DIFFERENTIAL EVOLUTION WITH DYNAMIC ADAPTATION OF PARAMETERS BASED ON A FUZZY LOGIC AUGMENTATION APPROACH
Oscar Castillo, Patricia Ochoa, José Soria · Journal of Universal Mathematics · 2019
This paper proposes an improvement to the Differential Evolution algorithm using a fuzzy logic augmentation. The main contribution is to dynamically adapt the parameters of mutation (F) and crossover (CR) using a fuzzy system, with the aim that the fuzzy system calculates the optimal parameters of the differential evolution algorithm during execution for obtaining better solutions, in this way arriving to the proposed new fuzzy differential evolution algorithm. In this paper experiments are performed with a set of mathematical functions using the original algorithm and the proposed method. Based on a statistical comparison of the original and proposed method, we can state that the fuzzy differential evolution algorithm outperforms the original differential evolution method.