摘要
求解TSP问题是遗传算法应用的一个重要领域,其本质是TSP问题中巡回路径编码串的组合最优化问题。对于符号编码方式的遗传算法,通常需要设计特定的交叉算子以提高算法的运行效率和性能。该文针对自然数编码的方式,提出了一种较适合于大规模TSP问题求解的遗传交叉算子:插入交叉(InsertCrossover,简称IX)算子。该算子以优良的交叉策略,保证了算法的快速收敛和全局寻优。仿真实验结果证明,IX算子对于大规模TSP问题具有比较好的性能。
Solving Traveling Salesman Problem(TSP)is a n important problem in Genetic Algorithm's Application,it is an optimization p roblem of the TSP path encoding in essence.In order to improve the algorithm's rate,some special crossover operators are needed for the genetic algorithm of character encoding.In this paper,a new genetic crossover operator,insert cros sover operator,suited for large-scale TSP is proposed for natural number enco ding scheme.This operator can guarantee the rapid convergence and globe optimum of genetic algorithm.The simulation experimental results show that the insert crossover operator is effective for large-scale TSP.
出处
《计算机工程与应用》
CSCD
北大核心
2003年第33期67-69,共3页
Computer Engineering and Applications
基金
武汉大学测绘遥感信息工程国家重点实验室开放研究基金
