鉴于船用核动力系统特殊的运行环境以及运行要求，其安全运行概念已经超出了核安全设计的框架，要加强其运行状态的安全评估。而核动力系统运行工况多变，运行状态存在多种设备的组合，以及各类设备健康状态的不确定性。本文概要介绍了当前核动力系统运行状态评估的背景，需求以及引发的下游问题，基于已经具有核动力系统的可靠性信息，运行信息，故障诊断的方法以及FMEA表，大量调研国内外技术发展以及剖析核动力系统技术发展现状与需求的基础上，提出了结合贝叶斯网络评估装置运行不确定性问题的解决方案，分析核动力系统设计、运行的安全评估方法，弥补当前系统健康管理中缺失的一环。
In view of the special requirements of marine nuclear power plant environment and operational status, which has been beyond the concept of nuclear security design framework, we should strengthen its safety assessment of operational status. Due to some problems such as load changing rapidly, various combined status of the equipment, and more uncertainty of the data and health status, therefore, this research proposes to assess the safety of nuclear power systems based on the uncertainty method. Relying on some operating experience, as well as FMEA operational data tables, expert knowledge, combined probability graph theory which is suitable for resolving uncertainty, the research aims to build the Bayesian networks of nuclear power system, based on the calculating and processing method of Bayesian networks, analyzes and predicts the security state on design condition, steady state of operation and dynamic process for Nuclear Power System, in order to fetch up the missing link of health management.
2017,39(1): 79-84 收稿日期：2016-03-29
DOI：10.3404/j.issn.1672-7619.2017.01.016
分类号：U664.15
作者简介：邰云(1981-),男,博士,高级工程师。主要从事核动力系统设计、研发工作。
参考文献：
[1] 虞镇国. 核电站安全性的探讨[J]. 能源工程, 1990:21.
[2] 王乃彦. 核电站安全性分析[J]. 院士论坛, 2008, 24(6):14
[3] 陈济东. 大亚湾核电站PSA主报告[R]. 深圳:广东核电集团, 2001.
[4] 陈玲, 蔡琦, 蔡章生. 船用核动力系统技术状态综合评估模型研究. 船海工程[J], 2006, 35(2):22-24.
[5] 陈玲, 蔡琦, 蔡章生. 基于部分失效的潜艇核动力系统技术状态评估[J]. 系统工程理论与实践, 2006, 26(9):141-144.
[6] 张永发, 童节娟, 蔡琪, 周羽. 船舶核动力装置安全目标的初步研宄[J]. 核科学与工程, 2011, 31(1):55-56.
[7] 刘永阔. 核动力装置故障诊断智能技术的研究. 哈尔滨工程大学博士学位论文[D], 2006.
[8] 刘永阔, 夏虹, 谢春丽. 智能故障诊断技术在核动力装置中的应用研究. 哈尔滨工程大学学报[J]. 2007, 28(2), 241-246.
[9] XIE Chun-li, XIA Hong, LIU Yong-kuo. Data fusion method application to fault diagnosis of nuclear power plant[J]. Journal of Marine Science and Applcation. 2005, l(4):30-33.
[10] KUN MO, SEUNGJUN LEE, POONG HYUNSEONG. A dynamic neural network aggregation model for transient diagnosis in nuclear power plants[J]. Progress in Nuclear Energy, 2007, 30(3):1-11.
[11] JINKYUN PARK, WONDEA JUNG. A study on the systematic framework to develop effective diagnosis procedures of nuclear power plants[J]. Reliability Engineering and System Safety. 2004, 84:319-335.
[12] TAE-HO WOO, UN-CHUL LEE. Safeguard assessment in nuclear power plant operartions using analytic hierarchy process(AHP)and production function[J]. Multi Science Publishing, 2011, 29(3):337-356.
[13] YANG. M. Application of a semiotic analysis method based on multilevel flow models for real-time diagnosis system of PWR plant[C]//. 2004 Annual Meeting of the Atomic Nuclear Society of Japan, 2004:368.
