A Combined Genetic Algorithm-Fuzzy Logic Method (GA-FL) to Design a 6 Bars Planar Mechanism

Abstract

This work deals with solution methods of optimal synthesis of planar mechanisms. A searching procedure presents a combined genetic algorithm–fuzzy logic method to solve the problem of path generation in mechanism synthesis. Previous works, dealing with the same problem and using the genetic algorithm method, suffered from the lack of precision, especially for large domain problems. The proposed method is made of a classical genetic algorithm coupled with a fuzzy logic controller (GA-FL). This controller monitors the variation of the design variables during the first run of the genetic algorithm and modifies the initial bounding intervals to restart a second round of the genetic algorithm. For both of these runs, we limited the number of generations to roughly half of the number found in the literature, without reducing the accuracy of the final solution. Compared to previous works on the same problem, our method proved to be more efficient in finding the optimal mechanism. The effectiveness of the proposed method has been demonstrated on a six bars synthesis example.