Fuzzy PID Controller Parameters Optimized by Genetic Algorithms

Abstract

Real systems have in general significant characteristics such as high-order, nonlinearities, deadtime, etc. and they can be affected by noise, load disturbances and other environment conditions that cause parameter variations and sudden modifications of the model structure. One of the main difficulties in the tuning of the PID parameters is to address at the same time different control specifications. In particular, achieving a high load disturbance rejection performance generally results in an aggressive tuning that provides a too oscillatory set-point step response. A fuzzy inference system is adopted to determine the value of the weight that multiplies the set-point for the proportional action. This contribution introduces genetic algorithms to change the shapes of the membership of the fuzzy controller by changing their basic parameters.

Neuro-Fuzzy Modified Smith predictor for IPDT and FOPDT Processes Control

“…However, the disadvantage of this proposed controller is that it is very complex and difficult to tune. Fuzzy logic attracted also special attention in recent research works [14][15][16][17][18][19] . Most of these applications have concentrated on achieving the desired system performance from the human operators' experience.…”