全局组搜索优化算法及其应用研究  

Global Group Search Optimizer Algorithm and Its Application Research

作　　者：张康[1] 顾幸生[1] 

机构地区：[1]华东理工大学化工过程先进控制与优化技术教育部重点实验室,上海200237

出　　处：《青岛科技大学学报（自然科学版）》2012年第5期529-534,共6页Journal of Qingdao University of Science and Technology:Natural Science Edition

基　　金：国家自然科学基金项目(61174040);国家863高技术研究发展计划项目(2009AA04Z141)

摘　　要：组搜索优化算法GSO(Group Search Optimizer)是一种基于动物捕食原理的新型群智能优化算法。本研究提出了一种改进的GSO优化算法:全局组搜索优化算法GGSO(Global GSO)。主要在两个方面对GSO算法进行了改进,一是在迭代过程中引入加速系数,加快种群收敛速度,增强算法的局部搜索能力;二是用高斯函数来产生随机位置变异,扩大搜索空间,从而增强算法的全局搜索能力。经过11个无约束测试函数和3个带约束问题的测试及与其他文献的比较可知,GGSO算法具有较好的局部和全局搜索能力,并且能够解决复杂的实际问题。The Group Search Optimizer （GSO） is a novel optimization algorithm, which is inspired by searching behavior of animals. In this paper we propose an improved GSO algorithm named Global Group Search Optimizer （GGSO） to balance the exploitation and exploration abilities of the algorithm. At first time, an improved search equation with an acceleration coefficient for the scroungers motion model is developed, which ac- celerates the moving speed of the scroungers toward the producer to improve the exploi- tation power. After that, a mutation operation using Gaussian function is introduced to enhance the rangers searching area which improve the exploration ability. The GGSO al- gorithm is evaluated on a set of 11 un-constrained numerical optimization problems and 3 constrained problems and compares favorably with the basic version of GSO. Experi- mental results indicate that the GGSO algorithm improves the performance on these problems significantly, and prove that the GGSO algorithm can be implied on practical problems.

关 键 词：组搜索优化算法 优化 全局数值优化 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]