 METHODOLOGY This methodology has four steps which are (1) develop the basic analysis model using RELAP5. | |||
(2) ranking of plant status parameters. (3) ranging of plant status uncertainty parameters (4) generate the input files by random sampling. | |||
(1) Develop the basic analysis model using RELAP5: | |||
RELAP5/MOD3.3 code with SNAP interface is used to do the calculations in this research. The RELAP5 Maanshan NPP analysis model had been developed few years ago. Several verifications and validations had been done, including main steam line isolation valve closure (MSIVC), feedwater pumps trip (FWPT), turbine trip (PAT50), | |||
partial loss of flow (PLOF), complete loss of flow (CLOF) [7]. | |||
(2) Ranking of plant status parameters: | |||
Before the uncertainty analysis, it is necessary to do the preliminary analysis or study to determine which uncertainty parameters are important. In this research, a LBLOCA base simulation was done to identify the uncertainty parameters. Initial conditions were according to the real operation conditions. Initial power was set to be 2822MWt and assumed all the systems were operated in the normal status. The uncertainty parameters were evaluated based on the results of the base case, including pressurizer pressure, accumulator (ACC) pressure, ACC volume, ACC water temperature, safety injection (SI) water temperature, feedwater temperature. Detail of the LBLOCA base analysis is indicated in next chapter. | |||
(3) Ranging of plant status uncertainty parameters: | |||
To define the uncertainty parameters of plant status, the uncertainty range and the distribution function are both needed. The uncertainty range in this research are according to Maanshan FSAR and Maanshan Training Manual. The distribution function is set to be uniform distribution in this research. Detail of the uncertainty parameters are listed in chapter. | |||
(4) Generate the input files by random sampling: | |||
DAKOTA code is used to do the sampling in this research. RELAP5 and DAKOTA code are coupled on the SNAP interface. The input uncertainty parameters were sampled and generated the RELAP5 input files. Figure 19 is an example of parameters input, uncertainty range and distribution function setting in DAKOTA. | |||
Wilks formula was adopted to determine the minimum number of runs. In nuclear safety analysis, only upper bond of PCT is the most concern. For the one-sided statistical tolerance limits, the minimum number of runs is given by Wilks formula: | |||
1 is the percentage of tolerance limit, is the confidence level, n is the minimum number of runs. For example, the analysis results in this research achieve / = 95/95 criterion, it means there is 95% confidence level that the results are below the 95% of tolerance limit. To achieve the criterion, the minimum number of runs n = 59. | |||
18 | |||
Figure 19 Uncertainty Parameters Setting in DAKOTA 19 | |||
4 LBLOCA BASE CASE 4.1 Initial Condition and Basic Assumption The initial conditions and the ECCS properties are listed in Table 1. The initial power is 2822 MWt. The LBLOCA base case was performed in this study. It was assumed that the break occurred at the primary side, at first loop cold-leg near the vessel. The break type was assumed to be double-ended and the break model is shown in Figure 20. Before the pipe rupture, 400 sec steady-state calculation had been done to ensure all the system operated correctly. In the LBLOCA transient, High Head Safety Injection (HHSI), ACC injection, and Low Head Safety Injection (LHSI) are included. It was assumed that all the safety injection systems could operate normally. The safety injection signal setpoint is 11.8 MPa and delay time is 27 sec. The ACC injection set point is 4.24 MPa. The minimum operating pressure of Low Head Safety Injection (LHSI) is 1.15 MPa. | |||
Table 1 Initial Condition and ECCS properties Parameters Values Initial power (MWt) 2822 Core flow rate (kg/s) 14176 Tavg* (K) 582.97 RCS* pressure (MPa) 15.862 Scram setpoint (MPa) 12.8 SI* signal setpoint (MPa) 11.8 SI signal delay (s) 27 ACC liquid volume (m3/tank) 27.9 ACC injection setpoint (MPa) 4.24 ACC liquid temperature (K) 311 LHSI operating pressure(MPa) 1.15 Break size (m2) 0.38 21 | |||
Figure 20 RELAP5/SNAP LBLOCA Break Model 4.2 Analysis Results The LBLOCA transient can be divided into six stages, which are critical heat flux phase, upward core flow phase, downward core flow phase, refill phase, reflood phase, and long term cooling phase. The exact postulated event sequence of Maanshan NPP is listed in Table 2. | |||
Table 2 Postulated LBLOCA Events Sequence of Maanshan NPP Action Time (sec) | |||
Pipe rupture 0 Scram 1.8 RCP trip 1.8 SI signal 2.8 ACC injection 5.5 HHSI 29.8 LHSI 29.8 ACC injection ended 69.5 End of event 700 4.2.1 Critical Heat Flux Phase Immediately following the cold-leg break, the fluid discharged out of the pipe with very high flow rates, which can be observed in Figure 21. Due to the extremely high break flow rate, the system pressure drop rapidly. This depressurized phenomenon resulted in water flashing in the 22 | |||
reactor vessel. The void fraction of vessel is shown in Figure 22, which indicates that the void fraction went up to one in very few seconds. Because of the flashing phenomenon in reactor core, the fuel cladding became fully uncovered. As shown in Figure 23, the cladding temperature increasing dramatically to the first peak in the very early period. | |||
Figure 21 LBLOCA Break Flow Rate 4.2.2 Upward Core Flow Phase The coolant in loop system still had great inertia at the very beginning (0 ~ 5 sec), and flashing phenomenon in the reactor vessel also generated a great pressure drop. Therefore, there was some two-phase mixture could be pulled into the core, which made the cladding temperature decrease. Because of the bottom-up mixture cooling effect, the temperature drop was more distinct at the lower part. This phenomenon can be seen in Figure 23, there was a temperature drop just after the first peak. | |||
23 | |||
Figure 22 Void Fraction in Vessel 4.2.3 Downward Core Flow Phase As pump head continues to degrade, and the coolant slow down. The pressure drop at break started to fully dominate and pull flow down through the core, up through the downcomer to the break. Figure 24 shows the vapor velocity in reactor core, which indicate a velocity increasing downward at about 6 ~ 10 sec. High speed vapor flow also provided good core cooling. This phenomenon can be seen in Figure 23, which shows a second temperature drop at about 10 sec. | |||
4.2.4 Refill Phase Following the blowdown phase, the water level was below the BAF (see Figure 25). Therefore, the cladding temperature increased (see Figure 23). This period is featured of rapid cladding temperature increasing because of lacking water and steam cooling. | |||
24 | |||
Figure 23 Cladding Temperature at the Lower Part Figure 24 Core Vapor Velocity 25 | |||
4.2.5 Reflood Phase Due to the ACC injection, the water level continued to increase. Figure 25 shows the core collapsed water level. At about 19 sec, the coolant reached the BAF and lower core region began to quench. During this period, core cooling might increase because of upward vapor flow and liquid entrainment. Therefore, the cladding temperature could decrease before the coolant recovered the fuel cladding. | |||
In general, if the income core flow velocity exceeded certain value, it was assumed that the flow regime near the quench front was Inverted-Slug flow. The references [8~10] indicate that the minimum core coolant inlet velocity should be 0.15m/s to form an Inverted-Slug flow regime. | |||
These studies also indicate that the quenching temperature depends strongly on the core inlet coolant velocity. | |||
The flow regime during reflooding phase was shown in Figure 26. Under the quench front, fuel cladding was covered by liquid and cladding temperature dropped to saturation temperature. At inverted-annular and droplet region, the two phase mixture still provided good cooling effect that made the cladding temperature go down. Therefore, this period is the most important in LBLOCA analysis. Figure 27 shows the quench and rewet points for the nodes. | |||
Figure 25 Core Collapsed Water Level 26 | |||
4.2.6 Long-Term Core Cooling Whole fuel cladding was recovered by coolant at about 500 sec, which can be seen in Figure | |||
: 27. With the pumped safety injection operating continuously, the water will never below the TAF again. | |||
Figure 26 Core Flow Regime during Reflooding phase 27 | |||
Figure 27 Cladding Temperature Axial Profile 28 | |||
5 LBLOCA UNCERTAINTY ANALYSIS 5.1 Uncertainty Parameters According to the previous Chapter, the evaluation of quench front progressing during the reflooding phase strongly affects the cladding temperature. The quench front velocity is a relation between TQ and h [11, 12]. TQ is the quenching temperature, and h is the heat transfer coefficient of the coolant under the quench front. TQ and h strongly depend on system pressure and subcooling of the coolant. In LBLOCA case, the performance of ECCS may affect the PCT, system pressure, and subcooling of the coolant. Therefore, the uncertainty analysis by using RELAP5, DAKOTA, and SNAP codes was performed in this study. The uncertainty parameters which are mainly the properties of ECCS are presented in Table 3. These parameters include pressurizer pressure, accumulator (ACC) pressure, ACC volume, ACC water temperature, safety injection (SI) water temperature, and feedwater temperature. The uncertainty ranges are based on FSAR and Training Manual of Maanshan NPP. The distribution functions are all set to be uniform distribution. According to the Wilks formula, 59 of runs should be conducted to achieve the 95/95 criterion. These uncertainty parameters would be randomly sampled and generated 59 RELAP5 input files. | |||
Table 3 Uncertainty Parameters Parameter Deviation PDFs Reference 15.86 MPa 1 PZR pressure FSAR | |||
(+/-0.21MPa) 2 ACC pressure 4.14 - 4.55 MPa Training manual 27.9 m3 (+/-0.425 3 ACC volume Training manual m3) Uniform distribution 4 ACC temperature 310.93 - 338.71 K Training manual 5 SI temperature 300 K (+19.4 K) FSAR Feedwater 6 500 K (+/-1.22 K) Training manual temperature 5.2 Results The core collapsed water level can be seen in Figure 28, which shows the time of end-of-blowdown is estimated at about 19 sec. Reflooding begin at about 37 sec, the time which the coolant reaches the BAF. PCT of 59 cases was shown in Figure 29. During blowdown and refilled phases, the temperature results of 59 cases are very similar. The time of the first temperature peak of 59 cases all occur at about 7 sec and the value of all cases are also almost identical (1293 K). In the beginning of the pipe break, a severe depressurization causes liquid in reactor core suddenly transfer to vapor phase. The heat transfer coefficient of vapor is much lower than liquid makes the cladding temperature reach to the first peak at about 7 sec. It is obvious that the first peak is caused by the heat transfer coefficient difference between liquid and vapor. Therefore, these six uncertainty parameters, which are related to ECCS, have no effect on the first peak. After the coolant reaches BAF, both PCT and water level curves start to have distinct deviations. During this period, the coolant started to attach the lower part of fuel 29 | |||
cladding. Some liquid droplets would be entrained by vapor from lower part to upper part that makes the cladding temperature do not increase dramatically like the first peak. Therefore, the deviations of ECCS properties made the cooling effect different and makes different PCT. The red curve in Figure 29 shows the highest PCT of all cases, which is about 1293 K. The PCT of LBLOCA in Maanshan FSAR is 1383.7 K. The difference of analysis results and FSAR data may be caused by the different calculation procedures, phenomenological modeling, and nodalization, etc. In addition, the PCT criteria of 10 CFR 50.46 is 1477.6 K (2200). Comparing the criteria and the results of uncertainty analysis, the results of uncertainty analysis are below the criteria. | |||
Figure 28 Core Collapsed Water Level of 59 Cases 30 | |||
Figure 29 PCT of 59 Cases 31 | |||
6 CONCLUSIONS This study presents an uncertainty analysis methodology of a postulated LBLOCA of Maanshan NPP using RELAP5/MOD3.3, DAKOTA, and SNAP codes. The results in base case analysis shows that the model of Maanshan NPP is able to manage a LBLOCA simulation. The phenomenon of the base case is consistent with the LOFT experiment and the analysis results of Westinghouse [13, 14]. The evaluation of quench front velocity during reflooding phase strongly affects the PCT. The performance of ECCS may affect reflooding phase. Therefore, an uncertainty analysis was performed in this study. | |||
. | For the uncertainty analysis, most of the uncertainty parameters are related to ECCS property. | ||
The ranges of uncertainty analysis are based on FSAR and Training Manual of Maanshan NPP. | |||
The maximum PCT of the analysis is 1293 K. This result includes the uncertainty effect. The PCT of Maanshan FSAR is 1383.7 K. Therefore, the prediction of the uncertainty analysis is below the FSAR data. The difference of analysis results and FSAR data may be caused by the different calculation procedures, phenomenological modeling, and nodalization, etc. In addition, the PCT criteria of 10 CFR 50.46 is 1477.6 K (2200). Hence, the result of the uncertainty analysis is also below the criteria of 10 CFR 50.46. This indicates that Maanshan NPP in LBLOCA transient is at a safety situation when ECCS is performed. | |||
