基于动态贝叶斯网络的乏燃料后处理核应急情景分析

doi:10.16265/j.cnki.issn1003-3033.2024.10.1738

摘要/Abstract

摘要：

为提升乏燃料后处理核事故应急准备与响应能力,针对核事故应急演变路径的不确定性、情景分析在突发事件应急决策的重要性、演变过程的复杂性以及组织实施难等问题,构建基于知识元理论的乏燃料后处理核事故应急情景。选取事件、致灾体、承灾体和应急响应为组成要素,建立基于动态贝叶斯网络(DBN)方法的乏燃料后处理核应急动态情景推演分析模型,计算关键情景的发生概率、推演情景的发展趋势、分析演化规律与路径。以马雅克乏燃料后处理厂的高放废液贮槽爆炸为例,开展基于知识元和DBN的乏燃料后处理核事故应急情景分析方法的过程推演,并分析结果。结果表明: 应急冷却供水丧失发生的概率为73%,高放废液贮槽爆炸发生的概率为86%,放射性核素通过多种途径转移到动植物产品和饮用水发生的概率为87%,部分地区有长寿命放射性核素沉积发生的概率为89%,事件平息与消亡发生的概率为72%;事故污染空气、土壤、河流的发生概率分别为89%、85%、81%,对公众健康和安全有影响的发生概率为86%。情景演化过程与后处理贮槽爆炸事故应急发展,以及对公众、环境的影响一致,验证了该模型的可行性和有效性。

关键词: 动态贝叶斯网络(DBN), 乏燃料后处理, 核事故应急, 情景推演, 知识元

Abstract:

To improve the emergency preparedness and response capabilities of spent fuel reprocessing nuclear accidents, a spent fuel reprocessing nuclear accident emergency scenario based on knowledge meta was proposed to address the uncertainty of the nuclear accident emergency evolution process, the importance of scenario analysis in emergency response decision-making, the complexity of the evolution process, and the difficulty of organization and implementation. The disaster event, causative agent, causal agent, and emergency response were determined, and then a dynamic scenario model for spent fuel reprocessing nuclear emergencies based on a DBN was developed to calculate the occurrence probability of key scenarios, evaluate the development trend of the scenarios, and analyze the evolution laws and paths. Taking the explosion of the high-release liquid storage tank of the Mayak spent fuel reprocessing plant as an example, the process of deduction of the scenario analysis method of the spent fuel reprocessing nuclear accident based on the knowledge meta theory and DBN was performed, and the results were further analyzed. The results showed that: the loss probability of emergency cooling water supply was 73%, the probability of explosion of the high-release waste liquid storage tank was 86%, the probability of radioactive nuclides transferred to animal and plant products and drinking water through multiple pathways was 87%, the probability of long-lived radioactive nuclides deposition in part of the area was 89%, and the probability of the event calming down and dying out was 72%. The probability of the accident contaminating the air, soil, and river was 89%, 85%, and 81%, respectively. The probability of affecting public health and safety was 86%. The scenario evolution process is consistent with the emergency response development of the reprocessing storage tank explosion accident and its impact on the public and the environment, validating the model performance.

Key words: dynamic Bayesian network(DBN), spent fuel reprocessing, nuclear emergency response, scenario projection, knowledge element

中图分类号:

X945

孙美兰, 邹树梁, 徐守龙, 陈甲华. 基于动态贝叶斯网络的乏燃料后处理核应急情景分析[J]. 中国安全科学学报, 2024, 34(10): 214-220.

SUN Meilan, ZOU Shuliang, XU Shoulong, CHEN Jiahua. Nuclear emergency scenario analysis for spent fuel reprocessing based on dynamic Bayesian network[J]. China Safety Science Journal, 2024, 34(10): 214-220.

