融入变异交叉的改进天牛须算法求解TSP及工程应用

为找到最短路径,克服传统算法收敛速度慢、求解精度低等问题,提出一种融入变异交叉的改进天牛群算法(MBSO).首先将个体天牛转换成群体天牛搜索寻优;在群体进化过程中融入变异和交叉,提高全局搜索到更优结果;最后加入天牛须间长度自适应和步长自适应机制的搜索算法,改善算法的探索能力.将改进的算法通过MATLAB对TSPLIB中的数据集进行仿真实验,并用于PON网络规划问题.证明改进的天牛须算法在收敛速度和求解精度两方面较其他算法都有所提升,算法运行时间平均减少0.3 s,实验结果更接近最优解.

Improved beetle antennae search algorithm with mutation crossover in TSP problem and engineering application

In order to find the shortest path and overcome the slow convergence speed and low accuracy of traditional algorithms. This paper proposed a mutation and crossover based on beetle swarm search algorithm(MBSO). Firstly, it transformed individual beetle into group beetle search for optimization. And then incorporation of mutation and crossover in the population evolution process to improve global search and get better results. Finally, it added the adaptive length and step size of beetle antennae to improve the exploration ability of the algorithm. By using MATLAB to simulate the data set in TSPLIB and applied to PON network planning problem, the experimental simulation results show that the improved algorithm has higher convergence speed and solution precision than other algorithms. The average running time of the algorithm is reduced by 0.3 s, the experimental results are closer to the optimal solution.