论文部分内容阅读
针对在处理约束优化问题时约束条件难以处理的问题,提出了一种求解约束优化问题的改进差分进化算法.即在每代进化前将群体分为可行个体和不可行个体两类,对不可行个体,用差量法将其逐个转化为可行个体,并保持种群规模不变,经过一序列的进化后,计算所有可行个体的适应度并找到问题的最优解.对5个经典函数进行了优化测试,测试结果表明提出的算法对求解约束优化问题是有效的.
Aiming at the problem that the constraint conditions are difficult to deal with when dealing with constrained optimization problems, an improved differential evolution algorithm is proposed to solve the constrained optimization problem, that is, before each generation, the group is divided into feasible and infeasible individuals, Individuals are transformed into feasible individuals one by one by the difference method and the population size is kept unchanged. After a series of evolutions, the fitness of all feasible individuals is calculated and the optimal solution of the problem is found. Five classical functions The optimization tests show that the proposed algorithm is effective in solving constrained optimization problems.