Fuzzy Non Linear PI Controller for High Performance

This paper presents a new technique for controlling different processes with high accuracy and fast response. The techniques is based on a fuzzy logic controller that was found to give better performance than a nonlinear PI compensator while maintaining the characteristics of a conventional PID controller. The approach uses two different control actions based on the feedback error obtained from the process. The error signal is used to determine the control output commands for the two different controller parameters (K_p and K_I). This approach gives the system flexibility as well as a fast method for performing the control calculations. Fast response is achieved based on the use of a high proportional gain at the beginning of the process which adjusts to provide a rapid convergence time when reaching the reference position. One of the main advantages of this technique is that it provides a closed form solution describing the controller actions in terms of the tuning parameters. Another advantage is that while other approaches are based on a manual technique for tuning the controller parameters, this one uses optimization techniques. Through these techniques the optimal values of the controller parameters, required to achieve fast response, can be determined while maintaining high accuracy and disturbance rejection. Simulation results of the proposed technique are presented for processes varying from single to five degree of freedom system. The same under damped linear single input single output plant as used by Shahruz and Schwartz (1997) is presented to highlight the improved performance of this new technique.