核电厂化学控制异常检测与故障诊断系统研究与实现  

Research and Implementation of Anomaly Detection and Fault Diagnosis System for Chemical Control in Nuclear Power Plants

作　　者：陈波 曹中才 姚祥英[1] 郭天宇 Chen Bo;Cao Zhongcai;Yao Xiangying;Guo Tianyu(Information Technology Research Center,Research Institute of Nuclear Power Operations,Wuhan,430223,China)

机构地区：[1]核动力运行研究所信息技术研究中心,武汉430223

出　　处：《核动力工程》2024年第S01期13-18,共6页Nuclear Power Engineering

摘　　要：目前核电厂主要依靠化学工程师进行化学控制异常检测与故障排查,受限于化学工程师个人经验和技能水平,造成异常检测与故障排查存在时效性差和准确性无法保证的问题。为解决此问题,设计了一套核电厂化学控制异常检测与故障诊断系统。在研究过程中重点探讨了核电厂化学控制异常数据检测规则库设计、故障诊断库设计以及异常检测与故障诊断等关键技术。通过在某核电厂的实际应用证明该系统能有效地自动检测化学监测数据异常并进行故障诊断。这一研究成果可显著提升核电厂的化学控制水平,为核电厂安全运行提供保障。At present,nuclear power plants mainly rely on chemical engineers for the anomaly detection and troubleshooting of chemical control,which is limited by the personal experience and skill level of chemical engineers,resulting in the problems of poor timeliness and inaccuracy in anomaly detection and troubleshooting.To address this issue,a set of anomaly detection and fault diagnosis system for chemical control in nuclear power plant is designed and implemented.During the research process,we focused on exploring key technologies such as anomaly detection rule databse,fault diagnosis database,and anomaly detection and fault diagnosis.Through the practical application in a nuclear power plant,it is proved that the system can effectively and automatically detect abnormalities in chemical monitoring data and diagnose abnormal faults,thereby improving the chemical control level of nuclear power plants and providing guarantees for the safe operation of nuclear power plants.

关 键 词：核电厂 专家知识库 化学控制 异常检测 故障诊断 

分 类 号：TL48[核科学技术—核技术及应用] TP319[自动化与计算机技术—计算机软件与理论]