Precise position tracking control based on adaptive neuron PID algorithm for automatic clutch driven by DC motor

Abstract

For automatic clutch control system, precise and quick position tracking is difficult on account of its nonlinear character. A new clutch actuating mechanism with a whirl spring was devised for torque compensation to reduce the load of DC actuator and improve the dynamic performance of clutch control system firstly, and then a neuron adaptive PID controller with feedforward and adjustable gain was designed. The feedforward can accelerate the response of controller, and the weights of neuron adjust automatically through self-learning function so as to obtain adaptive control character, furthermore, PSD (proportion, sum and differentiation) algorithm was proposed to adjust its gain. The experiment results show that the position tracking error is smaller than that of common controller; and the controller can adapt the nonlinear change of clutch load and target engaging speed. It indicates that the controller has good robustness and can meet position tracking control command of clutch.