流程生产安全数智化监测系统传感器故障诊断研究  

Research on sensor fault diagnosis in digital intelligence monitoring system of process production safety

作　　者：张建荣[1] 张伟[1] 赵挺生[1] 苗雨[1] ZHANG Jianrong;ZHANG Wei;ZHAO Tingsheng;MIAO Yu(School of Civil and Hydraulic Engineering,Huazhong University of Science and Technology,Wuhan Hubei 430074,China)

机构地区：[1]华中科技大学土木与水利工程学院,湖北武汉430074

出　　处：《中国安全生产科学技术》2024年第4期34-41,共8页Journal of Safety Science and Technology

基　　金：国家重点研发计划项目(2021YFB3301100)。

摘　　要：为保障流程生产安全监测数据的准确性,提出1种结合核主元分析和累积残差贡献率法的故障诊断方法。首先提出“感知-汇聚-决策”的多层级数智化监控系统架构;针对感知层传感器,基于核主元分析构建故障检测模型并通过累积残差贡献率法定位故障传感器;以DYTG转炉厂连铸作业区进行实证分析。研究结果表明:该故障诊断方法在SPE指标上的平均检测率和平均误检率分别为95.28%和2.61%,在T^(2)指标上的平均检测率和平均误检率分别为84.36%和1.71%,且针对4种故障形式均能精准定位故障传感器。研究结果有助于降低监测系统的维护成本,提升流程生产安全管控水平。To ensure the accuracy of process production safety monitoring data,a fault diagnosis method that combined kernel principal component analysis and cumulative residual contribution rate method was proposed.A multi-level digital intelligence monitoring system architecture based on the“perception-aggregation-decision”paradigm was put forward.For the sensors in the perception layer,a fault detection model was constructed based on kernel principal component analysis,and the fault sensors were located by the cumulative residual contribution rate method.The continuous casting operation area in the DYTG converter plant was selected as the case analysis.The results show that the average detection rate and average false detection rate of the proposed fault diagnosis approach on the SPE index are 95.28%and 2.61%,respectively,while those on the T2 index are 84.36%and 1.71%,respectively.Furthermore,it can accurately locate the fault sensor for four kinds of fault forms.The research results are conducive to reduce the maintenance cost of the monitoring system,and improve the control level of process production safety.

关 键 词：流程生产 传感器 故障诊断 核主元分析 累积残差 

分 类 号：X924.2[环境科学与工程—安全科学]