图/表 6

图1

图2

表1

图3

表2

图4

参考文献 22

[1]	姜卉, 黄钧. 罕见重大突发事件应急实时决策中的情景演变[J]. 华中科技大学学报:社会科学版, 2009, 23(1):104-108.
	JIANG Hui, HUANG Jun. The study on the issues of scenario evolvement in real-time decision making of in frequent fatal emergencies[J]. Journal of Huazhong University of Science and Technology:Social Science Edition, 2009, 23(1):104-108.
[2]	巴锐, 张宇栋, 刘奕, 等. 城市复杂灾害“三层四域”情景分析方法及应用[J]. 清华大学学报:自然科学版, 2022, 62(10):1579-1590.
	BA Rui, ZHANG Yudong, LIU Yi, et al. "Three-layer and four-domain" scenario analysis method and its application to urban complex disasters[J]. Journal of Tsinghua University:Science and Technology, 2022, 62(10):1579-1590.
[3]	CHEN Chen, YANG Yabin, WANG Mengtong, et al. Characterization and evolution of emergency scenarios using hybrid Petri net[J]. Process Safety and Environmental Protection, 2018, 114: 133-142.
[4]	岳文静, 杜丽敬, 陈先锋, 等. 基于DBN的气体泄漏事故情景推演与节点分析[J]. 中国安全科学学报, 2022, 32(增1):165-170.
	YUE Wenjing, DU Lijing, CHEN Xianfeng, et al. Scenario deduction and node analysis of gas leakage accidents based on DBN[J]. China Safety Science Journal, 2022, 32(S1):165-170. doi: 10.16265/j.cnki.issn1003-3033.2022.S1.0893
[5]	鲁金涛. 应急演练“情景-响应”模型的结构相似度构建方法[J]. 中国安全科学学报, 2021, 31(10):182-188. doi: 10.16265/j.cnki.issn1003-3033.2021.10.025
	LU Jintao. Structural similarity construction method of “scenario-response” model for emergency drills[J]. China Safety Science Journal, 2021, 31(10) :182-188. doi: 10.16265/j.cnki.issn1003-3033.2021.10.025
[6]	WU Jiansong, XU Shengdi, ZHOU Rui, et al. Scenario analysis of mine water inrush hazard using Bayesian networks[J]. Safety Science, 2016, 89: 231-239.
[7]	AFENYO M, KHAN F, VEITCH B, et al. Arctic shipping accident scenario analysis using Bayesian network approach[J]. Ocean Engineering, 2017, 133: 224-230.
[8]	袁晓芳, 田水承, 王莉. 基于PSR与贝叶斯网络的非常规突发事件情景分析[J]. 中国安全科学学报, 2011, 21(1):169-176.
	YUAN Xiaofang, TIAN Shuicheng, WANG Li. Scenario analysis of unconventional emergency based on PSR model and Bayesian networks[J]. China Safety Science Journal, 2011, 21(1):169-176.
[9]	索传军, 盖双双. 知识元的内涵、结构与描述模型研究[J]. 中国图书馆学报, 2018, 44(4):54-72.
	SUO Chuanjun, GAI Shuangshuang. The connotation, structure and description model of knowledge unit[J]. Journal of Library Science in China, 2018, 44(4):54-72.
[10]	王喆, 孔维磊, 方丹辉, 等. 基于贝叶斯网络的城镇洪涝应急情景推演研究[J]. 中国安全科学学报, 2021, 31(6):182-188. doi: 10.16265/j.cnki.issn 1003-3033.2021.06.024
	WANG Zhe, KONG Weilei, FANG Danhui, et al. Research on urban flood and waterlog emergency scenario deduction based on Bayesian network[J]. China Safety Science Journal, 2021, 31(6):182-188. doi: 10.16265/j.cnki.issn 1003-3033.2021.06.024
[11]	张磊, 王延章, 陈雪龙. 基于知识元的非常规突发事件情景模糊推演方法[J]. 系统工程学报, 2016, 31(6):729-738.
	ZHANG Lei, WANG Yanzhang, CHEN Xuelong. Fuzzy inference method for unconventional events scenarios based on knowledge unit[J]. Journal of Systems Engineering, 2016, 31(6):729-738.
[12]	曲毅, 仲秋雁, 马骁霏, 等. 基于情景的突发事件演变概率规则推理方法[J]. 系统工程学报, 2014, 29(4):571-578.
	QU Yi, ZHONG Qiuyan, MA Xiaofei, et al. The event evolution scenario-based reasoning method via probability rules[J]. Journal of Systems Engineering, 2014, 29(4):571-578.
[13]	李健行, 夏登友, 武旭鹏. 基于知识元与动态贝叶斯网络的非常规突发灾害事故情景分析[J]. 安全与环境学报, 2014, 14(4):165-170.
	LI Jianhang, XIA Dengyou, WU Xupeng. Scenario analysis of unconventional emergency disaster accidents based on the knowledge elements and dynamic Bayesian networks[J]. Journal of Safety and Environment, 2014, 14(4):165-170.
[14]	夏登友, 李丞曜, 朱毅, 等. 基于情景元的案例推理及其在应急决策中的应用[J]. 安全与环境学报, 2020, 20(3):1028-1033.
	XIA Dengyou, LI Chengyao, ZHU Yi, et al. Case-based reasoning method based on the scenario element and its application in the emergency decision-making[J]. Journal of Safety and Environment, 2020, 20(3):1028-1033.
[15]	王延章. 模型管理的知识及其表示方法[J]. 系统工程学报, 2011, 26(6):850-856.
	WANG Yanzhang. Knowledge and representation of model management[J]. Journal of Systems Engineering, 2011, 26(6):850-856.
[16]	张志霞, 郝纹慧. 基于知识元的突发灾害事故动态情景模型[J]. 油气储运, 2019, 38(9):980-987,995.
	ZHANG Zhixia, HAO Wenhui. A dynamic scenario model based on knowledge element for sudden disasters and accidents[J]. Oil & Gas Storage and Transportation, 2019, 38(9):980-987,995.
[17]	张承伟, 李建伟, 陈雪龙. 基于知识元的突发事件情景建模[J]. 情报杂志, 2012, 31(7):11-15,43.
	ZHANG Chengwei, LI Jianwei, CHEN Xuelong. Emergency scenario model based on knowledge element[J]. Journal of Intelligence, 2012, 31(7):11-15,43.
[18]	宋英华, 刘含笑, 蒋新宇, 等. 基于知识元与贝叶斯网络的食品安全事故情景推演研究[J]. 情报学报, 2018, 37(7):712-720.
	SONG Yinghua, LIU Hanxiao, Jiang Xinyu, et al. Research on scenario evolution of food safety incidents based on knowledge element and Bayesian network[J]. Journal of Intelligence, 2018, 37(7):712-720.
[19]	盛勇, 孙庆云, 王永明. 突发事件情景演化及关键要素提取方法[J]. 中国安全生产科学技术, 2015, 11(1):17-21.
	SHENG Yong, SUN Qingyun, WANG Yongming. Emergency scenario evolution and extraction method of key element[J]. Journal of Safety Science and Technology, 2015, 11(1):17-21.
[20]	邹树梁, 黄斌海, 李书帆. 海上浮动核电站人员认知可靠性模型[J]. 中国安全科学学报, 2020, 30(1):1-6. doi: 10.16265/j.cnki.issn1003-3033.2020.01.001
	ZOU Shuliang, HUANG Binhai, LI Shufan. Human cognitive reliability model of marine floating nuclear power plants[J]. China Safety Science Journal, 2020, 30(1):1-6. doi: 10.16265/j.cnki.issn1003-3033.2020.01.001
[21]	吕丹, 杨晓伟, 汪世军, 等. 国外乏燃料后处理设施爆炸事故的经验教训[J]. 科技导报, 2020, 38(7):118-124. doi: 10.3981/j.issn.1000-7857.2020.07.015
	LYU Dan, YANG Xiaowei, WANG Shijun, et al. Experience and lessons learned from explosion accidents in foreign spent fuel reprocessing facilities[J]. Science & Technology Review, 2020, 38(7):118-124.
[22]	罗上庚. 美国高放废液贮罐整治的经验与教训[J]. 辐射防护, 2010, 30(2):122-130.
	LUO Shanggeng. Experiences and lessens from treatment of HLW-tank in USA[J]. Radiation Protection, 2010, 30(2):122-130.

