关键词: 演化算法, 高斯柯西变异, 多父体杂交, 自适应机制

Abstract: This paper proposes a new self-adaptive Multi-Parent Crossover evolutionary algorithm based on Gaussian and Cauchy mutation for solving constrained function optimization problems.The specialty of the algorithm include： using three novel multi-parent crossover operators which can speed up the constringency dramatically；introducing a new method based on Gaussian and Cauchy mutation for producing the new individual；introducing a self-adaptive mechanism to adjust the dimension of the search subspace as the searching range changed.This paper presents some results of numerical experiments which show the new algorithm is more universal，effective and robust than its competitors，especially the constringency and stability.

Key words: evolutionary algorithm, Gaussian and Cauchy mutations, multi-parent crossover, self-adaptive methods