Research on Fault Early Warning Method of Industrial Control System Based on TCN-MKELM
Aiming at the characteristics of the temporal nature of critical signals of industrial control system,a fault early warning method of industrial control system based on time convolution network-multi kernel extreme learning machine(TCN-MKELM)is proposed.Firstly,the historical operation data of industrial control system is used,an online prediction model based on time convolution network(TCN)is established to predict each critical signal of industrial control system online and generate residual data.Secondly,multi kernel extreme learning machine(MKELM)is constructed,and a fault early warning model based on MKELM is established by combining the residual data.Finally,the operation data of boiler temperature control system of a thermal power plant is taken as an example for experiment.The experimental results show that the prediction error of the TCN-based prediction model is smaller than that of the traditional recurrent neural network;the accuracy of the MKELM fault early warning model based on residual data is higher than that of the method that adopts the original fault data for direct fault early warning.The method can effectively discover the safety hazards of industrial production control system and guarantee the safe operation of industrial control systems.
Industrial control systemIndustrial control safetyFault early warningTime convolution network(TCN)Multi kernel extreme learning machine(MKELM)Temporal data
万方数据
维普
