摘要
温度仪表是流程工业仪控系统的核心组件,仪表漂移故障发生率较高但却难以察觉。为此,文中提出基于边云协同的温度仪表漂移故障诊断方法,在边缘端部署线性回归模型进行温度仪表测量值预测;在云端部署基于stacking的集成学习模型实现精确漂移故障诊断。仅当边缘端检测出异常时触发云端集成学习诊断模型。文中以风力发电机实际运行数据对于所述方法进行验证,实验结果表明文中所述方法能够在基本不损失诊断准确性的同时大幅缩减诊断时间与计算负载。
Temperature transmitter is the key components of control system in process industry.Transmitter drift fault occurs frequently,but it is difficult to detect.Therefore,this paper proposed a temperature transmitter drift fault diagnosis method based on edge-cloud cooperation.A linear regression model working on the edge side predicted the measured value of temperature transmitter.An ensemble learning model based on stacking was built on the cloud side to diagnosis accurately.The cloud diagnosis only triggered when the edge trend prediction algorithm detected an anomaly.The method in this paper was verified by the actual operation data of wind turbine.The experimental results show that the method in this paper can greatly reduce diagnosis time and computing load of diagnosis without losing the diagnosis accuracy.
作者
何天放
王锴
徐皑冬
曾鹏
HE Tian-fang;WANG Kai;XU Ai-dong;ZENG Peng(Key Laboratory of Networked Control Systems,Chinese Academy of Sciences,Shenyang 110016,China;Shenyang Institute of Automation,Chinese Academy of Sciences,Shenyang 110016,China;Institutes for Robotics and Intelligent Manufacturing,Chinese Academy of Sciences,Shenyang 110016,China;School of Computer Science and Technology,University of Chinese Academy of Sciences,Beijing 100049,China)
出处
《仪表技术与传感器》
CSCD
北大核心
2022年第6期88-94,共7页
Instrument Technique and Sensor
基金
国家重点研发计划项目(2020YFB2009404)
国家自然科学基金项目(62073313)。
关键词
温度仪表
故障诊断
边云协同
集成学习
故障预检
temperature transmitter
fault diagnosis
edge cloud cooperation
ensemble learning
fault predetection
