A dyadic floating-point mutation operator of EC

The performance of evolutionary computation (EC) is determined by many parameters among which the mutation operator plays an important role especially for floating-point EC. However, the traditional mutation operation can´t effectively keep EC from trapping in local extremum. In order to improve the efficiency of EC, a novel dyadic mutation operator is presented in this paper. Then we take genetic algorithm (GA) as an example to introduce the novel mutation operator in detail. The experimental results based on function optimization show that the improved mutation operator can effectively prevent premature convergence.