合肥工业大学学报(自然科学版)
閤肥工業大學學報(自然科學版)
합비공업대학학보(자연과학판)
JOURNAL OF HEFEI UNIVERSITY OF TECHNOLOGY(NATURAL SCIENCE)
2014年
8期
928-932
,共5页
杨再甫%黄友锐%曲立国%葛平平
楊再甫%黃友銳%麯立國%葛平平
양재보%황우예%곡입국%갈평평
常规蚁群算法%改进蚁群算法%旅行商问题%局部最优解%动态蚂蚁数目
常規蟻群算法%改進蟻群算法%旅行商問題%跼部最優解%動態螞蟻數目
상규의군산법%개진의군산법%여행상문제%국부최우해%동태마의수목
conventional ant colony algorithm%improved ant colony algorithm%travelling salesman problem(TSP)%local optimal solution%dynamic number of ants
蚂蚁数目是影响蚁群算法性能的重要参数,常规蚁群算法在求解TSP时易于陷入局部最优解。文章针对该问题,提出了一种蚂蚁数目动态改变的蚁群算法,即每次周游时的蚂蚁数目是在一个范围内随机取值,该改进算法借用遗传算法中的排序选择策略对每次遍历时的蚂蚁位置进行初始化;分别对常规蚁群算法的TSP求解和改进蚁群算法的TSP求解进行了原理阐述,并对2种算法求解 TSP的结果进行了Matlab仿真。对比仿真结果表明,改进的算法在求解TSP时,能够有效地跳出局部最优解,并能很好地收敛,它比常规蚁群算法的性能要优。
螞蟻數目是影響蟻群算法性能的重要參數,常規蟻群算法在求解TSP時易于陷入跼部最優解。文章針對該問題,提齣瞭一種螞蟻數目動態改變的蟻群算法,即每次週遊時的螞蟻數目是在一箇範圍內隨機取值,該改進算法藉用遺傳算法中的排序選擇策略對每次遍歷時的螞蟻位置進行初始化;分彆對常規蟻群算法的TSP求解和改進蟻群算法的TSP求解進行瞭原理闡述,併對2種算法求解 TSP的結果進行瞭Matlab倣真。對比倣真結果錶明,改進的算法在求解TSP時,能夠有效地跳齣跼部最優解,併能很好地收斂,它比常規蟻群算法的性能要優。
마의수목시영향의군산법성능적중요삼수,상규의군산법재구해TSP시역우함입국부최우해。문장침대해문제,제출료일충마의수목동태개변적의군산법,즉매차주유시적마의수목시재일개범위내수궤취치,해개진산법차용유전산법중적배서선택책략대매차편력시적마의위치진행초시화;분별대상규의군산법적TSP구해화개진의군산법적TSP구해진행료원리천술,병대2충산법구해 TSP적결과진행료Matlab방진。대비방진결과표명,개진적산법재구해TSP시,능구유효지도출국부최우해,병능흔호지수렴,타비상규의군산법적성능요우。
T he number of ants is an important parameter that affects the performance of ant colony al-gorithm .As the conventional ant colony algorithm for solving travelling salesman problem (TSP) is easy to fall into a local optimal solution ,an ant colony algorithm based on dynamic changes of the number of ants is proposed ,in which each traveling is to be with random number of ants in a certain range .Besides ,the ranking selection policy of genetic algorithm is used to initialize the location of ants each time when traveling .The theories of the conventional ant colony algorithm and the improved ant colony algorithm for solving TSP are both expatiated ,and the results of the mentioned algorithms for solving TSP are simulated by Matlab .The simulation results show that the performance of the im-proved algorithm is better than that of the conventional algorithm since the improved algorithm can ef-fectively jump out of local optimal solution and has better convergence performance .