几种局部优化算子在求解TSP中的性能比较

Several local optimization operators in TSP performance comparison

将局部优化算子引入遗传算法求解TSP问题,以求提高算法的性能。具体措施是在标准遗传算法的最后阶段增加一步,即对每代的最优个体进行一定次数的局部搜索,以求改善该最优个体。首先提出将反序-杂交法引入局部优化过程中。同几种常用的局部优化方法相比,反序-杂交法的性能最为突出。实验结果表明,该优化方法能有效求解300个城市以内的TSP问题。

Local optimization operator to introduce genetic algorithm for TSP in order to improve the algorithm＇s performance. In the final step of SGA, every generation of the best individual for a certain number of local search, in order to improve the best individual. First inver-cross method introduced into local optimization process is proposed. Several common with the local optimization method, the inver-crossay method act ofthe most outstanding performance. Experiments show that the new method can effectively solve 300 cities within the TSP.