CAJ | 학술논문

为优化旅行商问题(TSP),结合禁忌搜索算法(TS)和模拟退火算法(SA)的思想设计了混合退火算法( TSA)。针对模拟退火算法搜索效果不稳定等问题,在初始阶段TSA多次禁忌搜索并筛选初始解,确保算法稳定地收敛到全局最优值,在求解部分设计了快速退火算法,使其快速退火并收敛。与其他算法相比,TSA求解精度高,求解效果稳定鲁棒性强,并且求解时间短。 TSA对China31问题的优化效果尤为精良,优化结果包括15375,15363,15352和15335等,均优于已知最好解15383。
위우화여행상문제(TSP),결합금기수색산법(TS)화모의퇴화산법(SA)적사상설계료혼합퇴화산법( TSA)。침대모의퇴화산법수색효과불은정등문제,재초시계단TSA다차금기수색병사선초시해,학보산법은정지수렴도전국최우치,재구해부분설계료쾌속퇴화산법,사기쾌속퇴화병수렴。여기타산법상비,TSA구해정도고,구해효과은정로봉성강,병차구해시간단。 TSA대China31문제적우화효과우위정량,우화결과포괄15375,15363,15352화15335등,균우우이지최호해15383。
The hybrid annealing algorithm ( TSA) , combining with the ideal of the Tabu Search ( TS) algorithm and the Simulated Annealing ( SA) algorithm, is proposed to optimize the well-known Traveling Salesman Problem ( TSP) . In view of the defect that the search performance of SA is not stable, the TSA algorithm use TS to search initial solutions for several times, to ensure the stable convergence of the algorithm to the global optimal value. Compared with other algorithms, the TSA algorithm has a high precision, strong robustness and rapid convergence. The TSA algorithm has an excellent effect on China31 problem, its optimization results include 15 375, 15 363, 15 352, and 15 335, which all are better than the known best value 15 378.