Research on Mutation Operator of Diploid Genetic Algorithm and its Dynamic Adaptation Strategy

Diploid genetic algorithm (DGA) is a double gene model for the genetic algorithm. This paper theoretically analyses influence of mutation operator on population diversity by introducing an average schema similar rate as the measure criteria of population diversity in DGA. A conclusion is drawn that DGA has a better performance in terms of preserving the diversity than HGA. Furthermore, a dynamic adaptation strategy is proposed to regulate the mutation operator by Mexican hat wavelet along with iterative generations. A simple optimal problem has been chosen to test and simulate on Matlab. Results show that the dynamic adaptation strategy has a better performance in terms of solution accuracy and convergence speed. The simulation results are found to be satisfactory