磷虾群优化算法的改进及应用
发布时间：2018-07-24 11:07
【摘要】：磷虾群优化算法是由Gandomi和Alavi于2012年提出的一种新的算法.该算法具有收敛性强、编程简单和易于实现等优点,也具有收敛精度较差、计算效率较低和应用领域较少等缺点.本文对磷虾群优化算法进行了改进和应用,主要工作如下:1.提出了一种基于自然选择和随机扰动的改进磷虾群优化算法.在磷虾群优化算法的每一轮迭代中,对磷虾个体运动中的诱导权重和觅食权重采用基于时变的非线性递减策略,其次在新一代磷虾个体的生成过程中加入随机扰动,最后通过自然选择,提升新一代磷虾个体的质量,以此来提升算法的全局搜索和局部勘探能力.2.提出了一种基于改进的粒子群和磷虾群的混合算法.该算法首先对磷虾群优化算法中的觅食权重和诱导权重采用了一种新的非线性递减策略,然后将其与惯性权重指数递减的粒子群算法混合,采用双子种群策略共享有效信息,以此来提升运行效率.最后将自然界中生物优胜劣汰的进化机制引入.结果表明,求解精度和运行效率都得到了很大的提升.3.将上述用于解决无约束优化问题的基于改进的粒子群和磷虾群的混合算法在加入约束处理机制后应用于求解约束优化问题——非线性混合整数规划问题,并在求解精度和成功率上与其他算法进行了比较.
[Abstract]:Krill swarm optimization algorithm is a new algorithm proposed by Gandomi and Alavi in 2012. The algorithm has the advantages of strong convergence, simple programming and easy implementation, but also has the disadvantages of poor convergence accuracy, low computational efficiency and less application fields. In this paper, the optimization algorithm of krill colony is improved and applied, the main work is as follows: 1. An improved algorithm based on natural selection and random perturbation is proposed. In each iteration of the algorithm, the induced weight and the foraging weight in the individual movement of the krill are reduced by the nonlinear decreasing strategy based on time varying, and then the random disturbance is added to the process of the generation of the new generation of individuals of the krill. Finally, through natural selection, the quality of the new generation of krill individuals is improved to improve the global search and local exploration ability of the algorithm. A hybrid algorithm based on improved particle swarm and krill swarm is proposed. In this algorithm, a new nonlinear decreasing strategy is adopted for the feeding weight and induced weight of the population optimization algorithm of krill population, and then the algorithm is mixed with the particle swarm optimization algorithm with decreasing inertial weight index, and the effective information is shared by using the Gemini population strategy. In order to improve the operational efficiency. Finally, the evolutionary mechanism of survival of the fittest in nature is introduced. The results show that the accuracy and efficiency of the solution are greatly improved. The hybrid algorithm based on improved particle swarm and krill swarm for solving the unconstrained optimization problem is applied to the constrained optimization problem-nonlinear mixed integer programming problem after the constraint processing mechanism is added. The accuracy and success rate are compared with other algorithms.
【学位授予单位】：北方民族大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP18
本文编号：2141182
[Abstract]:Krill swarm optimization algorithm is a new algorithm proposed by Gandomi and Alavi in 2012. The algorithm has the advantages of strong convergence, simple programming and easy implementation, but also has the disadvantages of poor convergence accuracy, low computational efficiency and less application fields. In this paper, the optimization algorithm of krill colony is improved and applied, the main work is as follows: 1. An improved algorithm based on natural selection and random perturbation is proposed. In each iteration of the algorithm, the induced weight and the foraging weight in the individual movement of the krill are reduced by the nonlinear decreasing strategy based on time varying, and then the random disturbance is added to the process of the generation of the new generation of individuals of the krill. Finally, through natural selection, the quality of the new generation of krill individuals is improved to improve the global search and local exploration ability of the algorithm. A hybrid algorithm based on improved particle swarm and krill swarm is proposed. In this algorithm, a new nonlinear decreasing strategy is adopted for the feeding weight and induced weight of the population optimization algorithm of krill population, and then the algorithm is mixed with the particle swarm optimization algorithm with decreasing inertial weight index, and the effective information is shared by using the Gemini population strategy. In order to improve the operational efficiency. Finally, the evolutionary mechanism of survival of the fittest in nature is introduced. The results show that the accuracy and efficiency of the solution are greatly improved. The hybrid algorithm based on improved particle swarm and krill swarm for solving the unconstrained optimization problem is applied to the constrained optimization problem-nonlinear mixed integer programming problem after the constraint processing mechanism is added. The accuracy and success rate are compared with other algorithms.
【学位授予单位】：北方民族大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP18
【参考文献】
相关期刊论文 前3条
1 王磊;张汉鹏;;基于混沌搜索与精英交叉算子的磷虾觅食算法[J];计算机工程;2015年03期
2 白俊强;尹戈玲;孙智伟;;基于二阶振荡及自然选择的随机权重混合粒子群算法[J];控制与决策;2012年10期
3 李会荣;高岳林;李济民;;一种非线性递减惯性权重策略的粒子群优化算法[J];商洛学院学报;2007年04期
相关博士学位论文 前1条
1 王维博;粒子群优化算法研究及其应用[D];西南交通大学;2012年
，本文编号：2141182
本文链接：https://www.wllwen.com/kejilunwen/zidonghuakongzhilunwen/2141182.html