33 | |||
.. | |||
. | |||
... | |||
.. | |||
7 REFERENCES | |||
: 1. Taiwan Power Company, Final Safety Analysis Report for Maanshan Nuclear Power Station Units 1&2 (FSAR), Taiwan Power Company, Republic of China (Taiwan), 1983. | |||
: 2. Jong-Rong Wang, Che-Hao Chen, Hao-Tzu Lin, Chunkuan Shih, Assessment of LONF ATWS for Maanshan PWR Using TRACE Code, NUREG/IA-0436, 2014. | |||
: 3. Hao-Tzu Lin, Jong-Rong Wang, Kai-Chun Huang, Chunkuan Shih, Show-Chyuan Chiang, Chia-Chuan Liu, Station Blackout Mitigation Strategies Analysis for Maanshan PWR Plant using TRACE, Annals of Nuclear Energy, 89, 2016. | |||
: 4. Information Systems Laboratories, Inc., RELAP5/MOD3.3 Code Manual Volume I: Code Structure, System Models, and Solution Methods, Information Systems Laboratories, Inc., Rockville Maryland, 2010. | |||
: 5. Adams, B. M., et al., DAKOTA, a Multilevel Parallel Object-oriented Framework for Design Optimization, Parameter Estimation, Uncertainty Quantification, and Sensitivity Analysis, Sandia National Laboratories, 20-21, 66-67, 99, 2011. | |||
: 6. Applied Programming Technology, Symbolic Nuclear Analysis Package (SNAP) User's Manual, Applied Programming Technology Inc., Bloomsburg, 2007. | |||
: 7. Chunkuan Shih, Jong-Rong Wang, Shao-Wen Chen, Hao-Chun Chang, Show-Chyuan Chiang, and Tzu-Yao Yu, RELAP5/MOD3.3 Model Assessment of Maanshan Nuclear Power Plant with SNAP Interface, NUREG/IA-0472, 2017. | |||
: 8. A.D. Siegel, Rewetting Temperature during Top and Bottom Reflood, M.S. Thesis, University of Illinois, 1979. | |||
: 9. A.D. Siegel and J.J. Carbajo, Trans. ANS 35, pp. 327-328, 19. | |||
: 10. Kevin D. Kimball, Quench Front Propagation during Bottom Reflooding of a Heated Annular Channel, Nuclear Engineering and Design 76, pp. 79-88, 1983. | |||
: 11. Duffey, R.B., Porthouse, D.T.C., The Physics of Rewetting in Water Reactor Emergency Core Cooling. Nuclear Engineering and Design 25, pp. 379~394, 1973. | |||
: 12. B.D.G. Piggott and D.T.C. Porthouse, A Correlation of Rewetting Data. Nuclear Engineering and Design 32, pp. 171~181, 1975. | |||
: 13. S. M. Modro, S. M. Modro, V. T. Berta, A. B. Wahba, Review of LOFT Large Break Experiments, NUREG/IA-0028, 1989. | |||
: 14. Westinghouse Company, Best-Estimate Analysis of the Large-Break Loss-of-Coolant Accident for Maanshan Units 1 and 2 Nuclear Power Plant Using the ASTRUM Methodology, WCAP-17054-P, 2009. | |||
35 | |||
NUREG/IA-0509 LBLOCA Uncertainty Analysis of Maanshan Nuclear Power Plant with RELAP5/SNAP and DAKOTA April 2019 Chunkuan Shih, Jong-Rong Wang, Jiang-Ting Chen, Yuh-Ming Ferng, Technical Shao-Wen Chen, Ting-Yi Wang*, and Tzu-Yao Yu* | |||
Institute of Nuclear Engineering and Science | |||
* Department of Nuclear Safety, Taiwan Power Co. | |||
National Tsing Hua University 242, Section 3, Roosevelt Rd., Zhongzheng Dist. | |||
Nuclear and new Energy Education and Research Foundation Taipei, Taiwan 101 Section 2, Kuang Fu Rd., HsinChu, Taiwan Division of Systems Analysis Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington, D.C. 20555-0001 K. Tien, NRC Project Manager The objective of this study is to assess the applicability of the RELAP5/MOD3.3 model of Maanshan NPP on LBLOCA transient. Maanshan NPP was the first three-loop PWR in Taiwan constructed by Westinghouse. In 2015, the RELAP5/MOD3.3 model of Maanshan NPP was developed with SNAP interface. An uncertainty analysis methodology of the postulated LBLOCA for Maanshan NPP using RELAP5/MOD3.3, DAKOTA, and SNAP codes are established in this study. The uncertainty parameters are related to ECCS property. The ranges of uncertainty analysis are based on FSAR and Training Manual of Maanshan. The maximum PCT of the uncertainty analysis is 1293 K. The PCT of FSAR is 1383.7 K. Hence, the result of the uncertainty analysis is below the FSAR data. The difference of uncertainty analysis results and FSAR data may be caused by the different calculation procedures, phenomenological modeling, and nodalization, etc. | |||
Additionally, the PCT criteria of 10 CFR 50.46 is 1477.6 K (2200). The prediction of the uncertainty analysis is below the criteria of 10 CFR 50.46. This implies that Maanshan NPP in LBLOCA transient is at a safety situation when ECCS is performed. | |||
Probability Distribution Function (PDF) | |||
Feedwater Pump(s) Trip (FWPT) | |||
Complete Loss of Flow (CLOF) | |||
High Head Safety Injection (HHSI) | |||
Top of Active Fuel (TAF) | |||
Maanshan Nuclear Power Plant (PWR) | |||
The maximum PCT of the analysis is 1293 K | |||
NUREG/IA-0509 LBLOCA Uncertainty Analysis of Maanshan Nuclear Power Plant with RELAP5/SNAP and DAKOTA April 2019}} | |||
Latest revision as of 21:14, 19 October 2019
ML19112A268 | |
Person / Time | |
---|---|
Issue date: | 04/30/2019 |
From: | Chen J, Sunny Chen, Ferng Y, Shih C, Kirk Tien, Jamie Wang, Wang T, Yu T Office of Nuclear Regulatory Research, National Tsing Hua Univ, Institute of Nuclear Science, Taiwan |
To: | |
Meyd, Donald | |
References | |
NUREG/IA-0509 | |
Download: ML19112A268 (57) | |
Text
NUREG/IA-0509 International Agreement Report LBLOCA Uncertainty Analysis of Maanshan Nuclear Power Plant with RELAP5/SNAP and DAKOTA Prepared by:
Chunkuan Shih, Jong-Rong Wang, Jian-Ting Chen, Yuh-Ming Ferng, Shao-Wen Chen, Ting-Yi Wang *, and Tzu-Yao Yu*
Institute of Nuclear Engineering and Science, National Tsing Hua University Nuclear and New Energy Education and Research Foundation 101 Section 2, Kuang Fu Rd., HsinChu, Taiwan
- Department of Nuclear Safety, Taiwan Power Company 242, Section 3, Roosevelt Rd., Zhongzheng District, Taipei, Taiwan K. Tien, NRC Project Manager Division of Systems Analysis Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington, DC 20555-0001 Manuscript Completed: June 2018 Date Published: April 2019 Prepared as part of The Agreement on Research Participation and Technical Exchange Under the Thermal-Hydraulic Code Applications and Maintenance Program (CAMP)
Published by U.S. Nuclear Regulatory Commission
AVAILABILITY OF REFERENCE MATERIALS IN NRC PUBLICATIONS NRC Reference Material Non-NRC Reference Material As of November 1999, you may electronically access Documents available from public and special technical NUREG-series publications and other NRC records at libraries include all open literature items, such as books, NRCs Library at www.nrc.gov/reading-rm.html. Publicly journal articles, transactions, Federal Register notices, released records include, to name a few, NUREG-series Federal and State legislation, and congressional reports.
publications; Federal Register notices; applicant, Such documents as theses, dissertations, foreign reports licensee, and vendor documents and correspondence; and translations, and non-NRC conference proceedings NRC correspondence and internal memoranda; bulletins may be purchased from their sponsoring organization.
and information notices; inspection and investigative reports; licensee event reports; and Commission papers Copies of industry codes and standards used in a and their attachments. substantive manner in the NRC regulatory process are maintained at NRC publications in the NUREG series, NRC regulations, The NRC Technical Library and Title 10, Energy, in the Code of Federal Regulations Two White Flint North may also be purchased from one of these two sources. 11545 Rockville Pike Rockville, MD 20852-2738
- 1. The Superintendent of Documents U.S. Government Publishing Office These standards are available in the library for reference Mail Stop IDCC use by the public. Codes and standards are usually Washington, DC 20402-0001 copyrighted and may be purchased from the originating Internet: bookstore.gpo.gov organization or, if they are American National Standards, Telephone: (202) 512-1800 from Fax: (202) 512-2104 American National Standards Institute 11 West 42nd Street
- 2. The National Technical Information Service New York, NY 10036-8002 5301 Shawnee Road www.ansi.org Alexandria, VA 22312-0002 (212) 642-4900 www.ntis.gov 1-800-553-6847 or, locally, (703) 605-6000 Legally binding regulatory requirements are stated only in A single copy of each NRC draft report for comment is laws; NRC regulations; licenses, including technical available free, to the extent of supply, upon written specifications; or orders, not in NUREG-series publications.
request as follows: The views expressed in contractor prepared publications in this series are not necessarily those of the NRC.
Address: U.S. Nuclear Regulatory Commission The NUREG series comprises (1) technical and adminis-Office of Administration trative reports and books prepared by the staff (NUREG-XXXX) or agency contractors (NUREG/CR-XXXX), (2)
Multimedia, Graphics, and Storage &
proceedings of conferences (NUREG/CP-XXXX), (3) reports Distribution Branch resulting from international agreements (NUREG/IA-XXXX),
Washington, DC 20555-0001 (4) brochures (NUREG/BR-XXXX), and (5) compilations of E-mail: distribution.resource@nrc.gov legal decisions and orders of the Commission and Atomic Facsimile: (301) 415-2289 and Safety Licensing Boards and of Directors decisions under Section 2.206 of NRCs regulations (NUREG-0750).
Some publications in the NUREG series that are posted at NRCs Web site address www.nrc.gov/reading-rm/ DISCLAIMER: This report was prepared under an doc-collections/nuregs are updated periodically and may international cooperative agreement for the exchange of technical information. Neither the U.S. Government nor any differ from the last printed version. Although references to agency thereof, nor any employee, makes any warranty, material found on a Web site bear the date the material expressed or implied, or assumes any legal liability or was accessed, the material available on the date cited responsibility for any third partys use, or the results of such may subsequently be removed from the site. use, of any information, apparatus, product or process disclosed in this publication, or represents that its use by such third party would not infringe privately owned rights.
NUREG/IA-0509 International Agreement Report LBLOCA Uncertainty Analysis of Maanshan Nuclear Power Plant with RELAP5/SNAP and DAKOTA Prepared by:
Chunkuan Shih, Jong-Rong Wang, Jian-Ting Chen, Yuh-Ming Ferng, Shao-Wen Chen, Ting-Yi Wang *, and Tzu-Yao Yu*
Institute of Nuclear Engineering and Science, National Tsing Hua University Nuclear and New Energy Education and Research Foundation 101 Section 2, Kuang Fu Rd., HsinChu, Taiwan
- Department of Nuclear Safety, Taiwan Power Company 242, Section 3, Roosevelt Rd., Zhongzheng District, Taipei, Taiwan K. Tien, NRC Project Manager Division of Systems Analysis Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington, DC 20555-0001 Manuscript Completed: June 2018 Date Published: April 2019 Prepared as part of The Agreement on Research Participation and Technical Exchange Under the Thermal-Hydraulic Code Applications and Maintenance Program (CAMP)
Published by U.S. Nuclear Regulatory Commission
ABSTRACT The objective of this study is to assess the applicability of the RELAP5/MOD3.3 model of Maanshan NPP on LBLOCA transient. Maanshan NPP was the first three-loop PWR in Taiwan constructed by Westinghouse. The total power of the two units of the power plant is 2785 MWt, which consist of 2775 MWt for reactor power and 10 MWt for cooling pumps. For the last few years, the TRACE model of Maanshan NPP was developed and several kinds of transient event including the LOCA, AOO and the URG transient were performed and verified. Recently, the RELAP5/MOD3.3 code is another important development priority for our group. In 2015, the RELAP5/MOD3.3 model of Maanshan NPP was developed with SNAP interface.
An uncertainty analysis methodology of the postulated LBLOCA for Maanshan NPP using RELAP5/MOD3.3, DAKOTA, and SNAP codes are established in this study. The uncertainty parameters are related to ECCS property. The ranges of uncertainty analysis are based on FSAR and Training Manual of Maanshan. The maximum PCT of the uncertainty analysis is 1293 K. The PCT of FSAR is 1383.7 K. Hence, the result of the uncertainty analysis is below the FSAR data. The difference of uncertainty analysis results and FSAR data may be caused by the different calculation procedures, phenomenological modeling, and nodalization, etc.
Additionally, the PCT criteria of 10 CFR 50.46 is 1477.6 K (2200°F). The prediction of the uncertainty analysis is below the criteria of 10 CFR 50.46. This implies that Maanshan NPP in LBLOCA transient is at a safety situation when ECCS is performed.
iii
FOREWORD The U.S. NRC has developed a thermal hydraulic analysis code, RELAP5, has been designed to perform best-estimate analysis of loss-of-coolant accidents, operational transients, and other accident scenarios in reactor systems. Models used include multidimensional two-phase flow, non-equilibrium thermo-dynamics, generalized heat transfer, reflood, level tracking, and reactor kinetics. Traditionally, the RELAP5 code analysis model was developed by ASCII file, which was not intelligible for the beginners of computer analysis. Fortunately, and graphic input interface, SNAP is developed by Applied Programming Technology Inc. and conducted by the U.S. NRC, the model development process becomes more convenient.
