摘要
针对Hopfield网络求解TSP问题时出现无效解和收敛性能差的问题,对约束条件能量函数进行改进,构造了一种求解TSP问题的遗传Hopfield神经网络算法,并与经典Hopfield神经网络求解TSP方法进行对比。实验结果表明,本文算法具有更好的整体求解性能。
For the Hopfield network in solving traveling salesman problem often getting invalid and not optimal solution, an improved constrained optimization energy function is used as fitness function of the genetic algorithm. A solving traveling salesman problem algorithm based on the genetic Hopfield network is constructed. Compared with traditional Hopfield network algorithm, the solving algorithm in this paper can easy obtain effective global optimal solution is proved by simulation experiment results.
出处
《微型机与应用》
2009年第21期7-9,15,共4页
Microcomputer & Its Applications
关键词
TSP问题
HOPFIELD网络
能量函数
遗传算法
travel salesman problem（TSP）
Hopfield neural network
energy function
genetic algorithm
