The search ability of genetic algorithm relies mainly on two aspects: the coding method and the genetic operators. So many research works are focusing on these aspects. In this paper, we propose a new decoding method named fuzzy decoding, this method makes each binary string correspond to a fuzzy field in the phenotypic space. We mainly discuss the evaluation of the fitness function about the strings. For genetic operator, we give a new selection operation based on the distance between the fitness value of each string and the best string in the search space. All these works provide a new way to solve the optimization problems, in which the search space is continuous or infinite. At last the simulations verify these methods' efficiency.
展开▼
