Fuzzy Rule-Based Adaptive Proportional Derivative Controller

An improved fuzzy rule based auto-tuning PD controller is designed for integrating processes with dead-time. A large number of industrial processes are integrating in nature. It has been studied that for integrating systems, widely used Ziegler-Nichols (ZN) tuned PID controller (ZNPID) gives excessively large value of overshoot and settling time. Some improvement in overshoot and settling time has been possible for such systems by the development of dynamic set-point weighted PID (DSWPID) and augmented ZN tuned PID (AZNPID) controllers. Here, we propose a fuzzy auto-tuning PD (FAPD) controller where an updating factor 'alpha' continuously adjusts the derivative gain to provide an overall good performance during set point change and load disturbance. The value of alpha is updated online by 9 fuzzy if then rules defined on the value of error (e) and change of error (Delta e) of the controlled variable. To study the effectiveness of the proposed controller, FAPD is tested and compared with other PID controllers for different integrating systems with varying dead-time.