To obtain the authorization of these codes, Taiwan and the United States have signed an agreement on CAMP (Code Applications and Maintenance Program) which includes the development and maintenance of RELAP5 code. NTHU (National Tsing Hua University) is the organization in Taiwan responsible for the application RELAP5 and SNAP in thermal hydraulic safety analysis. Therefore, the RELAP5/MOD3.3 model of Maanshan nuclear power plant has been developed. To expand the applicability of the RELAP5/MOD3.3 model, an uncertainty analysis methodology of the postulated LBLOCA is established in this study.
v
TABLE OF CONTENTS ABSTRACT ............................................................................................................................... iii FOREWORD ............................................................................................................................... v TABLE OF CONTENTS ........................................................................................................... vii LIST OF FIGURES .................................................................................................................... ix EXECUTIVE
SUMMARY
........................................................................................................... xi ABBREVIATIONS AND ACRONYMS ..................................................................................... xiii 1 INTRODUCTION ............................................................................................................. 1 2 MODEL ESTABLISHMENT ............................................................................................ 3 2.1 Hydraulic Components ........................................................................................ 3 2.2 Control Systems ................................................................................................. 8 2.3 Reactor Kinetics................................................................................................. 11 3 METHODOLOGY .......................................................................................................... 19 4 LBLOCA BASE CASE...21 4.1 Initial Condition and Basic Assumption.....................21 4.2 Analysis Results.....22 4.2.1 Critical Heat Flux Phase...22 4.2.2 Upward Core Flow Phase....23 4.2.3 Downward Core Flow Phase.. 24 4.2.4 Refill Phase 24 4.2.5 Reflood Phase26 4.2.6 Long-Term Core Cooling..27 5 LBLOCA UNCERTAINTY ANALYSIS....29 5.1 Uncertainty Parameters....29 5.2 Results........29 6 CONCLUSIONS ............................................................................................................ 33 7 REFERENCES .............................................................................................................. 35 vii
LIST OF FIGURES Figure 1 Reactor Core Components of Maanshan NPP in SNAP Interface ........................ 4 Figure 2 Components of First Loop of Maanshan NPP in SNAP Interface ......................... 5 Figure 3 RCP Properties of Maanshan NPP in SNAP Interface ......................................... 6 Figure 4 Heat Exchanger Component 250 of Maanshan NPP in SNAP Interface .............. 7 Figure 5 Heat Structure 2500 Properties in SNAP Interface ............................................... 7 Figure 6 Components in Secondary Side of Maanshan NPP in SNAP Interface ................ 8 Figure 7 Feedwater Control System in SNAP Interface ...................................................... 9 Figure 8 Steam Dump Control System in SNAP Interface ................................................ 10 Figure 9 Heater of the Pressurizer Control System in SNAP Interface ............................. 11 Figure 10 Pressurizer Injection Control System in SNAP Interface .................................... 11 Figure 11 Properties in Axial Direction of Heat Structure 1201 in SNAP ............................ 12 Figure 12 Properties in Radial Direction of Heat Structure 1201 in SNAP .......................... 13 Figure 13 Power Ratio for Branch 100 to 116 of Heat Structure 1201 ................................ 13 Figure 14 Doppler Effect Reactivity Feedback Table ......................................................... 14 Figure 15 Doppler Effect Heat Structure Weighting Factor ................................................. 15 Figure 16 Density Effect Reactivity Feedback Table .......................................................... 16 Figure 17 Moderator Density Effect Volume Weighting Factor ........................................... 16 Figure 18 Reactor Scram Reactivity Feedback Properties ................................................. 17 Figure 19 Uncertainty Parameters Setting in DAKOTA ...................................................... 20 Figure 20 RELAP5/SNAP LBLOCA Break Model............................................................... 22 Figure 21 LBLOCA Break Flow Rate ................................................................................. 23 Figure 22 Void Fraction in Vessel ...................................................................................... 24 Figure 23 Cladding Temperature at the Lower Part ........................................................... 25 Figure 24 Core Vapor Velocity ........................................................................................... 25 Figure 25 Core Collapsed Water Level .............................................................................. 26 Figure 26 Core Flow Regime during Reflooding Phase...................................................... 27 Figure 27 Cladding Temperature Axial Profile .................................................................... 28 Figure 28 Core Collapsed Water Level of 59 Cases .......................................................... 30 Figure 29 PCT of 59 Cases................................................................................................ 31 ix
EXECUTIVE
SUMMARY
RELAP5/MOD3.3 Patch04 code, which was developed for light water reactor transient analysis at Idaho National Engineering Laboratory for U.S. NRC, is applied in this research. This code is often performed to support rulemaking, licensing audit calculations, evaluation of accident, mitigation strategies, evaluation of operator guidelines, and experiment planning analysis. Same as other thermal hydraulic analysis codes, RELAP5/MOD3.3 is based on nonhomogeneous and non-equilibrium model for the two-phase system. However, calculations in this code will be solved by a fast, partially implicit numerical scheme to permit economical calculation of system transients. It can produce accurate transient analysis results in relatively short time.
SNAP is an interface of NPP analysis codes which developed by U.S. NRC and Applied Programming Technology, Inc. Different from the traditional input deck in ASCII files, the graphical control blocks and thermal hydraulic connections make researches comprehend the whole power plant and control system more easily. Due to these advantages, the RELAP5/MOD3.3 model of Maanshan NPP was developed with SNAP interface.
DAKOTA code can be used in uncertainty analysis. DAKOTA is applied for the sampling of input parameters, the calculation of correlations, and ranking of input parameters. DAKOTA provides Monte Carlo sampling and Latin Hypercube sampling methods combined with various PDFs including normal, lognormal, uniform, logunifrom, hypergeometric, and user-supplied histograms. In addition, the uncertainty analysis of DAKOTA also can run with the SNAP interface.
Maanshan NPP is located on the southern coast of Taiwan. Its nuclear steam supply system is a type of PWR designed and built by Westinghouse for Taiwan Power Company. In this research, a RELAP5/MOD3.3 model of Maanshan NPP for LBLOCA transient was developed.
This research also presents an uncertainty analysis methodology of LBLOCA using RELAP5/MOD3.3, DAKOTA, and SNAP codes.
The results of the RELAP5 analysis for LBLOCA base case indicate that the RELAP5/SNAP model is able to manage a LBLOCA simulation. The phenomenon of the base case is consistent with the LOFT experiment and the analysis results of Westinghouse. According to analysis results, the evaluation of quench front velocity during reflooding phase strongly affects the PCT.
The performance of ECCS may affect reflooding phase. Therefore, an uncertainty analysis was performed in this research.
Most of the uncertainty parameters are related to ECCS property in the uncertainty analysis.
The ranges of uncertainty analysis are based on FSAR and Training Manual of Maanshan NPP.
The maximum PCT of the analysis is 1293 K that includes the uncertainty effect. However, the PCT of Maanshan FSAR is 1383.7 K. The difference of analysis results and FSAR data may be caused by the different calculation procedures, phenomenological modeling, and nodalization, etc. In addition, the PCT criteria of 10 CFR 50.46 is 1477.6 K (2200). Hence, the result of the uncertainty analysis is below the criteria of 10 CFR 50.46. This indicates that Maanshan NPP in LBLOCA transient is at a safety situation when ECCS is performed.
xi
ABBREVIATIONS AND ACRONYMS ACC Accumulator BAF Bottom of Active Fuel CAMP Code Applications and Maintenance Program CLOF Completed Loss of Flow ECCS Emergency Core Cooling System FWPT Feedwater pump(s) Trip FSAR Final Safety Analysis Report HHSI High Head Safety Injection kg kilogram(s)
LBLOCA Large Break Loss Of Coolant Accident LHSI Low Head Safety Injection MPa Megapascal(s)
MSIVC Main Steam Isolation Valves Closure NPP Nuclear Power Plant NRC Nuclear Regulatory Commission NRWL Narrow Range Water Level NSSS Nuclear Steam Supply System NTHU National Tsing Hua University PDF Probability Distribution Function PORVs Power-Operated Relief Valves PWR Pressurized Light Water Reactor PLOF Partial Loss of Flow RCS Reactor Coolant System RCP Reactor Coolant Pump S/G Steam Generator SI Safety Injection SNAP Symbolic Nuclear Analysis Program SRV Safety/Relief Valves TAF Top of Active Fuel TRACE TRAC/RELAP Advanced Computational Engine WRWL Wide Range Water Level xiii
1 INTRODUCTION Maanshan Nuclear Power Plant (NPP) is the third NPP in Taiwan. Also, it is the first Pressurized Water Reactor (PWR) located at the south of Taiwan. There are two units in the Maanshan NPP. The total power of the nuclear steam supply system is 2785 MWt, which consist of 2775 MWt for reactor power and 10 MWt for cooling pumps [1]. For the last few years, our group has developed the models of Taiwan NPPs with TRACE code in SNAP interface [2, 3]. Further, it is necessary to perform the NPP transients with several analysis codes so that the data results could be compared with each other to ensure the consistency. Therefore, the RELAP5/MOD 3.3 Patch04 code was chosen to develop a new Maanshan NPP model. Different from the traditional ASCII input deck, the RELAP5/MOD 3.3 model was developed with SNAP interface.
DAKOTA is an uncertainty analysis code [5] and is applied for the sampling of input parameters, the calculation of correlations, and ranking of input parameters. The uncertainty quantification package of DAKOTA provides Monte Carlo sampling and Latin Hypercube sampling methods combined with various PDFs including normal, lognormal, uniform, logunifrom, hypergeometric, and user-supplied histograms.
SNAP is an interface of NPP analysis codes which developed by U.S. NRC and Applied Programming Technology, Inc. Different from the traditional input deck in ASCII files, the graphical control blocks and thermal hydraulic connections make researches comprehend the whole power plant and control system more easily [6]. Due to these advantages, an uncertainty analysis methodology of LBLOCA using RELAP5/MOD3.3 and DAKOTA with SNAP interface is established in this study. Moreover, due to the SNAP interface, the analysis results could be transferred into animations which were more attractive and more understandable. With the animation, interactions of different components and parameters could be easily observed.
To ensure the applicability of this model, three startup tests cases including feedwater pumps trip (FWPT), turbine trip (PAT50) and main isolation valves closure (MSIVC) had been analyzed in 2015 [7]. With the comparison of RELAP5 results and startup tests data, it shows that the RELAP5/MOD 3.3 model of Maanshan NPP is consistent with the startup tests data. To expand the applicability of this RELAP5 model, the postulated LBLOCA is performed. Two main steps are in this study. First, the establishment of RELAP5/MOD3.3 model for the postulated LBLOCA of Maanshan NPP is performed. An analysis of LBLOCA base case is performed by using this model to confirm the phenomenon of LBLOCA. Second, to confirm the uncertainty of ECCS that may affect the PCT, the uncertainty analysis of RELAP5/MOD 3.3 model combining with DAKOTA is performed in this study.
1
2 MODEL ESTABLISHMENT As the Maanshan NPP operated in normal conditions, coolant water in primary system will carry the heat generated by the fuel rods to the steam generator. Feedwater in the secondary system then obtain the heat, evaporate and drive turbines to generate electricity. According to the energy conservation principle, internal energy of steam, which had driven turbines, will decrease. This lower internal energy steam will then go through the condenser and be transferred into feedwater and re-injected into the steam generator. However, it is difficult to develop the entire recirculation system with the analysis code. The computational time will be impractically long and the mass balance will be hard to reach. Hence, as developing the RELAP5 model, it is practical to define the feedwater pumps and the turbines as the boundary conditions because the main purpose of this model is to obtain the NSSS reactions during the transient. For the NSSS system of Maanshan NPP, the feedwater pumps, auxiliary feedwater pumps, turbines, safety/relief valves, steam dump valves and PORVs were defined as the boundary conditions and developed by the Time Dependent Volume component in the RELAP5 program [7].
2.1 Hydraulic Components As mentioned in section 2, there are three recirculation loops in the Maanshan NPP. In each loop, there is a RCP and S/G. On the hot leg of second loop, a pressurizer which can adjust the pressure of RCS with the spray valves and electronic heater was developed. In this analysis model, there are several Branch components developed to simulate the reactor vessel.
According to the core arrangement, Branch components from number 140 to 156 were connected together as the average fuel channel. Branch components from number 120 to 136 were connected together as the hottest fuel channel. Branch components from number 100 to 116 were connected together as the bypass flow channel, as shown in Figure 1. Also, these channels will be connected to the heat structure components to obtain the heat and do the reactor kinetic analysis.
For those three recirculation loops in primary side, they were developed by Pipe, Valve, Branch, Jump and Single Volume, as shown in Figure 2. For these three-digit components, the first digit stands for the loop number (2 for first loop, 3 for second loop and 4 for third loop). Further, the other digits of these components represents to the component types. For instance, component 280 is the recirculation pumps in first loop and component 380 is the recirculation pumps in second loop. Though the Pump component in RELAP5 code has been developed with the pump parameters from Westinghouse, pump characters of the RCPs in this model was input according to Taiwan Power Company NPP training materials and past research models which were calculated by RELAP5-3D and TRACE codes, as shown in Figure 3.
In addition to RCPs, another important thermal hydraulic component in the primary side is heat exchanger. Pipe 250, which was developed for heat exchanger in first loop, was divided into eight nodes. According to the geometry of heat exchanger, junction between fourth cell and fifth cell was 180 degree as shown in Figure 4. Further, the heat structure component can be view as structural component once the both side of the heat structure were connected to thermal hydraulic components. Hence, the Pipe 250 was connected to the left boundary of heat structure 2500, as shown in Figure 5. Likewise, the Pipe 350 of secondary loop was connected to the left boundary of heat structure 3500 and Pipe 450 of third loop was connected to the left boundary of heat structure 4500.
3
Figure 1 Reactor Core Components of Maanshan NPP in SNAP Interface 4
Figure 2 Components of First Loop of Maanshan NPP in SNAP Interface Similar to primary loops, the secondary loops of Maanshan NPP were developed with Pipe, Valve, Branch, Pump and Single Volume. Specially, to simulate the feedwater, auxiliary feedwater and steam dump systems, which flow rate was determined by system feedback, the Time Dependent Junction was used. With the same rules of primary loops, the components number in secondary loops was numbered in three digits. The first digit stands for the loops number and the other two digits stand for the component types. For instance, the component 520 were heat exchanger in first loop because the first digit 5 represents the first loop and the latter two digits 20 represents the heat exchanger. Component 520 was connected to the right boundary of heat structure 2500, which allows the heat transfer from component 250 in primary side to component 520 in secondary side. Due the heat from primary side, the water in component 520 will evaporate and go through the next component 522. Component 522 was a separator which can increase the quality up to 99.7%. This dried steam will then leave the separator and go through the main steam line isolation valve (component 543), turbine control valve (component 774), turbine stop valve (component 775) and drive turbine, as shown in Figure 6.
As mentioned in section 2, the steam dump system was composed by 10 steam dump valves, six turbine bypass valves and several controlling equipment. To save the computational time, this RELAP5 model merged 10 steam dump valves into four groups. Each group was developed by a Time Dependent Junction component which the total steam flow rate was consistent to the operating conditions. Likewise, six turbine bypass valves were developed by two Time Dependent Junction components.
5
Figure 3 RCP Properties of Maanshan NPP in SNAP Interface 6
Figure 4 Heat Exchanger Component 250 of Maanshan NPP in SNAP Interface Figure 5 Heat Structure 2500 Properties in SNAP Interface To simplify the feedwater control system, the feedwater pumps and valves were developed by Time Dependent Volume and Time Dependent Junction respectively. For the Time Dependent Volume components, the fluid boundary conditions were referred to the thermal hydraulic properties of feedwater during operation. Therefore, the control system need only concern the effect of NRWL, steam flow rate and feedwater flow rate to determine the feedwater flow rate.
Once the flow rate was determined, Time Dependent Junctions which related to the control 7
blocks in the feedwater control system will inject the adequate feedwater into recirculation loops.
Details of the feedwater control system will be discussed in the following section.
Figure 6 Components in Secondary Side of Maanshan NPP in SNAP Interface 2.2 Control Systems In operation, the purpose of water level/feedwater control system is to ensure that water in the steam generator can cover the heat exchanger. For Maanshan NPP, the feedwater flow rate was determined by three units including NRWL in steam generator, steam flow rate and feedwater flow rate. As the water level deviate the setting values, the control system will adjust the injection of the feedwater flow rate to maintain the water level of the steam generator.
