一种求解TSP的自适应蚁群优化算法

针对标准蚁群算法在求解旅行商问题（TSP）时存在收敛速度慢，易陷入局部最优等缺陷，提出一种自适应蚁群优化算法．该算法设计了一种实时监测机制和一种新的搜索方向，实时监测机制可以让算法跳出局部最优值，并且当算法跳出局部最优值时，沿着新的搜索方向搜索，可以加快算法收敛到全局最优．通过对典型的TSP实例仿真实验，与基本蚁群算法、MMAS算法相对比，结果显示该算法在克服早熟现象和收敛速度方面有很大的优越性．

Adaptive ant colony optimization algorithm and TSP simulation

According to the standard ant colony algorithm in solving traveling salesman problem of （TSP） has the defects,such as slow convergence speed and easily trapped in local most superior,an adaptive ant colony optimization algorithm was proposed . The algorithm designs a real-time monitoring mechanism and a new search direction. Real-time monitoring mechanism can let the algorithm to jump out of local optimal value,and when the algorithm jump out of local optimal value, the new search direction can speed up the algorithm convergence to global optimal. Through the typical TSP instance simulation, com- pared with the basic ant colony algorithm and MMAS algorithm, the resuh shows that the algorithm has great superiority in overcoming the premature phenomenon and the convergence speed.