基于遗传算法的TSP问题求解

A Solution to TSP Problem Based on Genetic Algorithm

遗传算法（ＧＡ）是一种基于自然群体遗传机制的高效搜索算法，由于它在搜索空间中同时考虑许多点，这样就减少了收敛于局部极小的可能，同时也增加了处理的并行性。因此，可以利用遗传算法研究典型的组合优化实例———ＴＳＰ问题的求解问题，相应的求解方法称为遗传优化算法。计算机模拟结果表明，与Ｈｏｐｆｉｅｌｄ神经网络算法相比较，遗传优化算法不仅在收敛速度方面优于神经网络算法，而且通过较少的计算量就可以得到优于神经网络算法而更接近于最优解的满意解。

Genetic Algorithm (GA) is a effective searching algorithm based on genetic mechanism of natural colony, consulting simultaneously several points in searching space, GA can reduce the possibility of converging on local minimum and enhances the processing parallelization . Therefore,GA can be used in studying the solution of typical instance about combinatorial optimization——TSP problem , which is called Genetic Optimization Algorithm. Being compared with using Hopfiled Neural Network, the computer simulated results show that using Genetic Optimization Algorithm not only improve convergent speed, but also gain satisfied results, which is more close to the optimized result than using Hopfiled Neural Network.