Further, two water level measuring systems including the NRWL and WRWL calculate water level with pressure difference. Different from the TRACE model of Maanshan NPP which our group had developed before, there is no water level sensor signal component in the RELAP5 code. As a result, the measurements of water level were developed and composed with density, pressure and volume signal components which was shown in Figure 7.
8
Figure 7 Feedwater Control System in SNAP Interface In addition to the feedwater control, the steam dump system was also an important response mechanism. As mentioned above, the steam dump system of Maanshan NPP can be divided into two types including the pressure control mode and the Tave mode. The pressure mode was initiated as core power was in range from 0% to 10%, which will not be discussed and applied in this research. Hence, the setting of the steam dump system was only referred to the response of Tave mode. As shown in Figure 8, there are three control blocks with sum function calculated the average core temperature values of loop 1 to loop 3 respectively. Then, the control block 308 with max function will compare the maximum of average core temperature (Tave) in loop 1 to loop 3 with No Load Temperature (Tno load, 564K in Maanshan NPP).
Referring this comparison, the control blocks 318 can convert the difference of Tave and Tno load into steam dump flow rate with Table 15. As the temperature difference exceeded 0%
(0), the first group of dump valves was opened. As the temperature difference exceeded 16%
(15.8), the first groups of dump valves were fully opened and the second group of dump valves started to open and so on.
The pressure and water level control system of pressurizer includes the heater and the spray valve. There two types of heater including control heater and backup heater. The control heater and spray valves were applied for adjusting the pressure inside the pressurizer. From Figure 9, the pressure of pressurizer will be compared to rated pressure in control block 120. With the comparison of these two pressure values, the difference can be transferred into the open of the spray valve (control blocks 123) and the power of heater (control block 121 and heat structure 1212 and 1222). However, the control heater is also related to the lower water level of 9
pressurizer (control block 121). As the water level has been lower than 14%, the power of control heater will be zero (control block 124), which means the heater trip. In addition, if the trip setting of control block 121 was assigned to other trip signals, then the control heater can be tripped manually.
Figure 8 Steam Dump Control System in SNAP Interface The backup heater is related to the charging control system of pressurizer. As shown in Figure 10, the maximum core temperature will be transferred into program water level through control block 130. Then, the water level will be subtracted from the actual water level. If the difference of these two water levels is larger than 5%, the backup heater will be initiated (control block 132). Further, the water level will be transferred into charging flow rate (control block 136) to adjust the water level inside the pressurizer. However, as the safety injection signal is initiated, the charging flow rate will be forced to zero.
10
Figure 9 Heater of the Pressurizer Control System in SNAP Interface Figure 10 Pressurizer Injection Control System in SNAP Interface 2.3 Reactor Kinetics In this RELAP5 model of Maanshan NPP, there are two sets of heat structures, which component numbers are 1201 and 1601, developed to simulate the hot fuel channel and average fuel channel. These heat structures were divided into 16 nodes (shown in Figure 11) in axial and 7 nodes in radial (shown in Figure 12). For the axial nodes, they were connected to the Brach components of the reactor core respectively. For radial nodes, the first 4 nodes stand for fuel pellets; the fifth node is filled helium inside the fuel rod and the sixth and seventh nodes 11
are fuel cladding. The materials of each node can be defined manually. In this model, thermal properties (thermal conductivity and thermal capacity) of material 1 for the first 4 nodes were referred to that of Uranium dioxide. The material 2 for node 5 was referred to the helium thermal properties and the material 3 for node 6 and 7 was referred to that of zircalloy.
Figure 11 Properties in Axial Direction of Heat Structure 1201 in SNAP Heat source of the heat structure can be set with the total reactor power or power table. In this model, at the beginning of the model assessment the heat source was set with power table which was referred to the startup test data results of Maanshan NPP to ensure the applicability of the thermal hydraulic components. After that, heat source of the heat structure would be set with total reactor power to ensure the point kinetic feedback calculations. For those heat structure components which were developed as the fuel bundles, the left boundaries were set as symmetry and the right boundaries were connected to the Branch components. For these connections of Brach components and heat structures, the power ratio should be defined (as shown in Figure 13) respectively to calculate the correct heat transfer. The power ratio setting of this RELAP5 model was referred to the TRACE models which were fully developed and assessed before.
12
Figure 12 Properties in Radial Direction of Heat Structure 1201 in SNAP Figure 13 Power Ratio for Branch 100 to 116 of Heat Structure 1201 13
For the point kinetic model, in addition to defining the power ratio of each node of heat structures, the ratio and position of reactivity feedback should also be defined. The reactivity feedback is dominated by Doppler Effect and Moderator density effect. The previous one is related to the temperature of fuel rods; hence, the check list of fuel temperature and reactivity should be added into the Power components. With this table (as shown in Figure 14), the RELAP5 code can calculate the corresponding reactivity feedback due to fuel temperature.
Further, the fuel temperature feedback ratio should also be defined manually in the Heat Weighting settlement of Power component (shown in Figure 15). Similarity, to calculate the Moderator feedback, the checklist of coolant density and reactivity should be defined (shown in Figure 16). Then, with the volume weighting list (shown in Figure 17), the Branch components which were developed as the reactor core would be connected to the point kinetic calculation.
With these settings, the RELAP5 code could calculate the power variation due to temperature and density changes inside the reactor core during transient events.
Figure 14 Doppler Effect Reactivity Feedback Table As mentioned above, the startup assessment transient events were calculated with power table first to ensure the applicability of thermal hydraulic components. Then, the point kinetic model would be applied to do the whole assessment of Maanshan RELAP5 model. As performed with power table, the RELAP5 model needs no control system to simulate reactor scram. However, when performed with point kinetic model, the reactor scram control is required. For instance, Table 100 is the scram reactivity feedback table which will start to dominate the power variation once the trip logic/variable gate is initiated as shown in Figure 18. From this figure, it is obvious that the table could cause a large negative reactivity feedback in few seconds to simulate the control rods insertion.
14
Figure 15 Doppler Effect Heat Structure Weighting Factor 15
Figure 16 Density Effect Reactivity Feedback Table Figure 17 Moderator Density Effect Volume Weighting Factor 16
Figure 18 Reactor Scram Reactivity Feedback Properties 17
3 METHODOLOGY This methodology has four steps which are (1) develop the basic analysis model using RELAP5.
(2) ranking of plant status parameters. (3) ranging of plant status uncertainty parameters (4) generate the input files by random sampling.
(1) Develop the basic analysis model using RELAP5:
RELAP5/MOD3.3 code with SNAP interface is used to do the calculations in this research. The RELAP5 Maanshan NPP analysis model had been developed few years ago. Several verifications and validations had been done, including main steam line isolation valve closure (MSIVC), feedwater pumps trip (FWPT), turbine trip (PAT50),
partial loss of flow (PLOF), complete loss of flow (CLOF) [7].
(2) Ranking of plant status parameters:
Before the uncertainty analysis, it is necessary to do the preliminary analysis or study to determine which uncertainty parameters are important. In this research, a LBLOCA base simulation was done to identify the uncertainty parameters. Initial conditions were according to the real operation conditions. Initial power was set to be 2822MWt and assumed all the systems were operated in the normal status. The uncertainty parameters were evaluated based on the results of the base case, including pressurizer pressure, accumulator (ACC) pressure, ACC volume, ACC water temperature, safety injection (SI) water temperature, feedwater temperature. Detail of the LBLOCA base analysis is indicated in next chapter.
(3) Ranging of plant status uncertainty parameters:
To define the uncertainty parameters of plant status, the uncertainty range and the distribution function are both needed. The uncertainty range in this research are according to Maanshan FSAR and Maanshan Training Manual. The distribution function is set to be uniform distribution in this research. Detail of the uncertainty parameters are listed in chapter.
(4) Generate the input files by random sampling:
DAKOTA code is used to do the sampling in this research. RELAP5 and DAKOTA code are coupled on the SNAP interface. The input uncertainty parameters were sampled and generated the RELAP5 input files. Figure 19 is an example of parameters input, uncertainty range and distribution function setting in DAKOTA.
Wilks formula was adopted to determine the minimum number of runs. In nuclear safety analysis, only upper bond of PCT is the most concern. For the one-sided statistical tolerance limits, the minimum number of runs is given by Wilks formula:
1 is the percentage of tolerance limit, is the confidence level, n is the minimum number of runs. For example, the analysis results in this research achieve / = 95/95 criterion, it means there is 95% confidence level that the results are below the 95% of tolerance limit. To achieve the criterion, the minimum number of runs n = 59.
18
Figure 19 Uncertainty Parameters Setting in DAKOTA 19
4 LBLOCA BASE CASE 4.1 Initial Condition and Basic Assumption The initial conditions and the ECCS properties are listed in Table 1. The initial power is 2822 MWt. The LBLOCA base case was performed in this study. It was assumed that the break occurred at the primary side, at first loop cold-leg near the vessel. The break type was assumed to be double-ended and the break model is shown in Figure 20. Before the pipe rupture, 400 sec steady-state calculation had been done to ensure all the system operated correctly. In the LBLOCA transient, High Head Safety Injection (HHSI), ACC injection, and Low Head Safety Injection (LHSI) are included. It was assumed that all the safety injection systems could operate normally. The safety injection signal setpoint is 11.8 MPa and delay time is 27 sec. The ACC injection set point is 4.24 MPa. The minimum operating pressure of Low Head Safety Injection (LHSI) is 1.15 MPa.
Table 1 Initial Condition and ECCS properties Parameters Values Initial power (MWt) 2822 Core flow rate (kg/s) 14176 Tavg* (K) 582.97 RCS* pressure (MPa) 15.862 Scram setpoint (MPa) 12.8 SI* signal setpoint (MPa) 11.8 SI signal delay (s) 27 ACC liquid volume (m3/tank) 27.9 ACC injection setpoint (MPa) 4.24 ACC liquid temperature (K) 311 LHSI operating pressure(MPa) 1.15 Break size (m2) 0.38 21
Figure 20 RELAP5/SNAP LBLOCA Break Model 4.2 Analysis Results The LBLOCA transient can be divided into six stages, which are critical heat flux phase, upward core flow phase, downward core flow phase, refill phase, reflood phase, and long term cooling phase. The exact postulated event sequence of Maanshan NPP is listed in Table 2.
Table 2 Postulated LBLOCA Events Sequence of Maanshan NPP Action Time (sec)
Pipe rupture 0 Scram 1.8 RCP trip 1.8 SI signal 2.8 ACC injection 5.5 HHSI 29.8 LHSI 29.8 ACC injection ended 69.5 End of event 700 4.2.1 Critical Heat Flux Phase Immediately following the cold-leg break, the fluid discharged out of the pipe with very high flow rates, which can be observed in Figure 21. Due to the extremely high break flow rate, the system pressure drop rapidly. This depressurized phenomenon resulted in water flashing in the 22
reactor vessel. The void fraction of vessel is shown in Figure 22, which indicates that the void fraction went up to one in very few seconds. Because of the flashing phenomenon in reactor core, the fuel cladding became fully uncovered. As shown in Figure 23, the cladding temperature increasing dramatically to the first peak in the very early period.
Figure 21 LBLOCA Break Flow Rate 4.2.2 Upward Core Flow Phase The coolant in loop system still had great inertia at the very beginning (0 ~ 5 sec), and flashing phenomenon in the reactor vessel also generated a great pressure drop. Therefore, there was some two-phase mixture could be pulled into the core, which made the cladding temperature decrease. Because of the bottom-up mixture cooling effect, the temperature drop was more distinct at the lower part. This phenomenon can be seen in Figure 23, there was a temperature drop just after the first peak.
23
Figure 22 Void Fraction in Vessel 4.2.3 Downward Core Flow Phase As pump head continues to degrade, and the coolant slow down. The pressure drop at break started to fully dominate and pull flow down through the core, up through the downcomer to the break. Figure 24 shows the vapor velocity in reactor core, which indicate a velocity increasing downward at about 6 ~ 10 sec. High speed vapor flow also provided good core cooling. This phenomenon can be seen in Figure 23, which shows a second temperature drop at about 10 sec.
4.2.4 Refill Phase Following the blowdown phase, the water level was below the BAF (see Figure 25). Therefore, the cladding temperature increased (see Figure 23). This period is featured of rapid cladding temperature increasing because of lacking water and steam cooling.
24
Figure 23 Cladding Temperature at the Lower Part Figure 24 Core Vapor Velocity 25
4.2.5 Reflood Phase Due to the ACC injection, the water level continued to increase. Figure 25 shows the core collapsed water level. At about 19 sec, the coolant reached the BAF and lower core region began to quench. During this period, core cooling might increase because of upward vapor flow and liquid entrainment. Therefore, the cladding temperature could decrease before the coolant recovered the fuel cladding.
In general, if the income core flow velocity exceeded certain value, it was assumed that the flow regime near the quench front was Inverted-Slug flow. The references [8~10] indicate that the minimum core coolant inlet velocity should be 0.15m/s to form an Inverted-Slug flow regime.
These studies also indicate that the quenching temperature depends strongly on the core inlet coolant velocity.
The flow regime during reflooding phase was shown in Figure 26. Under the quench front, fuel cladding was covered by liquid and cladding temperature dropped to saturation temperature. At inverted-annular and droplet region, the two phase mixture still provided good cooling effect that made the cladding temperature go down. Therefore, this period is the most important in LBLOCA analysis. Figure 27 shows the quench and rewet points for the nodes.
Figure 25 Core Collapsed Water Level 26
4.2.6 Long-Term Core Cooling Whole fuel cladding was recovered by coolant at about 500 sec, which can be seen in Figure
- 27. With the pumped safety injection operating continuously, the water will never below the TAF again.
