改进的连续Hopfield网络求解组合优化问题——以TSP求解为例

Improved Continuous Hopfield Neural Network for Solving Combinatorial Optimization Problems:An Example to Solve the TSP

利用神经网络求解组合优化问题,是一种有效的途径。对连续Hopfield网络的数学模型及稳定性进行了分析;探讨了组合优化问题的神经网络求解方法,针对传统方法参数配置复杂、收敛速度慢等不足,提出了改进算法;最后,通过系统仿真与性能测试验证了该算法的可行性。

Using neural networks to solve combinatorial optimization problems is an effective approach.Analyzes the mathematical model and stability of continuous Hopfield neural network,discusses the use of CHNN to solve combinatorial optimization problem and puts forward the improved algorithm aiming at the insufficiency of the traditional method of parameters configuration complex and convergence rate slower.Finally,through the system simulation and performance testing,demonstrates the algorithm feasible.