Research Of Adaptive Control For DC Motor Based On Ethernet

In recent years, with the fast development network technology, the network control system enters into the field of the modern industrial control, the control method has solved the limitations of location and space in the traditional control, and the motor control system begins to develop in the direction of network. However, the network time delay is inevitable. So it is inevitable cause the unstable performance or oscillation of DC motor control system.In this paper, first introduces the characteristics and problems of network control system. It also researches the impact of network time delay on the stability of the control system. At the same time, the network control system simulation model is set up with the DC motor for the controlled object, and the network time delay is researched for the performance of the motor control system in this paper.Then, aim at the impact of the network time delay, the new Smith predictor with single neuron adaptive PID controller and the new Smith predictor with fuzzy immune adaptive PID controller are designed to compensate the network time delay. Due to the network time delay exists in the forward channel and feedback channel of the network control system, for the network time delay of the forward channel, respectively using single neuron adaptive PID and fuzzy immune adaptive PID control method, the adaptive controller is to modify their own characteristics so as to adapt to the characteristics of controlled object, meeting the requirements of control system, making the system to achieve optimal control. Single neuron adaptive PID controller is based on the adjusting of weighted coefficient to realize adaptive, self-organizing function, with the constant change of control environment. The controller can meet the requirements of the network control system. Fuzzy immune adaptive PID controller possesses the nonlinear characteristics of fuzzy control and adaptability, self-study and anti-interference of immune feedback control, completes correction of PID parameters on-line, finally achieves the purpose of self-tuning, to adapt to the change in the network control system. And for network time delay of feedback channel, using new Smith predictor compensates the network time delay. The delay controlled quantity can be received by the controller in advance through using the new Smith predictor, so that the controller is able to take action early, which reduces overshoot and accelerates adjustment process. The simulation analysis shows that this control strategy has a better compensation for the impact of network time delay, which makes the dynamic performance of the control system improve.Finally, using the DC motor for the controlled object, in the Matlab/simulink environment, building simulation model of network control system based on Ethernet, by changing the network time delay, observing the system simulation diagram in different network time delay, and analyzing the simulation results, according to the result, the control strategy can effectively compensate network time delay, it is an effective method of control. At the same time, through the comparison of simulation results, it can be seen that the effect of the new Smith fuzzy immune adaptive PID control is better than new Smith single neuron adaptive PID control effect.