Figure 26 Core Flow Regime during Reflooding phase 27
Figure 27 Cladding Temperature Axial Profile 28
5 LBLOCA UNCERTAINTY ANALYSIS 5.1 Uncertainty Parameters According to the previous Chapter, the evaluation of quench front progressing during the reflooding phase strongly affects the cladding temperature. The quench front velocity is a relation between TQ and h [11, 12]. TQ is the quenching temperature, and h is the heat transfer coefficient of the coolant under the quench front. TQ and h strongly depend on system pressure and subcooling of the coolant. In LBLOCA case, the performance of ECCS may affect the PCT, system pressure, and subcooling of the coolant. Therefore, the uncertainty analysis by using RELAP5, DAKOTA, and SNAP codes was performed in this study. The uncertainty parameters which are mainly the properties of ECCS are presented in Table 3. These parameters include pressurizer pressure, accumulator (ACC) pressure, ACC volume, ACC water temperature, safety injection (SI) water temperature, and feedwater temperature. The uncertainty ranges are based on FSAR and Training Manual of Maanshan NPP. The distribution functions are all set to be uniform distribution. According to the Wilks formula, 59 of runs should be conducted to achieve the 95/95 criterion. These uncertainty parameters would be randomly sampled and generated 59 RELAP5 input files.
Table 3 Uncertainty Parameters Parameter Deviation PDFs Reference 15.86 MPa 1 PZR pressure FSAR
(+/-0.21MPa) 2 ACC pressure 4.14 - 4.55 MPa Training manual 27.9 m3 (+/-0.425 3 ACC volume Training manual m3) Uniform distribution 4 ACC temperature 310.93 - 338.71 K Training manual 5 SI temperature 300 K (+19.4 K) FSAR Feedwater 6 500 K (+/-1.22 K) Training manual temperature 5.2 Results The core collapsed water level can be seen in Figure 28, which shows the time of end-of-blowdown is estimated at about 19 sec. Reflooding begin at about 37 sec, the time which the coolant reaches the BAF. PCT of 59 cases was shown in Figure 29. During blowdown and refilled phases, the temperature results of 59 cases are very similar. The time of the first temperature peak of 59 cases all occur at about 7 sec and the value of all cases are also almost identical (1293 K). In the beginning of the pipe break, a severe depressurization causes liquid in reactor core suddenly transfer to vapor phase. The heat transfer coefficient of vapor is much lower than liquid makes the cladding temperature reach to the first peak at about 7 sec. It is obvious that the first peak is caused by the heat transfer coefficient difference between liquid and vapor. Therefore, these six uncertainty parameters, which are related to ECCS, have no effect on the first peak. After the coolant reaches BAF, both PCT and water level curves start to have distinct deviations. During this period, the coolant started to attach the lower part of fuel 29
cladding. Some liquid droplets would be entrained by vapor from lower part to upper part that makes the cladding temperature do not increase dramatically like the first peak. Therefore, the deviations of ECCS properties made the cooling effect different and makes different PCT. The red curve in Figure 29 shows the highest PCT of all cases, which is about 1293 K. The PCT of LBLOCA in Maanshan FSAR is 1383.7 K. The difference of analysis results and FSAR data may be caused by the different calculation procedures, phenomenological modeling, and nodalization, etc. In addition, the PCT criteria of 10 CFR 50.46 is 1477.6 K (2200). Comparing the criteria and the results of uncertainty analysis, the results of uncertainty analysis are below the criteria.
Figure 28 Core Collapsed Water Level of 59 Cases 30
Figure 29 PCT of 59 Cases 31
6 CONCLUSIONS This study presents an uncertainty analysis methodology of a postulated LBLOCA of Maanshan NPP using RELAP5/MOD3.3, DAKOTA, and SNAP codes. The results in base case analysis shows that the model of Maanshan NPP is able to manage a LBLOCA simulation. The phenomenon of the base case is consistent with the LOFT experiment and the analysis results of Westinghouse [13, 14]. The evaluation of quench front velocity during reflooding phase strongly affects the PCT. The performance of ECCS may affect reflooding phase. Therefore, an uncertainty analysis was performed in this study.
For the uncertainty analysis, most of the uncertainty parameters are related to ECCS property.
The ranges of uncertainty analysis are based on FSAR and Training Manual of Maanshan NPP.
The maximum PCT of the analysis is 1293 K. This result includes the uncertainty effect. The PCT of Maanshan FSAR is 1383.7 K. Therefore, the prediction of the uncertainty analysis is below the FSAR data. The difference of analysis results and FSAR data may be caused by the different calculation procedures, phenomenological modeling, and nodalization, etc. In addition, the PCT criteria of 10 CFR 50.46 is 1477.6 K (2200). Hence, the result of the uncertainty analysis is also below the criteria of 10 CFR 50.46. This indicates that Maanshan NPP in LBLOCA transient is at a safety situation when ECCS is performed.
33
7 REFERENCES
- 1. Taiwan Power Company, Final Safety Analysis Report for Maanshan Nuclear Power Station Units 1&2 (FSAR), Taiwan Power Company, Republic of China (Taiwan), 1983.
- 2. Jong-Rong Wang, Che-Hao Chen, Hao-Tzu Lin, Chunkuan Shih, Assessment of LONF ATWS for Maanshan PWR Using TRACE Code, NUREG/IA-0436, 2014.
- 3. Hao-Tzu Lin, Jong-Rong Wang, Kai-Chun Huang, Chunkuan Shih, Show-Chyuan Chiang, Chia-Chuan Liu, Station Blackout Mitigation Strategies Analysis for Maanshan PWR Plant using TRACE, Annals of Nuclear Energy, 89, 2016.
- 4. Information Systems Laboratories, Inc., RELAP5/MOD3.3 Code Manual Volume I: Code Structure, System Models, and Solution Methods, Information Systems Laboratories, Inc., Rockville Maryland, 2010.
- 5. Adams, B. M., et al., DAKOTA, a Multilevel Parallel Object-oriented Framework for Design Optimization, Parameter Estimation, Uncertainty Quantification, and Sensitivity Analysis, Sandia National Laboratories, 20-21, 66-67, 99, 2011.
- 6. Applied Programming Technology, Symbolic Nuclear Analysis Package (SNAP) User's Manual, Applied Programming Technology Inc., Bloomsburg, 2007.
- 7. Chunkuan Shih, Jong-Rong Wang, Shao-Wen Chen, Hao-Chun Chang, Show-Chyuan Chiang, and Tzu-Yao Yu, RELAP5/MOD3.3 Model Assessment of Maanshan Nuclear Power Plant with SNAP Interface, NUREG/IA-0472, 2017.
- 8. A.D. Siegel, Rewetting Temperature during Top and Bottom Reflood, M.S. Thesis, University of Illinois, 1979.
- 9. A.D. Siegel and J.J. Carbajo, Trans. ANS 35, pp. 327-328, 19.
- 10. Kevin D. Kimball, Quench Front Propagation during Bottom Reflooding of a Heated Annular Channel, Nuclear Engineering and Design 76, pp. 79-88, 1983.
- 11. Duffey, R.B., Porthouse, D.T.C., The Physics of Rewetting in Water Reactor Emergency Core Cooling. Nuclear Engineering and Design 25, pp. 379~394, 1973.
- 12. B.D.G. Piggott and D.T.C. Porthouse, A Correlation of Rewetting Data. Nuclear Engineering and Design 32, pp. 171~181, 1975.
- 13. S. M. Modro, S. M. Modro, V. T. Berta, A. B. Wahba, Review of LOFT Large Break Experiments, NUREG/IA-0028, 1989.
- 14. Westinghouse Company, Best-Estimate Analysis of the Large-Break Loss-of-Coolant Accident for Maanshan Units 1 and 2 Nuclear Power Plant Using the ASTRUM Methodology, WCAP-17054-P, 2009.
35
NUREG/IA-0509 LBLOCA Uncertainty Analysis of Maanshan Nuclear Power Plant with RELAP5/SNAP and DAKOTA April 2019 Chunkuan Shih, Jong-Rong Wang, Jiang-Ting Chen, Yuh-Ming Ferng, Technical Shao-Wen Chen, Ting-Yi Wang*, and Tzu-Yao Yu*
Institute of Nuclear Engineering and Science
- Department of Nuclear Safety, Taiwan Power Co.
National Tsing Hua University 242, Section 3, Roosevelt Rd., Zhongzheng Dist.
Nuclear and new Energy Education and Research Foundation Taipei, Taiwan 101 Section 2, Kuang Fu Rd., HsinChu, Taiwan Division of Systems Analysis Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington, D.C. 20555-0001 K. Tien, NRC Project Manager The objective of this study is to assess the applicability of the RELAP5/MOD3.3 model of Maanshan NPP on LBLOCA transient. Maanshan NPP was the first three-loop PWR in Taiwan constructed by Westinghouse. In 2015, the RELAP5/MOD3.3 model of Maanshan NPP was developed with SNAP interface. An uncertainty analysis methodology of the postulated LBLOCA for Maanshan NPP using RELAP5/MOD3.3, DAKOTA, and SNAP codes are established in this study. The uncertainty parameters are related to ECCS property. The ranges of uncertainty analysis are based on FSAR and Training Manual of Maanshan. The maximum PCT of the uncertainty analysis is 1293 K. The PCT of FSAR is 1383.7 K. Hence, the result of the uncertainty analysis is below the FSAR data. The difference of uncertainty analysis results and FSAR data may be caused by the different calculation procedures, phenomenological modeling, and nodalization, etc.
Additionally, the PCT criteria of 10 CFR 50.46 is 1477.6 K (2200). The prediction of the uncertainty analysis is below the criteria of 10 CFR 50.46. This implies that Maanshan NPP in LBLOCA transient is at a safety situation when ECCS is performed.
Probability Distribution Function (PDF)
Feedwater Pump(s) Trip (FWPT)
Complete Loss of Flow (CLOF)
High Head Safety Injection (HHSI)
Top of Active Fuel (TAF)
Maanshan Nuclear Power Plant (PWR)
NUREG/IA-0509 International Agreement Report LBLOCA Uncertainty Analysis of Maanshan Nuclear Power Plant with RELAP5/SNAP and DAKOTA Prepared by:
Chunkuan Shih, Jong-Rong Wang, Jian-Ting Chen, Yuh-Ming Ferng, Shao-Wen Chen, Ting-Yi Wang *, and Tzu-Yao Yu*
Institute of Nuclear Engineering and Science, National Tsing Hua University Nuclear and New Energy Education and Research Foundation 101 Section 2, Kuang Fu Rd., HsinChu, Taiwan
- Department of Nuclear Safety, Taiwan Power Company 242, Section 3, Roosevelt Rd., Zhongzheng District, Taipei, Taiwan K. Tien, NRC Project Manager Division of Systems Analysis Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington, DC 20555-0001 Manuscript Completed: June 2018 Date Published: April 2019 Prepared as part of The Agreement on Research Participation and Technical Exchange Under the Thermal-Hydraulic Code Applications and Maintenance Program (CAMP)
Published by U.S. Nuclear Regulatory Commission
AVAILABILITY OF REFERENCE MATERIALS IN NRC PUBLICATIONS NRC Reference Material Non-NRC Reference Material As of November 1999, you may electronically access Documents available from public and special technical NUREG-series publications and other NRC records at libraries include all open literature items, such as books, NRCs Library at www.nrc.gov/reading-rm.html. Publicly journal articles, transactions, Federal Register notices, released records include, to name a few, NUREG-series Federal and State legislation, and congressional reports.
publications; Federal Register notices; applicant, Such documents as theses, dissertations, foreign reports licensee, and vendor documents and correspondence; and translations, and non-NRC conference proceedings NRC correspondence and internal memoranda; bulletins may be purchased from their sponsoring organization.
and information notices; inspection and investigative reports; licensee event reports; and Commission papers Copies of industry codes and standards used in a and their attachments. substantive manner in the NRC regulatory process are maintained at NRC publications in the NUREG series, NRC regulations, The NRC Technical Library and Title 10, Energy, in the Code of Federal Regulations Two White Flint North may also be purchased from one of these two sources. 11545 Rockville Pike Rockville, MD 20852-2738
- 1. The Superintendent of Documents U.S. Government Publishing Office These standards are available in the library for reference Mail Stop IDCC use by the public. Codes and standards are usually Washington, DC 20402-0001 copyrighted and may be purchased from the originating Internet: bookstore.gpo.gov organization or, if they are American National Standards, Telephone: (202) 512-1800 from Fax: (202) 512-2104 American National Standards Institute 11 West 42nd Street
- 2. The National Technical Information Service New York, NY 10036-8002 5301 Shawnee Road www.ansi.org Alexandria, VA 22312-0002 (212) 642-4900 www.ntis.gov 1-800-553-6847 or, locally, (703) 605-6000 Legally binding regulatory requirements are stated only in A single copy of each NRC draft report for comment is laws; NRC regulations; licenses, including technical available free, to the extent of supply, upon written specifications; or orders, not in NUREG-series publications.
request as follows: The views expressed in contractor prepared publications in this series are not necessarily those of the NRC.
Address: U.S. Nuclear Regulatory Commission The NUREG series comprises (1) technical and adminis-Office of Administration trative reports and books prepared by the staff (NUREG-XXXX) or agency contractors (NUREG/CR-XXXX), (2)
Multimedia, Graphics, and Storage &
proceedings of conferences (NUREG/CP-XXXX), (3) reports Distribution Branch resulting from international agreements (NUREG/IA-XXXX),
Washington, DC 20555-0001 (4) brochures (NUREG/BR-XXXX), and (5) compilations of E-mail: distribution.resource@nrc.gov legal decisions and orders of the Commission and Atomic Facsimile: (301) 415-2289 and Safety Licensing Boards and of Directors decisions under Section 2.206 of NRCs regulations (NUREG-0750).
Some publications in the NUREG series that are posted at NRCs Web site address www.nrc.gov/reading-rm/ DISCLAIMER: This report was prepared under an doc-collections/nuregs are updated periodically and may international cooperative agreement for the exchange of technical information. Neither the U.S. Government nor any differ from the last printed version. Although references to agency thereof, nor any employee, makes any warranty, material found on a Web site bear the date the material expressed or implied, or assumes any legal liability or was accessed, the material available on the date cited responsibility for any third partys use, or the results of such may subsequently be removed from the site. use, of any information, apparatus, product or process disclosed in this publication, or represents that its use by such third party would not infringe privately owned rights.