情景阶段	情景要素表示
发生情景	事件E:冷却与应急冷却供水丧失E₀; 致灾体D:冷却系统失控D₀; 承灾体H:贮槽H₁; 应急响应R:将设备室集水坑的溶液转送到适当的容器中R₀;去污设备室地面以及设备室集水坑R₁;更换损坏的部件R₂
发展情景	事件E:高放废液贮槽爆炸E₁; 致灾体D:温度升高,贮槽内物质蒸干D₁; 承灾体H:设备室H₂;建(构)筑物H₃; 应急响应R:辐射环境监测与后果评价R₃;非应急工作人员与公众撤离R₄;人员去污R₅
蔓延情景	事件E:放射性核素通过多种途径转移到动植物产品和饮用水E₂; 致灾体D:放射性核素D₂; 承灾体H:空气H₄;土壤H₅;河流H₆ 应急响应R:食物和饮水控制R₆;公众沟通R₇
消退情景	事件E:部分地区有长寿命放射性核素沉积E₃; 承灾体H:公众健康与安全H₇; 应急响应R:地区去污R₈;临时避迁或永久再定居R₉
消失情景	事件E:事件平息与消亡E₄

节点	节点变量概率			先验概率
	条件	概率
	条件	T	F
E₀	D₀=G, R₀=T, R₁=T, R₂=T	0.70	0.30	P(D₀=G)=0.95 P(D₀=N)=0.05 P(R₀=T)=0.75 P(R₀=F)=0.25 P(R₁=T)=0.70 P(R₁=F)=0.30 P(R₂=T)=0.90 P(R₂=F)=0.10
	D₀= G, R₀=T, R₁=T, R₂=F	0.85	0.15
	D₀= G, R₀=T, R₁= F, R₂=T	0.85	0.15
	D₀=G, R₀=T, R₁= F, R₂= F	0.90	0.10
	D₀= G, R₀= F, R₁= T, R₂=T	0.80	0.20
	D₀= G, R₀= F, R₁= T, R₂= F	0.90	0.10
	D₀= G, R₀= F, R₁= F, R₂=T	0.75	0.25
	D₀= G, R₀= F, R₁= F, R₂= F	0.95	0.05
	D₀=N, R₀=T, R₁=T, R₂=T	0.05	0.95
	D₀= N, R₀=T, R₁=T, R₂=F	0.10	0.90
	D₀=N, R₀=T, R₁= F, R₂=T	0.05	0.95
	D₀ =N, R₀=T, R₁= F, R₂= F	0.10	0.90
	D₀=N, R₀= F, R₁= T, R₂=T	0.05	0.95
	D₀=N, R₀= F, R₁= T, R₂= F	0.10	0.90
	D₀=N, R₀= F, R₁= F, R₂=T	0.05	0.95
	D₀=N, R₀= F, R₁= F, R₂= F	0.10	0.90