A new technique with improved control quality of nonlinear systems using an optimized fuzzy logic controller

摘要

Fuzzy logic controllers are increasingly applied to control complex systems because they have several advantages. The objective of this work is to propose a new technique to optimize a Takagi-Sugeno fuzzy logic controller with quality using the Non-dominated Sorting Genetic Algorithm-II by optimizing three objectives functions which are a cost function, the number of fuzzy inference rules, and the maximum instantaneous quadratic error. In this technique, the Multi-Criteria Decision-Making approach is used to choose one of the best controllers from the Pareto set of the last generation of the genetic algorithm. The proposed technique ensure: (i) that the output of the controlled system correctly follows the desired reference. (ii) the acceleration of the control process and (iii) avoid the existence of large overshoots: which are usually observed when applying commands to complex processes with variable behavior. At the end of the control process, a robustness test is performed to verify the efficiency of the proposed technique. It is shown here that the optimization of the third objective function, allows the improvement of the control quality. This new technique can be used to improve expert and intelligent systems based on fuzzy rules to control high complex systems with variable behavior which they have disturbing overshoots during their control. This technique; allows to accelerate the calculation of the control law of expert systems based on fuzzy rules; while ensuring that the quality of the control and the output signal are good. (C) 2019 Elsevier Ltd. All rights reserved.