NUREG/IA-0509 International Agreement Report LBLOCA Uncertainty Analysis of Maanshan Nuclear Power Plant with RELAP5/SNAP and DAKOTA Prepared by:
Chunkuan Shih, Jong-Rong Wang, Jian-Ting Chen, Yuh-Ming Ferng, Shao-Wen Chen, Ting-Yi Wang *, and Tzu-Yao Yu*
Institute of Nuclear Engineering and Science, National Tsing Hua University Nuclear and New Energy Education and Research Foundation 101 Section 2, Kuang Fu Rd., HsinChu, Taiwan
- Department of Nuclear Safety, Taiwan Power Company 242, Section 3, Roosevelt Rd., Zhongzheng District, Taipei, Taiwan K. Tien, NRC Project Manager Division of Systems Analysis Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington, DC 20555-0001 Manuscript Completed: June 2018 Date Published: April 2019 Prepared as part of The Agreement on Research Participation and Technical Exchange Under the Thermal-Hydraulic Code Applications and Maintenance Program (CAMP)
Published by U.S. Nuclear Regulatory Commission
ABSTRACT The objective of this study is to assess the applicability of the RELAP5/MOD3.3 model of Maanshan NPP on LBLOCA transient. Maanshan NPP was the first three-loop PWR in Taiwan constructed by Westinghouse. The total power of the two units of the power plant is 2785 MWt, which consist of 2775 MWt for reactor power and 10 MWt for cooling pumps. For the last few years, the TRACE model of Maanshan NPP was developed and several kinds of transient event including the LOCA, AOO and the URG transient were performed and verified. Recently, the RELAP5/MOD3.3 code is another important development priority for our group. In 2015, the RELAP5/MOD3.3 model of Maanshan NPP was developed with SNAP interface.
An uncertainty analysis methodology of the postulated LBLOCA for Maanshan NPP using RELAP5/MOD3.3, DAKOTA, and SNAP codes are established in this study. The uncertainty parameters are related to ECCS property. The ranges of uncertainty analysis are based on FSAR and Training Manual of Maanshan. The maximum PCT of the uncertainty analysis is 1293 K. The PCT of FSAR is 1383.7 K. Hence, the result of the uncertainty analysis is below the FSAR data. The difference of uncertainty analysis results and FSAR data may be caused by the different calculation procedures, phenomenological modeling, and nodalization, etc.
Additionally, the PCT criteria of 10 CFR 50.46 is 1477.6 K (2200°F). The prediction of the uncertainty analysis is below the criteria of 10 CFR 50.46. This implies that Maanshan NPP in LBLOCA transient is at a safety situation when ECCS is performed.
iii
FOREWORD The U.S. NRC has developed a thermal hydraulic analysis code, RELAP5, has been designed to perform best-estimate analysis of loss-of-coolant accidents, operational transients, and other accident scenarios in reactor systems. Models used include multidimensional two-phase flow, non-equilibrium thermo-dynamics, generalized heat transfer, reflood, level tracking, and reactor kinetics. Traditionally, the RELAP5 code analysis model was developed by ASCII file, which was not intelligible for the beginners of computer analysis. Fortunately, and graphic input interface, SNAP is developed by Applied Programming Technology Inc. and conducted by the U.S. NRC, the model development process becomes more convenient.
To obtain the authorization of these codes, Taiwan and the United States have signed an agreement on CAMP (Code Applications and Maintenance Program) which includes the development and maintenance of RELAP5 code. NTHU (National Tsing Hua University) is the organization in Taiwan responsible for the application RELAP5 and SNAP in thermal hydraulic safety analysis. Therefore, the RELAP5/MOD3.3 model of Maanshan nuclear power plant has been developed. To expand the applicability of the RELAP5/MOD3.3 model, an uncertainty analysis methodology of the postulated LBLOCA is established in this study.
v
TABLE OF CONTENTS ABSTRACT ............................................................................................................................... iii FOREWORD ............................................................................................................................... v TABLE OF CONTENTS ........................................................................................................... vii LIST OF FIGURES .................................................................................................................... ix EXECUTIVE
SUMMARY
........................................................................................................... xi ABBREVIATIONS AND ACRONYMS ..................................................................................... xiii 1 INTRODUCTION ............................................................................................................. 1 2 MODEL ESTABLISHMENT ............................................................................................ 3 2.1 Hydraulic Components ........................................................................................ 3 2.2 Control Systems ................................................................................................. 8 2.3 Reactor Kinetics................................................................................................. 11 3 METHODOLOGY .......................................................................................................... 19 4 LBLOCA BASE CASE...21 4.1 Initial Condition and Basic Assumption.....................21 4.2 Analysis Results.....22 4.2.1 Critical Heat Flux Phase...22 4.2.2 Upward Core Flow Phase....23 4.2.3 Downward Core Flow Phase.. 24 4.2.4 Refill Phase 24 4.2.5 Reflood Phase26 4.2.6 Long-Term Core Cooling..27 5 LBLOCA UNCERTAINTY ANALYSIS....29 5.1 Uncertainty Parameters....29 5.2 Results........29 6 CONCLUSIONS ............................................................................................................ 33 7 REFERENCES .............................................................................................................. 35 vii
LIST OF FIGURES Figure 1 Reactor Core Components of Maanshan NPP in SNAP Interface ........................ 4 Figure 2 Components of First Loop of Maanshan NPP in SNAP Interface ......................... 5 Figure 3 RCP Properties of Maanshan NPP in SNAP Interface ......................................... 6 Figure 4 Heat Exchanger Component 250 of Maanshan NPP in SNAP Interface .............. 7 Figure 5 Heat Structure 2500 Properties in SNAP Interface ............................................... 7 Figure 6 Components in Secondary Side of Maanshan NPP in SNAP Interface ................ 8 Figure 7 Feedwater Control System in SNAP Interface ...................................................... 9 Figure 8 Steam Dump Control System in SNAP Interface ................................................ 10 Figure 9 Heater of the Pressurizer Control System in SNAP Interface ............................. 11 Figure 10 Pressurizer Injection Control System in SNAP Interface .................................... 11 Figure 11 Properties in Axial Direction of Heat Structure 1201 in SNAP ............................ 12 Figure 12 Properties in Radial Direction of Heat Structure 1201 in SNAP .......................... 13 Figure 13 Power Ratio for Branch 100 to 116 of Heat Structure 1201 ................................ 13 Figure 14 Doppler Effect Reactivity Feedback Table ......................................................... 14 Figure 15 Doppler Effect Heat Structure Weighting Factor ................................................. 15 Figure 16 Density Effect Reactivity Feedback Table .......................................................... 16 Figure 17 Moderator Density Effect Volume Weighting Factor ........................................... 16 Figure 18 Reactor Scram Reactivity Feedback Properties ................................................. 17 Figure 19 Uncertainty Parameters Setting in DAKOTA ...................................................... 20 Figure 20 RELAP5/SNAP LBLOCA Break Model............................................................... 22 Figure 21 LBLOCA Break Flow Rate ................................................................................. 23 Figure 22 Void Fraction in Vessel ...................................................................................... 24 Figure 23 Cladding Temperature at the Lower Part ........................................................... 25 Figure 24 Core Vapor Velocity ........................................................................................... 25 Figure 25 Core Collapsed Water Level .............................................................................. 26 Figure 26 Core Flow Regime during Reflooding Phase...................................................... 27 Figure 27 Cladding Temperature Axial Profile .................................................................... 28 Figure 28 Core Collapsed Water Level of 59 Cases .......................................................... 30 Figure 29 PCT of 59 Cases................................................................................................ 31 ix
EXECUTIVE
SUMMARY
RELAP5/MOD3.3 Patch04 code, which was developed for light water reactor transient analysis at Idaho National Engineering Laboratory for U.S. NRC, is applied in this research. This code is often performed to support rulemaking, licensing audit calculations, evaluation of accident, mitigation strategies, evaluation of operator guidelines, and experiment planning analysis. Same as other thermal hydraulic analysis codes, RELAP5/MOD3.3 is based on nonhomogeneous and non-equilibrium model for the two-phase system. However, calculations in this code will be solved by a fast, partially implicit numerical scheme to permit economical calculation of system transients. It can produce accurate transient analysis results in relatively short time.
SNAP is an interface of NPP analysis codes which developed by U.S. NRC and Applied Programming Technology, Inc. Different from the traditional input deck in ASCII files, the graphical control blocks and thermal hydraulic connections make researches comprehend the whole power plant and control system more easily. Due to these advantages, the RELAP5/MOD3.3 model of Maanshan NPP was developed with SNAP interface.
DAKOTA code can be used in uncertainty analysis. DAKOTA is applied for the sampling of input parameters, the calculation of correlations, and ranking of input parameters. DAKOTA provides Monte Carlo sampling and Latin Hypercube sampling methods combined with various PDFs including normal, lognormal, uniform, logunifrom, hypergeometric, and user-supplied histograms. In addition, the uncertainty analysis of DAKOTA also can run with the SNAP interface.
Maanshan NPP is located on the southern coast of Taiwan. Its nuclear steam supply system is a type of PWR designed and built by Westinghouse for Taiwan Power Company. In this research, a RELAP5/MOD3.3 model of Maanshan NPP for LBLOCA transient was developed.
This research also presents an uncertainty analysis methodology of LBLOCA using RELAP5/MOD3.3, DAKOTA, and SNAP codes.
The results of the RELAP5 analysis for LBLOCA base case indicate that the RELAP5/SNAP model is able to manage a LBLOCA simulation. The phenomenon of the base case is consistent with the LOFT experiment and the analysis results of Westinghouse. According to analysis results, the evaluation of quench front velocity during reflooding phase strongly affects the PCT.
The performance of ECCS may affect reflooding phase. Therefore, an uncertainty analysis was performed in this research.
Most of the uncertainty parameters are related to ECCS property in the uncertainty analysis.
The ranges of uncertainty analysis are based on FSAR and Training Manual of Maanshan NPP.
The maximum PCT of the analysis is 1293 K that includes the uncertainty effect. However, the PCT of Maanshan FSAR is 1383.7 K. The difference of analysis results and FSAR data may be caused by the different calculation procedures, phenomenological modeling, and nodalization, etc. In addition, the PCT criteria of 10 CFR 50.46 is 1477.6 K (2200). Hence, the result of the uncertainty analysis is below the criteria of 10 CFR 50.46. This indicates that Maanshan NPP in LBLOCA transient is at a safety situation when ECCS is performed.
xi
ABBREVIATIONS AND ACRONYMS ACC Accumulator BAF Bottom of Active Fuel CAMP Code Applications and Maintenance Program CLOF Completed Loss of Flow ECCS Emergency Core Cooling System FWPT Feedwater pump(s) Trip FSAR Final Safety Analysis Report HHSI High Head Safety Injection kg kilogram(s)
LBLOCA Large Break Loss Of Coolant Accident LHSI Low Head Safety Injection MPa Megapascal(s)
MSIVC Main Steam Isolation Valves Closure NPP Nuclear Power Plant NRC Nuclear Regulatory Commission NRWL Narrow Range Water Level NSSS Nuclear Steam Supply System NTHU National Tsing Hua University PDF Probability Distribution Function PORVs Power-Operated Relief Valves PWR Pressurized Light Water Reactor PLOF Partial Loss of Flow RCS Reactor Coolant System RCP Reactor Coolant Pump S/G Steam Generator SI Safety Injection SNAP Symbolic Nuclear Analysis Program SRV Safety/Relief Valves TAF Top of Active Fuel TRACE TRAC/RELAP Advanced Computational Engine WRWL Wide Range Water Level xiii
1 INTRODUCTION Maanshan Nuclear Power Plant (NPP) is the third NPP in Taiwan. Also, it is the first Pressurized Water Reactor (PWR) located at the south of Taiwan. There are two units in the Maanshan NPP. The total power of the nuclear steam supply system is 2785 MWt, which consist of 2775 MWt for reactor power and 10 MWt for cooling pumps [1]. For the last few years, our group has developed the models of Taiwan NPPs with TRACE code in SNAP interface [2, 3]. Further, it is necessary to perform the NPP transients with several analysis codes so that the data results could be compared with each other to ensure the consistency. Therefore, the RELAP5/MOD 3.3 Patch04 code was chosen to develop a new Maanshan NPP model. Different from the traditional ASCII input deck, the RELAP5/MOD 3.3 model was developed with SNAP interface.
DAKOTA is an uncertainty analysis code [5] and is applied for the sampling of input parameters, the calculation of correlations, and ranking of input parameters. The uncertainty quantification package of DAKOTA provides Monte Carlo sampling and Latin Hypercube sampling methods combined with various PDFs including normal, lognormal, uniform, logunifrom, hypergeometric, and user-supplied histograms.
SNAP is an interface of NPP analysis codes which developed by U.S. NRC and Applied Programming Technology, Inc. Different from the traditional input deck in ASCII files, the graphical control blocks and thermal hydraulic connections make researches comprehend the whole power plant and control system more easily [6]. Due to these advantages, an uncertainty analysis methodology of LBLOCA using RELAP5/MOD3.3 and DAKOTA with SNAP interface is established in this study. Moreover, due to the SNAP interface, the analysis results could be transferred into animations which were more attractive and more understandable. With the animation, interactions of different components and parameters could be easily observed.
To ensure the applicability of this model, three startup tests cases including feedwater pumps trip (FWPT), turbine trip (PAT50) and main isolation valves closure (MSIVC) had been analyzed in 2015 [7]. With the comparison of RELAP5 results and startup tests data, it shows that the RELAP5/MOD 3.3 model of Maanshan NPP is consistent with the startup tests data. To expand the applicability of this RELAP5 model, the postulated LBLOCA is performed. Two main steps are in this study. First, the establishment of RELAP5/MOD3.3 model for the postulated LBLOCA of Maanshan NPP is performed. An analysis of LBLOCA base case is performed by using this model to confirm the phenomenon of LBLOCA. Second, to confirm the uncertainty of ECCS that may affect the PCT, the uncertainty analysis of RELAP5/MOD 3.3 model combining with DAKOTA is performed in this study.
1
2 MODEL ESTABLISHMENT As the Maanshan NPP operated in normal conditions, coolant water in primary system will carry the heat generated by the fuel rods to the steam generator. Feedwater in the secondary system then obtain the heat, evaporate and drive turbines to generate electricity. According to the energy conservation principle, internal energy of steam, which had driven turbines, will decrease. This lower internal energy steam will then go through the condenser and be transferred into feedwater and re-injected into the steam generator. However, it is difficult to develop the entire recirculation system with the analysis code. The computational time will be impractically long and the mass balance will be hard to reach. Hence, as developing the RELAP5 model, it is practical to define the feedwater pumps and the turbines as the boundary conditions because the main purpose of this model is to obtain the NSSS reactions during the transient. For the NSSS system of Maanshan NPP, the feedwater pumps, auxiliary feedwater pumps, turbines, safety/relief valves, steam dump valves and PORVs were defined as the boundary conditions and developed by the Time Dependent Volume component in the RELAP5 program [7].
