一种新的约束优化遗传算法及其工程应用

摘要

提出一种新的用于求解约束优化问题的遗传算法,该算法利用佳点集方法初始化个体以维持种群的多样性.在进化过程中,通过可行解与不可行解算术交叉对问题的决策空间进行搜索；对可行种群与不可行种群分别采用高斯变异和柯西变异,从而协调算法的勘探和开采能力.几个标准测试问题的实验结果表明该算法的有效性；应用新算法求解两个工程优化设计问题,结果表明该算法的可行性.%This paper proposed a novel genetic algorithm to solve constrained optimization problems. It introduced the individual generation based on good-point-set method into the genetic algorithm initial step, which maintained the population diversity of the genetic algorithm. In the evolution process, it searched the decision space of a problem through the arithmetic crossover operator of feasible and infeasible solutions. In order to coordinate the exploitation and the exploration ability of the algorithm, it used Gaussian and Cauchy mutation operators to the feasible and infeasible subpopulation respectively. It tested several benchmark problems and two engineering design problems. The results show that the proposed method is an effective way for constrained optimization problems.