Fuzzy Anti-windup Compensator Design for Linear Active Disturbance Rejection Control

Aiming at the problem that it's hard for the constant coefficient compensator to effectively compensate strong nonlinear systems, a fuzzy anti-windup compensator is proposed in this paper. The error between controller's and actuator's outputs, and error derivative are used as inputs for fuzzy compensator. The error compensation coefficient is adjusted in a fuzzy manner. Applied in a permanent magnet DC motor servo system, the results show that the fuzzy compensator can reduce resources occupancy and improve dynamic characteristics significantly. The overshoot of step response reduces by 23 %, and the adjustment time (2% error band) is shortened by 0.8s than that in usual cases. Compared with the constant coefficient compensator, the fuzzy compensator performs better in transient stability and robustness, which is a feasible solution for anti-windup problem.