2.1 Hydraulic Components As mentioned in section 2, there are three recirculation loops in the Maanshan NPP. In each loop, there is a RCP and S/G. On the hot leg of second loop, a pressurizer which can adjust the pressure of RCS with the spray valves and electronic heater was developed. In this analysis model, there are several Branch components developed to simulate the reactor vessel.
According to the core arrangement, Branch components from number 140 to 156 were connected together as the average fuel channel. Branch components from number 120 to 136 were connected together as the hottest fuel channel. Branch components from number 100 to 116 were connected together as the bypass flow channel, as shown in Figure 1. Also, these channels will be connected to the heat structure components to obtain the heat and do the reactor kinetic analysis.
For those three recirculation loops in primary side, they were developed by Pipe, Valve, Branch, Jump and Single Volume, as shown in Figure 2. For these three-digit components, the first digit stands for the loop number (2 for first loop, 3 for second loop and 4 for third loop). Further, the other digits of these components represents to the component types. For instance, component 280 is the recirculation pumps in first loop and component 380 is the recirculation pumps in second loop. Though the Pump component in RELAP5 code has been developed with the pump parameters from Westinghouse, pump characters of the RCPs in this model was input according to Taiwan Power Company NPP training materials and past research models which were calculated by RELAP5-3D and TRACE codes, as shown in Figure 3.
In addition to RCPs, another important thermal hydraulic component in the primary side is heat exchanger. Pipe 250, which was developed for heat exchanger in first loop, was divided into eight nodes. According to the geometry of heat exchanger, junction between fourth cell and fifth cell was 180 degree as shown in Figure 4. Further, the heat structure component can be view as structural component once the both side of the heat structure were connected to thermal hydraulic components. Hence, the Pipe 250 was connected to the left boundary of heat structure 2500, as shown in Figure 5. Likewise, the Pipe 350 of secondary loop was connected to the left boundary of heat structure 3500 and Pipe 450 of third loop was connected to the left boundary of heat structure 4500.
3
Figure 1 Reactor Core Components of Maanshan NPP in SNAP Interface 4
Figure 2 Components of First Loop of Maanshan NPP in SNAP Interface Similar to primary loops, the secondary loops of Maanshan NPP were developed with Pipe, Valve, Branch, Pump and Single Volume. Specially, to simulate the feedwater, auxiliary feedwater and steam dump systems, which flow rate was determined by system feedback, the Time Dependent Junction was used. With the same rules of primary loops, the components number in secondary loops was numbered in three digits. The first digit stands for the loops number and the other two digits stand for the component types. For instance, the component 520 were heat exchanger in first loop because the first digit 5 represents the first loop and the latter two digits 20 represents the heat exchanger. Component 520 was connected to the right boundary of heat structure 2500, which allows the heat transfer from component 250 in primary side to component 520 in secondary side. Due the heat from primary side, the water in component 520 will evaporate and go through the next component 522. Component 522 was a separator which can increase the quality up to 99.7%. This dried steam will then leave the separator and go through the main steam line isolation valve (component 543), turbine control valve (component 774), turbine stop valve (component 775) and drive turbine, as shown in Figure 6.
As mentioned in section 2, the steam dump system was composed by 10 steam dump valves, six turbine bypass valves and several controlling equipment. To save the computational time, this RELAP5 model merged 10 steam dump valves into four groups. Each group was developed by a Time Dependent Junction component which the total steam flow rate was consistent to the operating conditions. Likewise, six turbine bypass valves were developed by two Time Dependent Junction components.
5
Figure 3 RCP Properties of Maanshan NPP in SNAP Interface 6
Figure 4 Heat Exchanger Component 250 of Maanshan NPP in SNAP Interface Figure 5 Heat Structure 2500 Properties in SNAP Interface To simplify the feedwater control system, the feedwater pumps and valves were developed by Time Dependent Volume and Time Dependent Junction respectively. For the Time Dependent Volume components, the fluid boundary conditions were referred to the thermal hydraulic properties of feedwater during operation. Therefore, the control system need only concern the effect of NRWL, steam flow rate and feedwater flow rate to determine the feedwater flow rate.
Once the flow rate was determined, Time Dependent Junctions which related to the control 7
blocks in the feedwater control system will inject the adequate feedwater into recirculation loops.
Details of the feedwater control system will be discussed in the following section.
Figure 6 Components in Secondary Side of Maanshan NPP in SNAP Interface 2.2 Control Systems In operation, the purpose of water level/feedwater control system is to ensure that water in the steam generator can cover the heat exchanger. For Maanshan NPP, the feedwater flow rate was determined by three units including NRWL in steam generator, steam flow rate and feedwater flow rate. As the water level deviate the setting values, the control system will adjust the injection of the feedwater flow rate to maintain the water level of the steam generator.
Further, two water level measuring systems including the NRWL and WRWL calculate water level with pressure difference. Different from the TRACE model of Maanshan NPP which our group had developed before, there is no water level sensor signal component in the RELAP5 code. As a result, the measurements of water level were developed and composed with density, pressure and volume signal components which was shown in Figure 7.
8
Figure 7 Feedwater Control System in SNAP Interface In addition to the feedwater control, the steam dump system was also an important response mechanism. As mentioned above, the steam dump system of Maanshan NPP can be divided into two types including the pressure control mode and the Tave mode. The pressure mode was initiated as core power was in range from 0% to 10%, which will not be discussed and applied in this research. Hence, the setting of the steam dump system was only referred to the response of Tave mode. As shown in Figure 8, there are three control blocks with sum function calculated the average core temperature values of loop 1 to loop 3 respectively. Then, the control block 308 with max function will compare the maximum of average core temperature (Tave) in loop 1 to loop 3 with No Load Temperature (Tno load, 564K in Maanshan NPP).
Referring this comparison, the control blocks 318 can convert the difference of Tave and Tno load into steam dump flow rate with Table 15. As the temperature difference exceeded 0%
(0), the first group of dump valves was opened. As the temperature difference exceeded 16%
(15.8), the first groups of dump valves were fully opened and the second group of dump valves started to open and so on.
The pressure and water level control system of pressurizer includes the heater and the spray valve. There two types of heater including control heater and backup heater. The control heater and spray valves were applied for adjusting the pressure inside the pressurizer. From Figure 9, the pressure of pressurizer will be compared to rated pressure in control block 120. With the comparison of these two pressure values, the difference can be transferred into the open of the spray valve (control blocks 123) and the power of heater (control block 121 and heat structure 1212 and 1222). However, the control heater is also related to the lower water level of 9
pressurizer (control block 121). As the water level has been lower than 14%, the power of control heater will be zero (control block 124), which means the heater trip. In addition, if the trip setting of control block 121 was assigned to other trip signals, then the control heater can be tripped manually.
Figure 8 Steam Dump Control System in SNAP Interface The backup heater is related to the charging control system of pressurizer. As shown in Figure 10, the maximum core temperature will be transferred into program water level through control block 130. Then, the water level will be subtracted from the actual water level. If the difference of these two water levels is larger than 5%, the backup heater will be initiated (control block 132). Further, the water level will be transferred into charging flow rate (control block 136) to adjust the water level inside the pressurizer. However, as the safety injection signal is initiated, the charging flow rate will be forced to zero.
10
Figure 9 Heater of the Pressurizer Control System in SNAP Interface Figure 10 Pressurizer Injection Control System in SNAP Interface 2.3 Reactor Kinetics In this RELAP5 model of Maanshan NPP, there are two sets of heat structures, which component numbers are 1201 and 1601, developed to simulate the hot fuel channel and average fuel channel. These heat structures were divided into 16 nodes (shown in Figure 11) in axial and 7 nodes in radial (shown in Figure 12). For the axial nodes, they were connected to the Brach components of the reactor core respectively. For radial nodes, the first 4 nodes stand for fuel pellets; the fifth node is filled helium inside the fuel rod and the sixth and seventh nodes 11
are fuel cladding. The materials of each node can be defined manually. In this model, thermal properties (thermal conductivity and thermal capacity) of material 1 for the first 4 nodes were referred to that of Uranium dioxide. The material 2 for node 5 was referred to the helium thermal properties and the material 3 for node 6 and 7 was referred to that of zircalloy.
Figure 11 Properties in Axial Direction of Heat Structure 1201 in SNAP Heat source of the heat structure can be set with the total reactor power or power table. In this model, at the beginning of the model assessment the heat source was set with power table which was referred to the startup test data results of Maanshan NPP to ensure the applicability of the thermal hydraulic components. After that, heat source of the heat structure would be set with total reactor power to ensure the point kinetic feedback calculations. For those heat structure components which were developed as the fuel bundles, the left boundaries were set as symmetry and the right boundaries were connected to the Branch components. For these connections of Brach components and heat structures, the power ratio should be defined (as shown in Figure 13) respectively to calculate the correct heat transfer. The power ratio setting of this RELAP5 model was referred to the TRACE models which were fully developed and assessed before.
12
Figure 12 Properties in Radial Direction of Heat Structure 1201 in SNAP Figure 13 Power Ratio for Branch 100 to 116 of Heat Structure 1201 13
For the point kinetic model, in addition to defining the power ratio of each node of heat structures, the ratio and position of reactivity feedback should also be defined. The reactivity feedback is dominated by Doppler Effect and Moderator density effect. The previous one is related to the temperature of fuel rods; hence, the check list of fuel temperature and reactivity should be added into the Power components. With this table (as shown in Figure 14), the RELAP5 code can calculate the corresponding reactivity feedback due to fuel temperature.
Further, the fuel temperature feedback ratio should also be defined manually in the Heat Weighting settlement of Power component (shown in Figure 15). Similarity, to calculate the Moderator feedback, the checklist of coolant density and reactivity should be defined (shown in Figure 16). Then, with the volume weighting list (shown in Figure 17), the Branch components which were developed as the reactor core would be connected to the point kinetic calculation.
With these settings, the RELAP5 code could calculate the power variation due to temperature and density changes inside the reactor core during transient events.
Figure 14 Doppler Effect Reactivity Feedback Table As mentioned above, the startup assessment transient events were calculated with power table first to ensure the applicability of thermal hydraulic components. Then, the point kinetic model would be applied to do the whole assessment of Maanshan RELAP5 model. As performed with power table, the RELAP5 model needs no control system to simulate reactor scram. However, when performed with point kinetic model, the reactor scram control is required. For instance, Table 100 is the scram reactivity feedback table which will start to dominate the power variation once the trip logic/variable gate is initiated as shown in Figure 18. From this figure, it is obvious that the table could cause a large negative reactivity feedback in few seconds to simulate the control rods insertion.
14
Figure 15 Doppler Effect Heat Structure Weighting Factor 15
Figure 16 Density Effect Reactivity Feedback Table Figure 17 Moderator Density Effect Volume Weighting Factor 16
Figure 18 Reactor Scram Reactivity Feedback Properties 17
3 METHODOLOGY This methodology has four steps which are (1) develop the basic analysis model using RELAP5.
(2) ranking of plant status parameters. (3) ranging of plant status uncertainty parameters (4) generate the input files by random sampling.
(1) Develop the basic analysis model using RELAP5:
RELAP5/MOD3.3 code with SNAP interface is used to do the calculations in this research. The RELAP5 Maanshan NPP analysis model had been developed few years ago. Several verifications and validations had been done, including main steam line isolation valve closure (MSIVC), feedwater pumps trip (FWPT), turbine trip (PAT50),
partial loss of flow (PLOF), complete loss of flow (CLOF) [7].
(2) Ranking of plant status parameters:
Before the uncertainty analysis, it is necessary to do the preliminary analysis or study to determine which uncertainty parameters are important. In this research, a LBLOCA base simulation was done to identify the uncertainty parameters. Initial conditions were according to the real operation conditions. Initial power was set to be 2822MWt and assumed all the systems were operated in the normal status. The uncertainty parameters were evaluated based on the results of the base case, including pressurizer pressure, accumulator (ACC) pressure, ACC volume, ACC water temperature, safety injection (SI) water temperature, feedwater temperature. Detail of the LBLOCA base analysis is indicated in next chapter.
(3) Ranging of plant status uncertainty parameters:
To define the uncertainty parameters of plant status, the uncertainty range and the distribution function are both needed. The uncertainty range in this research are according to Maanshan FSAR and Maanshan Training Manual. The distribution function is set to be uniform distribution in this research. Detail of the uncertainty parameters are listed in chapter.
(4) Generate the input files by random sampling:
DAKOTA code is used to do the sampling in this research. RELAP5 and DAKOTA code are coupled on the SNAP interface. The input uncertainty parameters were sampled and generated the RELAP5 input files. Figure 19 is an example of parameters input, uncertainty range and distribution function setting in DAKOTA.
Wilks formula was adopted to determine the minimum number of runs. In nuclear safety analysis, only upper bond of PCT is the most concern. For the one-sided statistical tolerance limits, the minimum number of runs is given by Wilks formula:
1 is the percentage of tolerance limit, is the confidence level, n is the minimum number of runs. For example, the analysis results in this research achieve / = 95/95 criterion, it means there is 95% confidence level that the results are below the 95% of tolerance limit. To achieve the criterion, the minimum number of runs n = 59.
18
Figure 19 Uncertainty Parameters Setting in DAKOTA 19
4 LBLOCA BASE CASE 4.1 Initial Condition and Basic Assumption The initial conditions and the ECCS properties are listed in Table 1. The initial power is 2822 MWt. The LBLOCA base case was performed in this study. It was assumed that the break occurred at the primary side, at first loop cold-leg near the vessel. The break type was assumed to be double-ended and the break model is shown in Figure 20. Before the pipe rupture, 400 sec steady-state calculation had been done to ensure all the system operated correctly. In the LBLOCA transient, High Head Safety Injection (HHSI), ACC injection, and Low Head Safety Injection (LHSI) are included. It was assumed that all the safety injection systems could operate normally. The safety injection signal setpoint is 11.8 MPa and delay time is 27 sec. The ACC injection set point is 4.24 MPa. The minimum operating pressure of Low Head Safety Injection (LHSI) is 1.15 MPa.
