多元宇宙优化算法及应用研究

发布时间：2018-05-31 05:26

本文选题：多元宇宙优化算法 + 指数步长　；参考：《广西民族大学》2017年硕士论文

【摘要】：多元宇宙优化算法是受自然界中多元宇宙现象的启发而提出的一种元启发式算法,该算法结构简单、参数少易于理解且具有较强的搜索能力。近年来,该算法受到了国内外学者的极大关注。本论文针对多元宇宙优化算法所存在的一些不足,从指数步长增长和位置更新策略等方面进行改进,目的在于改进多元宇宙算法的整体性能,完善算法的理论基础,并将改进后的算法应用于函数优化,旅行商问题和聚类分析等,从而拓宽其应用领域。本文的工作主要包括以下3个方面:(1)采用指数步长增长的方式以增强算法的深度搜索性能,通过引入膨胀系数参数可以有效地缩小解的空间,使算法具有更快的搜索速度,同时也促使种群向最优个体附近迅速靠拢,使得算法在提高精度的同时加快了收敛速度。(2)针对传统方法在求解聚类分析问题时精度低,聚类效率不高,易陷入局部最优等问题,本文采用指数步长增长的方式加强算法局部搜索能力,提高了算法的执行效率,优化了多元宇宙算法对数据集的聚类分析能力。实验仿真结果表明,改进过的多元宇宙算法在求解聚类分析问题中是可行和高效的。(3)针对传统算法在求解城市数量较大的旅行商问题时运行速度较慢,在多元宇宙优化算法的基础上引入指数步长增长的方式,在求解旅行商问题可以加快搜索速度,并对求解过程中出现的异常解后及时进行修正,可更高效地对旅行商问题进行求解。
[Abstract]:The multi-universe optimization algorithm is a meta-heuristic algorithm inspired by the phenomenon of multi-universe in nature. The algorithm is simple in structure, easy to understand with few parameters and has strong searching ability. In recent years, the algorithm has attracted great attention from scholars at home and abroad. In order to improve the overall performance of the multiverse algorithm and improve the theoretical basis of the algorithm, this paper aims at improving the exponential step size growth and location updating strategy in view of the shortcomings of the multiverse optimization algorithm. The improved algorithm is applied to function optimization, traveling salesman problem and clustering analysis. The work of this paper mainly includes the following three aspects: 1) the method of exponential step growth is adopted to enhance the depth search performance of the algorithm. By introducing the expansion coefficient parameter, the space of the solution can be reduced effectively, and the search speed of the algorithm is faster. At the same time, it also makes the population close to the optimal individual rapidly, which makes the algorithm improve the accuracy and speed of convergence. (2) aiming at the problems of low precision, low clustering efficiency and easy to fall into the local optimum, the traditional methods can solve the cluster analysis problem with low precision, low clustering efficiency, and easy to fall into local optimization. In this paper, the exponential step size is used to enhance the local search ability of the algorithm, to improve the efficiency of the algorithm, and to optimize the clustering analysis ability of the multiverse algorithm to the data set. The experimental results show that the improved multiverse algorithm is feasible and efficient in clustering analysis. On the basis of the multi-universe optimization algorithm, the exponential step growth method is introduced. The search speed can be accelerated by solving the traveling salesman problem, and the abnormal solutions in the process of solving the problem can be corrected in time. The traveling salesman problem can be solved more efficiently.
【学位授予单位】：广西民族大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP18

【参考文献】