Evaluation of Interinfluence between Adjacent Units in seismic PRA

Transcription

1 22410 Evaluation of Interinfluence between Adjacent Units in seismic PRA Shuhei Matsunaka Hiroshi Abe Manabu Watanabe Yoshihiro Oyama (Tepco Systems Corporation) (Tepco Systems Corporation) (Tokyo Electric Power Company Holdings) (Tokyo Electric Power Company Holdings) PSA2017 September 24-28, 2017, Pittsburgh, PA, USA

2 1. Objectives 1 Multi-Unit PRA (MUPRA) is an integrated safety assessment of multi-unit sites. In order to express the total risk of multi units, it is necessary to consider multiple units and multiple hazards. Focused on radiation influence in this research, though several problems peculiar to multi-unit evaluation exist. Incorporated radiation influence from adjacent units into PRA model and confirmed sensitivity to risk. Ref. Sakura Jade House (

3 2. Outline of the Evaluation 2 Although it may be considered to apply dynamic PRA method to implement detailed multi-unit PRA, there are still problems such as lack of maturity of method and computational complexity. In the model of this research, in order to evaluate the sensitivity of the focused challenge simply, the existing single unit PRA models structure have been updated by static PRA method. Evaluate based on the actual plant.

4 2. Outline of the Evaluation 3 Ref.

5 2. Outline of the Evaluation 4 Generation capacity : 8,212 MWe (The world s largest NPS), All units have been in plant outage since Mar.26,2012. Units 1 to 5 : BWR5 (1,100 MWe), Units 6, 7 : ABWR (1,356 MWe) S W E N Unit 7 Unit 6 Unit 5

6 2. Outline of the Evaluation 5 Scope and Assumptions (1/2) Set Unit 6 and Unit 7 to Target Plants. Particularly when computing three or more units, the calculation becomes complicated, so it is limited to 2 units. An initiating event (hazard) is the earthquake Because an earthquake is a typical initiating event with a strong common factor to multi-units.

7 2. Outline of the Evaluation 6 Scope and Assumptions (2/2) Assuming that the accident scenario of one plant begins after the accident scenario of the other plant is finished. Even with this method, the sensitivity to the result can be grasped roughly. Radiation influence are evaluated based on deterministic evaluation. (described later) Perfect correlations are assumed for intra-units, and Zero correlations are assumed for inter-units.

8 2. Outline of the Evaluation 7 PRA taking into consideration the radiation influence and PRA not taking into consideration it were carried out, and the sensitivity was confirmed. Case 1: Case considering radiation influence Unit 6 ET(including CET) Scenario without radiation influence Scenario with radiation influence Unit 7 Event Tree ー Credit was taken for operator action (normal ET) Event Tree ー Consider radiation influence on the operability of operator action Case 2: Case not considering radiation influence Unit 6 ET(including CET) Scenario without radiation influence Unit 7 Event Tree ー Credit was taken for operator action (normal ET)

9 3. Accident Sequence Analysis 8 Unit 6 ET(including CET) Scenario without radiation influence Scenario with radiation influence Unit 7 Event Tree ー Credit was taken for operator action (normal ET) Event Tree ー Consider radiation influence on the operability of operator action

10 3. Accident Sequence Analysis 9 Hierarchy Event Tree Seismic Bldg/Str uc. (R/B) PCV RPV PCV BYPASS E-LOCA I&C DC Pipings between the Bldgs / R/B(part) RCW RSW AC (D/G) RHR LOCA FDW/MS BYPASS Offsite Power Sequence Transient* LOP* LOCA1* LOCA2* SBO* LUHSSBO1* LUHSSBO2* TBD CI LOCA BYPASS PCVR RBR *: Transfer

11 4. Radiation Influence 10 Unit 6 ET(including CET) Scenario without radiation influence Scenario with radiation influence Unit 7 Event Tree ー Credit was taken for operator action (normal ET) Event Tree ー Consider radiation influence on the operability of operator action

12 4. Radiation Influence 11 No. Task Contents Simplified Method (applied in this study) i) Setting of release category ii) Release condition setting for release category Of the scenarios modeled in the PRA, the scenarios that may have adverse effects on operation by radioactive substances are extracted and classified into categories based on release timing etc. Set release condition (route, amount, start time, duration, etc.) for each release category. Use deterministic radiation influence assessment that has already been implemented. For each evaluation pattern that has been assessed, whether or not to be operable in the adjacent plant is set. iii) iv) Extraction of operations that may be affected by radiation Radiation influence assessment Extract human error modeled in single reactor PRA and identify work area. Quantitative evaluation on radiation influence or qualitative evaluation based on existing evaluations. v) Confirmation of operation feasibility Considering the maneuverability expected of the single reactor PRA in consideration of the operation timing and operation time for the human operation extracted in iii) vi) PRA modelling Detailed modelling of time series. It is modeled that Unit 6 as a victimizer side, Unit 7 as a victim side. Depending on the release timing, there are operations that can be credited, but not credited in the model.

13 4. Radiation Influence 12 No. Release patterns Core Release Route RPV PRA PDS Deterministic Assessment Results 1 No damage Anticipatory venting No damage L1_OK (anticipatory venting) No influence with operation. 2 3 Design leakage (PCV: no damaged) No damage Damaged OKV OKP Evaluate the case of the core damage scenario following Large LOCA + SBO + all loss of ECCS, furthermore the venting in one unit. (and the design leakage for the other unit.) It can be judged that operation is possible due to several msv (integration value). 4 5 Damaged Venting after a core damage No damage Damaged VV VP The same as the above pattern. Note, however, that the scenario is the venting after 38 hours following the accident. 6 7 PCV overpressure failure (S/C) No damage Damaged FOPSP 8 9 PCV overpressure failure (D/W) No damage Damaged FOPDW Inoperable is assumed PCV hatch (PCV: isolation failure) No damage Damaged PBYP

14 4. Radiation Influence 13 No. 1 PRA PDS L1_OK (anticipatory venting) Deterministic Assessment Results No influence with operation. Availability of Operation PRA model Comments No radiation influence on both the MCR and the Remote. 2 3 OKV OKP Evaluate the case of the core damage scenario following Large LOCA + SBO + all loss of ECCS, furthermore the venting in one unit. (and the design leakage for the other unit.) It can be judged that operation is possible due to several msv (integration value). No radiation influence on both the MCR and the Remote. 4 5 VV VP The same as the above pattern. Note, however, that the scenario is the venting after 38 hours following the accident. Set inoperable conservatively, since scenarios with faster progress are included in the scenario modeled in PRA. 6 7 FOPSP 8 9 FOPDW Inoperable is assumed. Inoperable in both the MCR and the Remote PBYP

15 5. Results 14 Case 1: Case considering radiation influence Case 2: Case not considering radiation influence CDF 1 2.1E-6 CDF 1 3.9E-6 CDF 2 5.2E-6 CDF 2 3.4E-6 CDF TOTAL 7.3E-6 CDF TOTAL 7.3E-6 [/ry] [/ry]

16 5. Results 15 Frequency(/y) 1.0E E E E E E E E E E E E Acceleration(gal) Case1 CDF2[/y] Case2 CDF2[/y]

17 6. Summary and Conclusions 16 The sensitivity to the result was confirmed to be limited. The reason for this is that Regardless of the presence or absence of radiation influence, if one plant has reached a situation where radiation influence are exerted on adjacent unit (containment vessel breakage etc.), the possibility that adjacent unit is similarly in a severe situation is high. This tendency is also considered to be the same for Tsunami PRA.(depending on the circumstance.)

18 Pittsburgh View from Mt. Washington (25/09/2017)