TSP问题是一类经典的NP问题,目前有很多方法对其求解,而用混合遗传算法对其求解取得了很好的成效。常见的混合遗传算法有遗传算法与最速下降法相结合(GACSDM)、遗传算法与模拟退火法相结合(SAGA)。设计了贪婪的复合变异算子(GCM),并引入隔代爬山法算子(Climb)增加遗传算法的局部搜索能力。实验结果表明该算法是有效的。
Traveling Salesman Problem(TSP) is a classical kind of nondeterministic polynomial completeness.Now these are many ways that can solve it,but a hybrid genetic algorithm is useful way for TSP.The normal hybrid genetic algorithms include genetic algorithm combined with steepest descent method and simulated annealing genetic algorithm.In the text,a greedy composite operator and climb operator are introduced for increasing the local searching ability.The experiment result demonstrates that the method is valid and effective.