摘要
TSP问题是组合优化领域新算法设计的一个理想测试平台。当前比较成功的算法大都基于局部优化 ,局部优化速度快 ,但收敛性差 ,往往无法求出最优解。最近 ,人们在局部优化策略的基础上提出了基于群体的局部优化算法 ,即 Memetic算法 ,试图将局部优化的高效性和遗传算法的鲁棒性结合起来。文章分析了 Memetic算法的性能和 TSP问题的特性 。
TSP is an ideal testing platform for the designing of new algorithms in combinatorial optimization.Currently,most successful algorithms are based on local optimization.Local optimization is quite fast but often not able to get the global optimum.Recently local optimization based on population is proposed in order to combine the high efficiency of local optimization and the robustness of genetic algorithms.In this paper,the performance of Memetic algorithms and the property of TSP are analyzed,and a new efficient Memetic algorithm is given based on our analysis.
出处
《交通与计算机》
2002年第1期14-17,共4页
Computer and Communications