[14] 詹静. 基于多层流模型的可靠性分析方法在核动力装置中的应用[D]. 哈尔滨:哈尔滨工程大学. 2007.
[15] PEARL J. Fusion, propagation and structuring in belief Networks[J]. Artificial Intelligence, 1986, 29:241-288.
[16] JENSEN F V. Propagation in DAGs. Introduciton to Bayesian networks[M]. London:UCL press, Manuscript in Progress, 2001.
[17] 张晓丹. 汽车发动机故障诊断中不确定性问题的贝叶斯网络解法[D]. 吉利:东北大学, 2005.
[18] JUN SU HA, POONG YYUN SEONG. A method for risk-informed safety significance categorization using the analytic hierarchy process and Bayesian belief networks[J], Reliability Engineering and System Safety, 2004, 83:1-15.
[19] M. ROEMER, E. NWADIOGHU, G. BLOOR. Development of diagnostic and prognostic technologies for aerospace health management applications[C]//. In IEEE Aerospace Conference Proceedings, 2001, 6:193-202.
[20] PRZYTULA K W, THOMPSON D. Construction of Bayesian networks for diagnostics[C]//. Aerospace Conference Proceedings. IEEE, 2000, 5:193-200.
[21] HUSSAIN S, HOSSEIN A, GABBAR H A. Tuning of fault semantic network using Bayesian theory for probabilistic fault diagnosis in process industry[C]//. International Conference on Quality, Reliability, Risk, Maintenance, and Safety Engineering (QR2MSE), 2013:1677-1682.
[22] XIE S, PENG X, ZHONG X, et al. Fault diagnosis of the satellite power system based on the Bayesian network[C]//. Computer Science & Education (ICCSE), 8th International Conference on. IEEE, 2013:1004-1008.
[23] SCUTARI M, BROGINI A. Bayesian network structure learning with permutation tests[J]. Communications in Statistics-Theory and Methods, 2012, 41(16-17):3233-3243.
[24] HECKERMAN D, GEIGER D, CHICKERING D M. Learning Bayesian networks:The combination of knowledge and statistical data[J]. Machine Learning, 1995, 20(3):197-243.
[25] PENA J M, LOZANO J A, LARRANAGA P. An improved bayesian structural EM algorithm for learning Bayesian networks for clustering[J], Pattern Recognition Letters, 2000, 21(8):779-786.
[26] KJRASULFF U B, MADSEN A L. Bayesian Networks and Influence Diagrams:A Guide to Construction and Analysis:A Guide to Construction and Analysis[M]. Springer, 2012.
[27] 石东昱. 适合分布计算环境不确定性处理的概率图模型若干问题研究[D]. 上海:上海大学, 2006.
[28] MILONE D H, GALLI J R, CANGIANO C A, et al. Automatic recognition of ingressive sounds of cattle based on hidden Markov models[J]. Computers and Electronics in Agriculture, 2012, 87:51-55.
[29] ALKHATEEB J H, PAUPLIN 0, REN J, et al. Performance of hidden Markov model and dynamic Bayesian network classifiers on handwritten Arabic word recognition[J]. Knowledge-Based Systems, 2011, 24(5):680-688.
[30] DOIJ E, CHEN C, PECHT M. A Bayesian Hidden Markov Model-based approach for anomaly detection in electronic systems[C]//. In:Proceedings of the 34th IEEE Aerospace Conference, Big Sky MT, USA:2013, 1-10.
[31] TORABIAN ESFAHANI M, VAHIDI B. A new stochastic model of electric arc furnace based on hidden Markov model:A study of its effects on the power system[J]. IEEE Transactions on Power Delivery, 2012, 27(4):1893-1901.
[32] GEORGOULAS G, MUSTAFA M O, TSOUMAS I P, et al. Principal Component Analysis of the start-up transient and Hidden Markov Modeling for broken rotor bar fault diagnosis in asynchronous machines[J]. Expert Systems with Applications, 2013, 40(17):7024-7033.
