A .COMPARATIVE STUDY OF FUZZY AND, CONVENTIONAL PI CONTROLLERS

Abstract

Industrial processes are usually non-linear with considerable dead time and their parameters may change with changes in ambient conditions or with time. PID control is the major practical technology that is widely used in industrial to control these processes. But it has the limitation that if process parameters changes then a significant amount of effort is required to manually tune the PID controller. Instead of using conventional linear controller, if fuzzy logic is used to implement a linear controller then effect of variation of parameters of the process on the desired response can be reduced to certain level. In other words, FLC (fuzzy logic controller) can operate over a wider range of operating conditions. In this dissertation, a PI type FLC is simulated for a first order process involving delay and its performance is compared with conventional PI type controller. The proposed FLC is designed using a very simple control rule-base and most natural and unbiased membership functions (symmetrical triangles with equal bases and 50% overlap with neighbouring MFS). 4