基于分层型的改进粒子群优化算法  

Improved Particle Swarm Optimization Algorithm Based on Hierarchical

作　　者：安无恙 吕柏权[1] 李嫽 AN Wu-yang;LU Bai-quan;LI Liao(College of Mechatronic Engineering and Automation,Shanghai Universit,Shanghai 200072,Chin)

机构地区：[1]上海大学机电工程与自动化学院,上海200072

出　　处：《自动化与仪表》2018年第7期19-24,共6页Automation & Instrumentation

摘　　要：针对粒子群算法容易陷入局部极小点而无法得到全局最优值,且对于复杂的多极值点的函数的优化效果较差的问题。该文提出了一种改进的分层型粒子群优化算法,引入了多维空间旋转矩阵,在第一层的基础上,通过旋转不同角度矩阵产生多个层。该改进算法对整个粒子群进行旋转分层,增加粒子的多样性,从不同角度和平面寻找全局最优解,提高算法的全局搜索能力和收敛性能。最后对9个基准测试函数的仿真以及其统计数据,通过分析分层粒子群优化算法对每个基准函数的收敛速度和搜寻精度评价该算法的优劣。验证了改进后的算法满足函数优化要求。This paper presents an improved hierarchical particle swarm optimization algorithm,the introduction of multi-dimensional space rotation matrix,on the basis of the first layer,multiple layers are generated by rotating different angle matrices. The improved algorithm is used to spin the stratification of the whole particle swarm,increase the diversity of the particles,and find the optimal solution from different angles and planes to improve the global search ability and convergence performance of the algorithm. Through the simulation of 9 benchmark functions and their statistical data,the advantages and disadvantages of the proposed algorithm are evaluated by analyzing the convergence rate and search precision of the hierarchical particle swarm optimization algorithm. It is proved that the improved algorithm meets the requirement of function optimization.

关 键 词：旋转矩阵 分层型粒子群算法 收敛性能 搜索能力 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]