社会蜘蛛群优化算法改进分析及应用研究

发布时间：2018-04-26 15:46

本文选题：社会蜘蛛群优化算法 + 0-1背包问题　；参考：《广西民族大学》2017年硕士论文

【摘要】：社会蜘蛛群优化算法是模拟自然界蜘蛛进行分工合作、信息交流和繁衍后代的行为而出现的一种新兴的群智能优化算法,该算法具有结构简单,稳定性较强和易于理解的特点。自从提出之后就受到该领域学者的广泛关注。但随着研究的深入,一些学者发现该算法存在寻优精度较差,收敛速度较慢等缺点,这在一定程度上限制了社会蜘蛛群优化算法的理论发展和应用范围。本论文主要是针对社会蜘蛛群优化算法中存在的一些不足进行改进,并将改进的优化算法应用于实际的优化问题中,目的是进一步完善社会蜘蛛群优化算法的理论和拓展其应用范围。本文的工作内容主要分为以下五个方面:(1)提出一种利用社会蜘蛛群优化算法求解0-1背包问题的算法,该算法的优势在于求解高维的0-1背包问题。实验结果表明,在求解高维的0-1背包问题时社会蜘蛛群优化算法具有一定的优势。(2)提出一种利用社会蜘蛛群优化算法求解无线传感器覆盖率问题的方法,可快速地找出无线传感器最佳布置方案,并以可视的方式表现出来。实验结果表明,利用社会蜘蛛群优化算法得到的优化方案最佳。(3)提出一种基于精英反向学习策略的社会蜘蛛群优化算法,克服社会蜘蛛群优化算法易陷入局部最优的缺点,将精英反向学习策略引入社会蜘蛛群优化算法,实现扩大其搜索空间和增强种群多样性的目的,并将基于精英反向学习策略的社会蜘蛛群优化算法用于函数优化问题。(4)提出一种具有差分进化算子的社会蜘蛛群优化算法,较大程度上克服社会蜘蛛群算法在一些情况下寻优能力较差且收敛速度较慢的缺点,将差分进化算子引入社会蜘蛛群优化算法,增强了蜘蛛个体的全局搜索能力,进而提升了算法的性能,并将具有差分进化算子的社会蜘蛛群优化算法用于流水线型生产车间调度问题。(5)提出一种基于量子编码的社会蜘蛛群优化算法,引入量子编码的思想,量子编码扩展了个体信息的多样性,量子旋转门作为更新策略,提升了算法的局部和全局搜索能力,且将改进后的量子社会蜘蛛群优化算法应用到水电站优化调度问题。
[Abstract]:The social spider swarm optimization algorithm is a new kind of swarm intelligence optimization algorithm which simulates the behavior of natural spiders working together, exchanging information and breeding offspring. The algorithm has the characteristics of simple structure, strong stability and easy to understand. Since it was put forward, it has received extensive attention from scholars in this field. However, with the development of the research, some scholars find that the algorithm has some shortcomings, such as poor precision and slow convergence rate, which limits the theoretical development and application scope of the social spider swarm optimization algorithm to a certain extent. This paper mainly aims at improving some shortcomings of the social spider swarm optimization algorithm, and applies the improved optimization algorithm to the actual optimization problem. The aim is to further improve the theory of social spider swarm optimization algorithm and expand its application scope. The main work of this paper is divided into the following five aspects: 1) an algorithm for solving 0-1 knapsack problem using social spider swarm optimization algorithm is presented. The advantage of this algorithm is to solve the 0-1 knapsack problem with high dimension. The experimental results show that the social spider swarm optimization algorithm has some advantages in solving the high dimensional 0-1 knapsack problem. It can quickly find out the best arrangement of wireless sensor and display it visually. The experimental results show that a social spider swarm optimization algorithm based on elite reverse learning strategy is proposed, which overcomes the shortcoming that the social spider swarm optimization algorithm is prone to fall into local optimum. The elite reverse learning strategy is introduced into the social spider swarm optimization algorithm to expand the search space and enhance the diversity of the population. The social spider swarm optimization algorithm based on elite reverse learning strategy is applied to the function optimization problem. (4) A social spider swarm optimization algorithm with differential evolution operator is proposed. In order to overcome the shortcomings of the social spider swarm algorithm in some cases, the differential evolution operator is introduced into the social spider swarm optimization algorithm, and the global search ability of the spider individual is enhanced. Furthermore, the performance of the algorithm is improved, and the social spider swarm optimization algorithm with differential evolution operator is applied to the production shop scheduling problem of pipeline type. (5) A social spider swarm optimization algorithm based on quantum coding is proposed, and the idea of quantum coding is introduced. Quantum coding expands the diversity of individual information. Quantum rotary gate is used as an update strategy to improve the local and global search ability of the algorithm. The improved quantum society spider swarm optimization algorithm is applied to the optimal operation of hydropower station.
【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】