Table 1 Initial Condition and ECCS properties Parameters Values Initial power (MWt) 2822 Core flow rate (kg/s) 14176 Tavg* (K) 582.97 RCS* pressure (MPa) 15.862 Scram setpoint (MPa) 12.8 SI* signal setpoint (MPa) 11.8 SI signal delay (s) 27 ACC liquid volume (m3/tank) 27.9 ACC injection setpoint (MPa) 4.24 ACC liquid temperature (K) 311 LHSI operating pressure(MPa) 1.15 Break size (m2) 0.38 21
Figure 20 RELAP5/SNAP LBLOCA Break Model 4.2 Analysis Results The LBLOCA transient can be divided into six stages, which are critical heat flux phase, upward core flow phase, downward core flow phase, refill phase, reflood phase, and long term cooling phase. The exact postulated event sequence of Maanshan NPP is listed in Table 2.
Table 2 Postulated LBLOCA Events Sequence of Maanshan NPP Action Time (sec)
Pipe rupture 0 Scram 1.8 RCP trip 1.8 SI signal 2.8 ACC injection 5.5 HHSI 29.8 LHSI 29.8 ACC injection ended 69.5 End of event 700 4.2.1 Critical Heat Flux Phase Immediately following the cold-leg break, the fluid discharged out of the pipe with very high flow rates, which can be observed in Figure 21. Due to the extremely high break flow rate, the system pressure drop rapidly. This depressurized phenomenon resulted in water flashing in the 22
reactor vessel. The void fraction of vessel is shown in Figure 22, which indicates that the void fraction went up to one in very few seconds. Because of the flashing phenomenon in reactor core, the fuel cladding became fully uncovered. As shown in Figure 23, the cladding temperature increasing dramatically to the first peak in the very early period.
Figure 21 LBLOCA Break Flow Rate 4.2.2 Upward Core Flow Phase The coolant in loop system still had great inertia at the very beginning (0 ~ 5 sec), and flashing phenomenon in the reactor vessel also generated a great pressure drop. Therefore, there was some two-phase mixture could be pulled into the core, which made the cladding temperature decrease. Because of the bottom-up mixture cooling effect, the temperature drop was more distinct at the lower part. This phenomenon can be seen in Figure 23, there was a temperature drop just after the first peak.
23
Figure 22 Void Fraction in Vessel 4.2.3 Downward Core Flow Phase As pump head continues to degrade, and the coolant slow down. The pressure drop at break started to fully dominate and pull flow down through the core, up through the downcomer to the break. Figure 24 shows the vapor velocity in reactor core, which indicate a velocity increasing downward at about 6 ~ 10 sec. High speed vapor flow also provided good core cooling. This phenomenon can be seen in Figure 23, which shows a second temperature drop at about 10 sec.
4.2.4 Refill Phase Following the blowdown phase, the water level was below the BAF (see Figure 25). Therefore, the cladding temperature increased (see Figure 23). This period is featured of rapid cladding temperature increasing because of lacking water and steam cooling.
24
Figure 23 Cladding Temperature at the Lower Part Figure 24 Core Vapor Velocity 25
4.2.5 Reflood Phase Due to the ACC injection, the water level continued to increase. Figure 25 shows the core collapsed water level. At about 19 sec, the coolant reached the BAF and lower core region began to quench. During this period, core cooling might increase because of upward vapor flow and liquid entrainment. Therefore, the cladding temperature could decrease before the coolant recovered the fuel cladding.
In general, if the income core flow velocity exceeded certain value, it was assumed that the flow regime near the quench front was Inverted-Slug flow. The references [8~10] indicate that the minimum core coolant inlet velocity should be 0.15m/s to form an Inverted-Slug flow regime.
These studies also indicate that the quenching temperature depends strongly on the core inlet coolant velocity.
The flow regime during reflooding phase was shown in Figure 26. Under the quench front, fuel cladding was covered by liquid and cladding temperature dropped to saturation temperature. At inverted-annular and droplet region, the two phase mixture still provided good cooling effect that made the cladding temperature go down. Therefore, this period is the most important in LBLOCA analysis. Figure 27 shows the quench and rewet points for the nodes.
Figure 25 Core Collapsed Water Level 26
4.2.6 Long-Term Core Cooling Whole fuel cladding was recovered by coolant at about 500 sec, which can be seen in Figure
- 27. With the pumped safety injection operating continuously, the water will never below the TAF again.
Figure 26 Core Flow Regime during Reflooding phase 27
Figure 27 Cladding Temperature Axial Profile 28
5 LBLOCA UNCERTAINTY ANALYSIS 5.1 Uncertainty Parameters According to the previous Chapter, the evaluation of quench front progressing during the reflooding phase strongly affects the cladding temperature. The quench front velocity is a relation between TQ and h [11, 12]. TQ is the quenching temperature, and h is the heat transfer coefficient of the coolant under the quench front. TQ and h strongly depend on system pressure and subcooling of the coolant. In LBLOCA case, the performance of ECCS may affect the PCT, system pressure, and subcooling of the coolant. Therefore, the uncertainty analysis by using RELAP5, DAKOTA, and SNAP codes was performed in this study. The uncertainty parameters which are mainly the properties of ECCS are presented in Table 3. These parameters include pressurizer pressure, accumulator (ACC) pressure, ACC volume, ACC water temperature, safety injection (SI) water temperature, and feedwater temperature. The uncertainty ranges are based on FSAR and Training Manual of Maanshan NPP. The distribution functions are all set to be uniform distribution. According to the Wilks formula, 59 of runs should be conducted to achieve the 95/95 criterion. These uncertainty parameters would be randomly sampled and generated 59 RELAP5 input files.
Table 3 Uncertainty Parameters Parameter Deviation PDFs Reference 15.86 MPa 1 PZR pressure FSAR
(+/-0.21MPa) 2 ACC pressure 4.14 - 4.55 MPa Training manual 27.9 m3 (+/-0.425 3 ACC volume Training manual m3) Uniform distribution 4 ACC temperature 310.93 - 338.71 K Training manual 5 SI temperature 300 K (+19.4 K) FSAR Feedwater 6 500 K (+/-1.22 K) Training manual temperature 5.2 Results The core collapsed water level can be seen in Figure 28, which shows the time of end-of-blowdown is estimated at about 19 sec. Reflooding begin at about 37 sec, the time which the coolant reaches the BAF. PCT of 59 cases was shown in Figure 29. During blowdown and refilled phases, the temperature results of 59 cases are very similar. The time of the first temperature peak of 59 cases all occur at about 7 sec and the value of all cases are also almost identical (1293 K). In the beginning of the pipe break, a severe depressurization causes liquid in reactor core suddenly transfer to vapor phase. The heat transfer coefficient of vapor is much lower than liquid makes the cladding temperature reach to the first peak at about 7 sec. It is obvious that the first peak is caused by the heat transfer coefficient difference between liquid and vapor. Therefore, these six uncertainty parameters, which are related to ECCS, have no effect on the first peak. After the coolant reaches BAF, both PCT and water level curves start to have distinct deviations. During this period, the coolant started to attach the lower part of fuel 29
cladding. Some liquid droplets would be entrained by vapor from lower part to upper part that makes the cladding temperature do not increase dramatically like the first peak. Therefore, the deviations of ECCS properties made the cooling effect different and makes different PCT. The red curve in Figure 29 shows the highest PCT of all cases, which is about 1293 K. The PCT of LBLOCA in Maanshan FSAR is 1383.7 K. The difference of analysis results and FSAR data may be caused by the different calculation procedures, phenomenological modeling, and nodalization, etc. In addition, the PCT criteria of 10 CFR 50.46 is 1477.6 K (2200). Comparing the criteria and the results of uncertainty analysis, the results of uncertainty analysis are below the criteria.
Figure 28 Core Collapsed Water Level of 59 Cases 30
Figure 29 PCT of 59 Cases 31
6 CONCLUSIONS This study presents an uncertainty analysis methodology of a postulated LBLOCA of Maanshan NPP using RELAP5/MOD3.3, DAKOTA, and SNAP codes. The results in base case analysis shows that the model of Maanshan NPP is able to manage a LBLOCA simulation. The phenomenon of the base case is consistent with the LOFT experiment and the analysis results of Westinghouse [13, 14]. The evaluation of quench front velocity during reflooding phase strongly affects the PCT. The performance of ECCS may affect reflooding phase. Therefore, an uncertainty analysis was performed in this study.
For the uncertainty analysis, most of the uncertainty parameters are related to ECCS property.
The ranges of uncertainty analysis are based on FSAR and Training Manual of Maanshan NPP.
The maximum PCT of the analysis is 1293 K. This result includes the uncertainty effect. The PCT of Maanshan FSAR is 1383.7 K. Therefore, the prediction of the uncertainty analysis is below the FSAR data. The difference of analysis results and FSAR data may be caused by the different calculation procedures, phenomenological modeling, and nodalization, etc. In addition, the PCT criteria of 10 CFR 50.46 is 1477.6 K (2200). Hence, the result of the uncertainty analysis is also below the criteria of 10 CFR 50.46. This indicates that Maanshan NPP in LBLOCA transient is at a safety situation when ECCS is performed.
33
7 REFERENCES
- 1. Taiwan Power Company, Final Safety Analysis Report for Maanshan Nuclear Power Station Units 1&2 (FSAR), Taiwan Power Company, Republic of China (Taiwan), 1983.
- 2. Jong-Rong Wang, Che-Hao Chen, Hao-Tzu Lin, Chunkuan Shih, Assessment of LONF ATWS for Maanshan PWR Using TRACE Code, NUREG/IA-0436, 2014.
- 3. Hao-Tzu Lin, Jong-Rong Wang, Kai-Chun Huang, Chunkuan Shih, Show-Chyuan Chiang, Chia-Chuan Liu, Station Blackout Mitigation Strategies Analysis for Maanshan PWR Plant using TRACE, Annals of Nuclear Energy, 89, 2016.
- 4. Information Systems Laboratories, Inc., RELAP5/MOD3.3 Code Manual Volume I: Code Structure, System Models, and Solution Methods, Information Systems Laboratories, Inc., Rockville Maryland, 2010.
- 5. Adams, B. M., et al., DAKOTA, a Multilevel Parallel Object-oriented Framework for Design Optimization, Parameter Estimation, Uncertainty Quantification, and Sensitivity Analysis, Sandia National Laboratories, 20-21, 66-67, 99, 2011.
- 6. Applied Programming Technology, Symbolic Nuclear Analysis Package (SNAP) User's Manual, Applied Programming Technology Inc., Bloomsburg, 2007.
- 7. Chunkuan Shih, Jong-Rong Wang, Shao-Wen Chen, Hao-Chun Chang, Show-Chyuan Chiang, and Tzu-Yao Yu, RELAP5/MOD3.3 Model Assessment of Maanshan Nuclear Power Plant with SNAP Interface, NUREG/IA-0472, 2017.
- 8. A.D. Siegel, Rewetting Temperature during Top and Bottom Reflood, M.S. Thesis, University of Illinois, 1979.
- 9. A.D. Siegel and J.J. Carbajo, Trans. ANS 35, pp. 327-328, 19.
- 10. Kevin D. Kimball, Quench Front Propagation during Bottom Reflooding of a Heated Annular Channel, Nuclear Engineering and Design 76, pp. 79-88, 1983.
- 11. Duffey, R.B., Porthouse, D.T.C., The Physics of Rewetting in Water Reactor Emergency Core Cooling. Nuclear Engineering and Design 25, pp. 379~394, 1973.
- 12. B.D.G. Piggott and D.T.C. Porthouse, A Correlation of Rewetting Data. Nuclear Engineering and Design 32, pp. 171~181, 1975.
- 13. S. M. Modro, S. M. Modro, V. T. Berta, A. B. Wahba, Review of LOFT Large Break Experiments, NUREG/IA-0028, 1989.
- 14. Westinghouse Company, Best-Estimate Analysis of the Large-Break Loss-of-Coolant Accident for Maanshan Units 1 and 2 Nuclear Power Plant Using the ASTRUM Methodology, WCAP-17054-P, 2009.
35
NUREG/IA-0509 LBLOCA Uncertainty Analysis of Maanshan Nuclear Power Plant with RELAP5/SNAP and DAKOTA April 2019 Chunkuan Shih, Jong-Rong Wang, Jiang-Ting Chen, Yuh-Ming Ferng, Technical Shao-Wen Chen, Ting-Yi Wang*, and Tzu-Yao Yu*
Institute of Nuclear Engineering and Science
- Department of Nuclear Safety, Taiwan Power Co.
National Tsing Hua University 242, Section 3, Roosevelt Rd., Zhongzheng Dist.
Nuclear and new Energy Education and Research Foundation Taipei, Taiwan 101 Section 2, Kuang Fu Rd., HsinChu, Taiwan Division of Systems Analysis Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington, D.C. 20555-0001 K. Tien, NRC Project Manager The objective of this study is to assess the applicability of the RELAP5/MOD3.3 model of Maanshan NPP on LBLOCA transient. Maanshan NPP was the first three-loop PWR in Taiwan constructed by Westinghouse. In 2015, the RELAP5/MOD3.3 model of Maanshan NPP was developed with SNAP interface. An uncertainty analysis methodology of the postulated LBLOCA for Maanshan NPP using RELAP5/MOD3.3, DAKOTA, and SNAP codes are established in this study. The uncertainty parameters are related to ECCS property. The ranges of uncertainty analysis are based on FSAR and Training Manual of Maanshan. The maximum PCT of the uncertainty analysis is 1293 K. The PCT of FSAR is 1383.7 K. Hence, the result of the uncertainty analysis is below the FSAR data. The difference of uncertainty analysis results and FSAR data may be caused by the different calculation procedures, phenomenological modeling, and nodalization, etc.
Additionally, the PCT criteria of 10 CFR 50.46 is 1477.6 K (2200). The prediction of the uncertainty analysis is below the criteria of 10 CFR 50.46. This implies that Maanshan NPP in LBLOCA transient is at a safety situation when ECCS is performed.
Probability Distribution Function (PDF)
Feedwater Pump(s) Trip (FWPT)
Complete Loss of Flow (CLOF)
High Head Safety Injection (HHSI)
Top of Active Fuel (TAF)
Maanshan Nuclear Power Plant (PWR)
NUREG/IA-0509 LBLOCA Uncertainty Analysis of Maanshan Nuclear Power Plant with RELAP5/SNAP and DAKOTA April 2019