摘要
为解决传统数据采集与监视控制(Supervisory Control and Data Acquisition,SCADA)系统在高压强磁与极端温度环境下存在的高延迟、感知盲区、推理低效问题,提出一种基于物联网通信的可靠性检测方法。在边缘部署多模态传感器,通过图卷积网络融合特征,并结合Weibull、自回归整合移动平均(Auto Regressive Integrated Moving Average,ARIMA)、长短期记忆(Long Short-Term Memory,LSTM)模型与熵权健康指数预测剩余寿命与智能告警。实验表明,该方法在准确率、响应时延、通信时延、丢包率以及预警正确率等方面均优于SCADA系统,证实其在严苛工况下具有工程可行性。
In order to solve the problems of high delay,blind area of perception and low reasoning efficiency in the traditional Supervisory Control and Data Acquisition(SCADA)system under high voltage,strong magnetic field and extreme temperature environment,a reliability detection method based on Internet of Things communication is proposed.Multi-modal sensors are deployed at the edge,and features are fused by graph convolution network,combined with Weibull,Auto Regressive Integrated Moving Average(ARIMA),Long Short-Term Memory(LSTM)model and entropy weight health index to predict the remaining life and intelligent alarm.Experiments show that this method is superior to SCADA system in accuracy,response delay,communication delay,packet loss rate and early warning accuracy,which proves its engineering feasibility under severe working conditions.
作者
曹善俊
施静林
CAO Shanjun;SHI Jinglin(Shandong Huitong Energy Group Co.,Ltd.,Yantai 264000,China)
出处
《通信电源技术》
2025年第14期72-74,共3页
Telecom Power Technology
关键词
物联网通信
电力工程自动设备
可靠性检测方法
Internet of Things communication
power engineering automation equipment
reliability detection method
