论文部分内容阅读
提出一种新的求解约束优化问题的进化算法.该算法在处理约束时不引入惩罚因子,使约束处理问题简单化.基于种群中个体的可行性,分别采用3种不同的交叉方式和混合变异机制用于指导算法快速搜索过程.为了求解位于边界附近的全局最优解,引入一种不可行解保存和替换机制,允许一定比例的最好不可行解进入下一代种群.标准测试问题的实验结果表明了该算法的可行性和有效性.
This paper proposes a new evolutionary algorithm for solving constrained optimization problems.The algorithm does not introduce penalties when dealing with constraints and simplifies the problem of constraint processing.Based on the feasibility of individuals in the population, three different crossover methods and mixed variants Mechanism is used to guide the fast searching process of the algorithm.In order to solve the global optimal solution located near the boundary, a mechanism of saving and replacing of infeasible solutions is introduced, allowing a certain proportion of the best infeasible solution to enter the next generation population. The results show that the algorithm is feasible and effective.