摘要
核电厂应急柴油发电机组监测数据是判断柴油发电机组运行状态的重要依据,其异常检测是增强应急柴油发电机组可靠性和安全运行的保证。因监测数据量大、参数间关系复杂等特点,导致监测数据异常检测很难实现高检测率和低误检率。本文讨论监测数据异常检测的方法及应用现状,对数据驱动的监测数据异常检测方法进行了详细分析,并在此基础上归纳发展方向,为相关内容研究提供参考。
The monitoring data of emergency diesel generator(EDG)at nuclear power plant data is the important basis for estimating EDG’s state,and the anomaly detection of which has become the important mean to enhance the EDG’s reliability and safety.The monitoring data has the characteristics of large amount of data and complex relation.The anomaly detection is hard to realize the high detection rate and low false detection rate.This paper discusses the methods and application status of monitoring data anomaly detection,and makes a detailed analysis of data-driven monitoring data anomaly detection methods.On this basis,the development direction is summarized,which provides a reference for related research.
作者
周勇
朱鹏树
陈星
马波
杜晓光
Zhou Yong;Zhu Pengshu;Chen Xing;Ma Bo;Du Xiaoguang(Daya Bay Nuclear Power Operations And Management Co.,Ltd.,Guangdong,Shenzhen,518031,China;Beijing University of Chemical Technology,Beijing,100029,China;Beijing Division,China Nuclear Power Technology Research Institute Co.,Ltd.,Beijing,100086,China)
出处
《仪器仪表用户》
2020年第10期52-55,78,共5页
Instrumentation
作者简介
周勇(1982-),男,湖北人,本科,高级工程师,从事核电应急柴油发电机组设备管理工作;通讯作者:杜晓光(1982-),男,河南人,硕士研究生,高级工程师,E-mail:15110142735@163.com。
