基于快速二层解修补策略的区间离散遗传算法

Interval Discrete Genetic Algorithms Based on Fast Two-Level Solution Repair Strategy

以等式约束下的区间离散多目标优化问题为研究对象,提出了快速二层解修补策略,其主要思想是:首先,用初始解生成器生成一个满足等式约束条件的种群,然后,将此种群中可修补个体以其违反约束度最小为目标函数,将落在未定义区间的个体修补至定义区间内,最后,在定义子区间内微调得到满足约束条件的个体,其调整方法如下:按照当前个体中的每个变量在其所处区间内的可调节上下限在此个体总的可调节上下限值内所占的比例进行调节,使得不满足等式约束的变量得到有效修补.最后,通过实验验证了本文算法的有效性.

A fast two-level solution repair strategy is proposed for interval discrete multi-objective optimization problem with equality constraints.The main idea is:first,a population satisfying equality constraints is generated by using the initial solution generator,and then,the repairable individuals in this population are repaired to definitions by taking the minimum degree of violation of constraints as the objective function.In the interval,finally,the individual satisfying the constraints can be fine-tuned in the definition sub-interval.The adjusting method is as follows:according to the proportion of the upper and lower limits of each variable in the current individual in its interval,the variable that does not satisfy the equality constraints can be effectively repaired by adjusting the proportion of the upper and lower limits of the individual in the total adjustable upper and lower limits.Finally,the effectiveness of the proposed algorithm is verified by experiments.