针对TSP问题,提出了一种基于自适应评价函数的改进的遗传算法,并且给出选择、交叉和变异操作的设计,实验表明算法维持了群体的多样性,防止算法过早收敛而陷入局部最优解,更有效地搜出全局近似最优解。 Aiming at the problem of TSP, an improved genetic algorithm based on adaptive evaluation function is proposed, and the design of selection, crossover and mutation operation is given. Experiments show that the algorithm maintains the diversity of groups and prevents the algorithm from converging prematurely and locally Optimal solution, more effectively find the global approximate optimal solution.