首页> 外文会议>IEEE International Conference on Artificial Intelligence and Information Systems >Research on Anomaly Location Method for Nuclear Pump Vibration Monitoring Data Based on Correlation Coefficient
【24h】

Research on Anomaly Location Method for Nuclear Pump Vibration Monitoring Data Based on Correlation Coefficient

机译:基于相关系数的核泵振动监测数据异常定位方法研究

获取原文

摘要

Nuclear power plants have accumulated a large amount of process monitoring data, but most of the data are not marked with specified patterns, which cannot be directly applied to the data-driven intelligent early warning and fault diagnosis. On-site alarm threshold can only locate a small number of abnormal vibration data, ignoring a large number of data that doesn't exceed alarm threshold but is obviously abnormal fluctuation of vibration phenomenon. To solve this problem, a method is proposed to locate abnormal vibration data based on correlation coefficient in this paper. This method takes the correlation coefficient of vibration data and corresponding time as the fluctuation index of measuring vibration data, and calculates the fluctuation threshold through historical data statistics, so as to locate abnormal vibration data. The vibration monitoring data of nuclear pump show that the proposed method can effectively detect the abnormal fluctuation of the data and locate the starting point of abnormal vibration.
机译:核电厂已经积累了大量的过程监控数据,但是大多数数据没有以指定的模式进行标记,因此无法直接应用于数据驱动的智能预警和故障诊断。现场告警阈值只能定位少量异常振动数据,而忽略大量未超过告警阈值的数据,显然是振动现象的异常波动。针对这一问题,本文提出了一种基于相关系数的异常振动数据定位方法。该方法将振动数据和相应时间的相关系数作为测量振动数据的波动指标,并通过历史数据统计计算波动阈值,以定位异常振动数据。核泵的振动监测数据表明,该方法可以有效地检测出数据的异常波动,并确定异常振动的起点。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号