Comparative analysis of robustness of optimally offline tuned PID controller and Fuzzy supervised PID controller

Abstract

Proportional Integral Derivative(PID) controllers are widely used controllers. Most of the process industries make use of this controller. The reason behind it is that it is known to give reliable operation, near optimal performance and has simple structure. However it is crucial that PID controller must be tuned properly. By various methods PID can be tuned in an offline fashion but because of parameter variation and disturbances taking place in system need for online tuning of these controller parameters arise. This paper deals with the fuzzy supervised PID controller which can online tune the PID parameters. Along with this the parameters of the Fuzzy supervised PID has been optimally adjusted by making use of Ant Colony Algorithm. A comparison of the robustness of the two controllers i.e. optimally offline tuned PID and Fuzzy supervised PID has been done on the basis of their performance when a system is subjected to parameter variations. From the conclusion of these results it is found that a combination of the two controllers is giving better performance as compared to the two controllers applied alone. Finally optimally offline tuned PID and the combination of the two controllers have been implemented on the Magnetic levitation system and results obtained from it also points towards the same.