Research On The Prediction And Compensation Method Of Time Delay Of Network Control System

With the rapid development of network technology,Networked Control System(NCS),which relies on network for system control signal transmission,has been gradually applied to all aspects of industry.However,the quality of the communication network will have a direct impact on the stability of the control system,so it is necessary to conduct fault diagnosis and timely processing research on the common delay problems in the network control system.In this paper,we focus on the network control system,focusing on the delay fault,and propose a diagnostic method for the long delay fault of the network control system,and a delay compensation control method for the delay fault of the network control system.First,the purpose and significance of carrying out research on NCS fault diagnosis and compensation control are discussed,and the current research status of NCS fault diagnosis and control methods is introduced.Then,the multiple problems existing in the network control system are analyzed.The common problems such as time delay,packet loss and external interference in NCS are introduced and analyzed in detail,and the impact of time delay and packet loss on stability is studied through simulation.Then,a neural network method is used to troubleshoot the system without accurate modeling for NCS with long delays.The sparrow search algorithm(SSA)is improved for the characteristics of NCS,the BP neural network weights and thresholds are optimized by using the powerful computational power of SSA,and the SSA-BP neural network is proposed to improve the SSA-BP neural network and be used to predict the real-time delay of NCS.Finally,SSA-BP is combined with the observer fault detection method to complete two tasks of delay prediction and fault diagnosis of the network control system.Finally,the prediction accuracy of the delay and the sensitivity of fault diagnosis are verified by simulation.Finally,the delay compensation control method is proposed for the nonlinear network control system.In order to reduce the impact of delay faults on the network control system,the U-model method is applied to model the controlled object,and an improved PI-type generalized predictive control method based on the U-model is proposed in combination with the generalized predictive control algorithm and the PI controller feedback structure.In combination with the improved SSA-BP delay predictor,a network control system delay compensation controller based on SSA-BP and U-PIGPC is designed.Simulation results show that the delay compensation control proposed in this paper can effectively compensate the system delay.