基于混沌粒子群算法和Hopfield网络求解TSP问题

Solving TSP Via Chaotic PSO and HNN Algorithm

针对Hopfield网络求解TSP问题经常出现局部最优解，该文将混沌粒子群算法（PSO）与之结合，提出一种基于混沌粒子群的Hopfield神经网络方法。通过实验将其与文献[5，8]以及“PSO＋HNN”策略比较，验证了该文算法不仅能够以更大概率收敛到全局最优，而且耗时更少。

Since the Hopfield network often suffers from being trapped in local extrema when used to solve the traveling salesman problem, this paper combines the chaotic particle swarm optimization （PSO） and Hopfield neural networks （HNN） to form a novel algorithm, CP- SO-HNN. Experiments show that the proposed method outperform References [5,8] and the strategy of ＂PSO plus HNN＂ in terms of both global convergence rate and computation time.