The Application of Genetic Algorithms in Designing Fuzzy Logic Controllers for Plastic Extruders

Abstract

AbstractThis paper investigates the application of Genetic Algorithms (GA) in the design and implementation of Fuzzy Logic Controllers (FLC) for temperature control in an extruder. The importance of FLC is during the process of selecting the membership functions. What is best to determine the membership functions is the first question that has be addressed. It is important therefore to select accurate membership functions but these methods possess one common weakness where conventional FLC use membership functions generated by human operators. In this situation the membership function selection process is done by trial and error and it runs step by step which is too long to arrive at a solution to the problem. This research proposes a method that may help users to determine the membership functions of FLC using GA optimization for the fastest process in solving problems. The data collection is based on simulation results and the results refer to the maximum overshoot. From the results presented, the system arrives at better and more exact results and the value of overshoot is decreased from 1.2800 for FLC without GA, to 1.0011 for FGA.KeywordsTemperatureExtruderFuzzy logicGenetic algorithmsMembership function