Loom Warp Tension Control Strategy

The main content of this thesis is tension control strategy of loom system. At presnt, many of the domestic looms use mechanical control or conventional PID control whose effect is not ideal, leading to textile production low, poor quality. The method of reasonable tension control will reduce the wave of the tension, so that the tension which warp bears stabilizes, so avoiding the probability of starting marks and so on. Therefore, research and development of high-precision tension control system is to enhance the key to loom grade.This thesis analyses the working principle of the loom, the factors which make the warp tension fluctuate, and the characteristic of the warp tension control, and establishs mathematical model of loom with Matlab toolbox identification system. Based on PID control, several improved PID controls are further studied. In practice , it gets a good result.Conventional PID control method, which has a lot of advantage: simple algorithm, good robustness and high reliability, relays on the mathematical model. Due to the control problem in loom control, it raises the neural network control strategy, in order to achive better control effect. Taking advantage of neural network which can approach any non-linear system, a method of combining neural network with PID controller is proposed, and use matlab to simulate the adaptive PID control method based on RBF neural network with a Kaman filter and it gets a good result. The simulate result indicates that control effect and dynamic performance for adaptive PID control for warp tension system based on neural network are obviously superior to conventional PID control.