一种新的自适应粒子群优化算法

《一种新的自适应粒子群优化算法》由会员上传分享，免费在线阅读，更多相关内容在行业资料-天天文库。

1、Seediscussions,stats,andauthorprofilesforthispublicationat:https://www.researchgate.net/publication/4094255AdaptiveparticleswarmoptimizationalgorithmConferencePaper·July2004DOI:10.1109/WCICA.2004.1341988 · Source:IEEEXploreCITATIONSREADS61103authors,including:FengPanJieChenBeijingInstituteofTec

2、hnologyUniversityofMassachusettsBoston59PUBLICATIONS296CITATIONS460PUBLICATIONS9,783CITATIONSSEEPROFILESEEPROFILESomeoftheauthorsofthispublicationarealsoworkingontheserelatedprojects:NovelTherapiesforProstateCancerViewprojectSensormeasurementandcalibrationViewprojectAllcontentfollowingthispagew

3、asuploadedbyFengPanon05July2015.Theuserhasrequestedenhancementofthedownloadedfile.第34卷第7期计算机工程2008年4月Vol.34No.7ComputerEngineeringApril2008·人工智能及识别技术·文章编号：1000—3428(2008)07—0181—03文献标识码：A中图分类号：TP18一种新的自适应粒子群优化算法林川，冯全源(西南交通大学信息科学与技术学院，成都610031)摘要：基于粒子分工与合作的思想，提出一种自适应粒子群优化(PSO)算法。该算法为不同的粒子分配不同的任务

4、，对性能较好的粒子使用较大的惯性权，对性能较差的粒子采用较小的惯性权，加速系数根据惯性权自适应调整。将标准PSO算法中的全局最优位置与个体最优位置分别替换为相关个体最优位置的加权平均，更好地平衡了算法的全局与局部搜索能力，提高了算法的多样性与搜索效率。5个经典测试函数的仿真结果及与其他PSO算法的比较结果验证了该算法的有效性。关键词：粒子群优化；自适应参数；分工；平衡点；多样性NewAdaptiveParticleSwarmOptimizationAlgorithmLINChuan,FENGQuan-yuan(SchoolofInformationScienceandTechnolo

5、gy,SouthwestJiaotongUniversity,Chengdu610031)【Abstract】Basedontheideaofspecializationandcooperation,anadaptiveParticleSwarmOptimization(PSO)algorithmisproposed.Inthenewalgorithm,differentparticlesareassignedspecifictasks.Betterparticlesaregivenlargerinertialweights,whileworseonesaregivensmaller

6、inertialweights.Andtheparticle’saccelerationcoefficientsareadaptivelyadjustedaccordingtoitsinertialweight.Besides,thepersonalbestpositionandglobalbestpositioninstandardPSOalgorithmarerespectivelyreplacedbytheweightedmeanofsomerelevantpersonalbestpositions.ThesestrategiesimprovethePSOalgorithmat

7、theaspectsofdiversityandthebalanceofexplorationandexploitation.TheefficiencyofthenewalgorithmisverifiedbythesimulationresultsoffiveclassicaltestfunctionsandthecomparisonwithotherPSOalgorithms.【Keywords】ParticleSwarmOptimization(PS