量化核三廠功率提升後中破口冷卻水流失事故之安全餘裕與爐心熔損頻率

Effect of Power Uprate on Safety Margin and Core Damage Frequency in a MBLOCA

陳宇民(Yu-Min Chen)

指導教授 ： 李敏 ； 陳詩奎

國立清華大學/原子科學院/核子工程與科學研究所/博士(2018年)

摘要

本論文研究參考Nuclear Energy Agency所屬的Committee on the Safety of Nuclear Installations (CSNI) 領導之Safety Margin Action Plan (SMAP) 以及Safety Margin Application and Assessment (SM2A)計畫，所發展的安全餘裕縮減 (Safety Margin Reduction，SMR) 與爐心熔損頻率（Core Damage Frequency，CDF）量化的方法論;此方法論通稱為Risk Informed Safety Margin Characterization (RISMC);量化台灣電力公司核能三廠功率提升對SMR與CDF的衝擊，期望能夠減少過去傳統分析方法加諸的保守性，更準確的量化SMR與CDF。台灣電力公司核能三廠採用美國西屋公司設計的三迴路壓水式反應器。分析事故為中破口冷卻水流失事故（Medium Break Loss Of Coolant Accident，MBLOCA）。本研究利用 RELAP5-3D/K 程式量化功率提升前後之安全餘裕的差異，將程式輸入參數的不準度納入考量後，估算事故發生後之最高燃料棒護套溫度 (Peak Cladding Temperature, PCT) 的機率分布函數 (Probability Density Function, PDF)，量化功率提升所造成的安全餘裕縮減。
功率提升亦會改變爐心熔損頻率，結合安全度評估方法 (Probabilistic Safety Assessment, PSA)，SM2A提供了量化CDF改變的方法論。本研究找出中破口冷卻水流失事故肇始事件中，對於CDF影響最大的人為誤失事件，設定人為動作執行時間的機率分佈，結合對REALP5-3D/K 結果有重大影響的參數的機率分布，以蒙地卡羅取樣，進行多次程式計算，依事件樹頂端事件的成功準則決定該人為誤失事件發生的機率，並重新計算爐心熔損頻率。估算結果顯示，功率提升後安全餘裕縮減了4.2%，MBLOCA肇始事件的爐心熔損頻率提高16%。

關鍵字

安全度評估 ； 人因失效機率 ； 不準度分析 ； 功率提升 ； 安全餘裕 ； 爐心熔損頻率

並列摘要

Nuclear Energy Agency of OECD has developed a methodology in the Safety Margin Action Plan (SMAP) and in the Safety Margin Application and Assessment (SM2A) to quantify the reduction of safety margin and the increase of core damage frequency (CDF) due to power uprate. The methodology combines the probabilistic and deterministic approaches of safety analyses. In the present study, the SMAP methodology is adopted to quantify the reduction of safety margin of a 5% power uprate of Maanshan nuclear power plant (NPP) of Taiwan Power Company. The sequence selected to demonstrate the methodology is medium break cold leg loss of coolant accident. An evaluation model code, RELAP5-3D/K is used in the LOCA analysis. The plant specific probabilistic safety assessment (PSA) model is used to determine the accident scenario analyzed. Uncertainty of the predicted peak cladding temperature (PCT) during the accident is quantified using the rank statistics. Phenomenon Identification and Ranking Table (PIRT) for parameters that are important for the medium break LOCA analyses and their distribution are identified. Two sets of calculations are performed to determine PCT of the 95th percentile with 95% of confidence at 102% and 105% power cases. When the best estimated value is determined non-parametrically, the safety margin before SPU is 10.17 and after SPU is 10.42. The margin increases slightly after SPU. From these results, it can be concluded that the impact of SPU on the safety margin of PCT is very insignificant. For CDF, there were two operator actions in the accident mitigation for the sequence selected. These actions were the emergency cooldown and depressurization (CND) of RCS and the lineup the high-head safety recirculation (HHSR) when RWST was empty. A new approach was proposed to quantify the failure probability of the operator actions. The time of operator to execute the action was included in the uncertainty analyses of code simulations. The branch probability of the event tree heading involved the operation action was determined based on the success criteria of the heading. The simulation results showed that the CND failure probability increased by 23% after the power uprate. The sequence CDF increased by 16% after the power uprate.

並列關鍵字

Probabilistic safety assessment ； Human error probability ； Uncertainty analysis ； Power uprate ； Safety margin ； Core damage frequency

參考文獻

[1] “Safety Margins Action Plan -– Final Report”, NEA/CSNI/R(2007)9.

[2] “Safety Margin Evaluation – SMAP Framework Assessment and Application”, NEA/ CSNI/R, (2011)3.

[3] https://www.nrc.gov/about-nrc/regulatory/risk-informed/history.html

[4] D.R. Bradley, “Overview of PSA Development and Risk-Informed Application in the U.S.”, Level 2 PSA and SAMG Workshop Beijing, China, 2016.

[5] S.K. Chen, “The History Perspective of PRA”, PRA II Course Lecture, NTHU NES, 2017.