文献类型：期刊文献

英文题名：A fuzzy-based design of networked control systems considering random time delays and packet dropouts in the forward communication channel

作者：Tong, Shiwen[1]; Yan, Xiaoyu[2]; Fang, Jianjun[3]

第一作者：佟世文

通讯作者：Tong, Shiwen

机构：[1] College of Robotics, Beijing Union University, Beijing, China; [2] Graduate School, Beijing Union University, Beijing, China; [3] College of Urban Rail Transit and Logistics, Beijing Union University, Beijing, China

第一机构：北京联合大学机器人学院

年份：2018

起止页码：171-189

外文期刊名：Adaptive Control: Methods, Applications and Research

收录：EI(收录号：20190906576480)

语种：英文

外文关键词：Backpropagation - Control theory - Controllers - Delay control systems - Electric control equipment - Learning systems - Networked control systems - Neural networks - Packet loss - Proportional control systems - Sliding mode control - System stability - Tanks (containers) - Three term control systems - Time delay - Water levels

摘要：The tuning of PID parameters is a key problem of PID control. The neural network technology is applied to the PID controller. Through the self-learning of neural networks and the adjustment of weight coefficients, the PID control parameters under some optimal control law can be obtained, which can improve the dynamic performance of the control system and enhance the stability of the system. In this paper, the realization method of BP neural network adaptive PID control algorithm in PLC is studied. The structure and algorithm of BP neural network adaptive PID control system are emphatically introduced. Taking S7-1200 PLC as the controller, the structure control language (SCL) is used to design the function block of BP neural network adaptive PID control algorithm. The explanation of the parameters of the function block is given. The three tank water object is a typical nonlinear and time-delay system. The neural network adaptive PID controller is applied to the three tank water level control system and the on-line PID parameter tuning is carried out by using the self-learning ability of the neural network. Experiments show that compared with the conventional PID control, the BP neural network adaptive PID control has higher accuracy and stronger adaptability and better control results can be obtained. The function block of BP neural network adaptive PID control algorithm has a certain versatility and portability. It provides reference for extending advanced control algorithm to engineering field. ? 2018 Nova Science Publishers, Inc. All rights reserved.