Loom Intelligent Control System Control Strategy

The main content of this thesis is intelligent control of loom system. With the development of textile industry in China, the demands of top-grade loom is becoming larger and larger, it will cost a lot of money to import top-grade loom. It requires us to develop top-grade loom, enhances rapidly the performance cost ration of loom. Due to the difficulty of acquiring mathematical model of loom system, conventional PID control method can not get a very good control result, so it needs to find other efficient control method to solve the problem.Based on the analysis of loom plant, we use the system identification method to set up the mathematical model. First, use the least square method to set up the linear model approximately. Because of the good approaching capability of neural networks to the nonlinear system, and it doesn't relay on the mathematical model, in order to reach more precisely control effect, we use BP neural networks to set up the model.Conventional PID control method has a lot of advantages: simple algorithm, good robustness, high reliability. The loom system is nonlinear and indefinite, and conventional PID method can't get a good control result. Due to the control problem in loom control, it needs to adopt intelligent control strategy, in order to adapt top-grade loom. We study expert PID control, fuzzy adaptive PID control, RBF neural network PID control, internal control based on RBF neural networks. In loom control, we use advanced PID control, get a good result, and use matlab to simulate the internal model control method based on neural networks and it gets a good result.