基于海潮启示改进的退火算法在TSP求解中的研究与应用

发布时间：2018-04-23 04:04

本文选题：海潮启示 + 启发式算法　；参考：《北京工业大学》2015年硕士论文

【摘要】：海洋潮汐是一种复杂的地球物理现象,月球和太阳的引潮力在引起了这种自然现象发生时,会导致地面倾斜潮汐、重力潮汐、地球应变潮汐的发生。从牛顿首先应用万有引力定律解释了潮汐现象,到拉普拉斯提出了潮汐动力学理论,直至当代的中外科学家对潮汐测算所衍生的数学模型,使得海潮的测算已经达到了较高的精确度。智能算法是与计算机科学密切相关的萌发于20世纪80年代的新兴领域,尽管只有三十几年的发展过程,却已经引起众多学科领域研究人员的关注,目前已经成为人工智能、经济、社会、生物、工程等许多学科的热点。模拟退火算法,是受固体加热后退火的粒子状态启发而形成的启发式智能算法。从渐进式特征来看,海洋潮汐与固体退火的数学模型有相似之处;受此启发,把海洋潮汐研究领域的科学成果借鉴到模拟退火算法的改进之中具有一定的可行性。经本文研究,基于海潮启示改进的模拟退火算法在求解TSP时,既保留了退火算法原有的避免陷入局部解而能全局寻优的优点,保持了原有的鲁棒性长处;同时,取得了TSP路径距离得以优化的进展,达到了改进原算法,解决TSP问题的基础研究意义。
[Abstract]:Ocean tide is a complex geophysical phenomenon. When the lunar and solar tidal force causes this natural phenomenon, it will lead to the earth tide, gravity tide and earth strain tide. From Newton's first application of the law of universal gravity to the explanation of tidal phenomena, to Laplacian's theory of tidal dynamics, and to the mathematical models derived from modern Chinese and foreign scientists' calculations of tides, So that the tide of the calculation has reached a high degree of accuracy. Intelligent algorithm is a new field which is closely related to computer science. Although it has only developed for more than 30 years, it has attracted the attention of researchers in many disciplines and has become artificial intelligence. Economic, social, biological, engineering and many other disciplines of hot spots. Simulated annealing algorithm is a heuristic intelligent algorithm inspired by the state of particles annealed after solid heating. From the point of view of progressive characteristics, the mathematical models of ocean tide and solid annealing are similar, and inspired by this, it is feasible to use the scientific achievements of ocean tide research field for reference to the improvement of simulated annealing algorithm. In this paper, the improved simulated annealing algorithm based on ocean tide inspiration not only preserves the advantages of annealing algorithm in solving TSP, but also preserves the original advantages of global optimization, such as avoiding falling into local solution, and keeping the original robustness, at the same time, The progress of TSP path distance optimization has been achieved, and the basic research significance of improving the original algorithm and solving the TSP problem has been achieved.
【学位授予单位】：北京工业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：P731.23;TP18

【相似文献】