Adaptive Hybrid Genetic Algorithm with Fuzzy Logic Controller

Abstract

This paper proposes adaptive genetic operators in the hybrid genetic algorithm with a fuzzy logic controller. For the hybrid genetic algorithm (HGA), a rough search technique and an iterative hill climbing technique are employed to a genetic algorithm. The rough search technique is used to initialize the population of the genetic algorithm; its strategy is to make large jumps in the search space in order to avoid being trapped in local optima and the iterative hill climbing technique is also applied to find a better solution in the convergence region within the genetic algorithm loop. A crossover operator and a mutation operator used in the HGA are automatically adjusted for the adaptive ability during the search process of the HGA, and the fuzzy logic controller (FLC) regulates the applying ratios of these operators. For the comparison of the adaptive ability in the HGA, a conventional heuristic method and the proposed FLC are analyzed and compared in a numerical example. Finally, the best hybrid genetic algorithm with an adaptive ability is recommended.