预览加载中,请您耐心等待几秒...
1/5
2/5
3/5
4/5
5/5

在线预览结束,喜欢就下载吧,查找使用更方便

如果您无法下载资料,请参考说明:

1、部分资料下载需要金币,请确保您的账户上有足够的金币

2、已购买过的文档,再次下载不重复扣费

3、资料包下载后请先用软件解压,在使用对应软件打开

42011,47(18)ComputerEngineeringandApplications计算机工程与应用 一种蛙跳和差分进化混合算法 何兵1,2,车林仙1,2,刘初升1 HEBing1,2,CHELinxian1,2,LIUChusheng1 1.中国矿业大学机电工程学院,江苏徐州221008 2.泸州职业技术学院机电工程研究所,四川泸州646005 1.SchoolofMechanicalandElectricalEngineering,ChinaUniversityofMiningandTechnology,Xuzhou,Jiangsu221008,China 2.InstituteofMechatronicsEngineering,LuzhouVocationalandTechnicalCollege,Luzhou,Sichuan646005,China HEBing,CHELinxian,LIUChusheng.Novelhybridshuffledfrogleapinganddifferentialevolutionalgorithm.Comput- erEngineeringandApplications,2011,47(18):4-8. Abstract:ShuffledLeapingFrogAlgorithm(SFLA)ischaracterizedbysimplicity,fewcontrolparametersrequired,andeasily beused,buthasthedisadvantagesofprematureconvergenceandlowprecisionforhardhigh-dimensionaloptimizationprob- lems,duetoitsrapidlossofthepopulationdiversityandthelackoflocalrefinedsearchabilitiesatthelaterstagesofgen- erations.InordertoovercometheeasyprematureorearlyconvergenceofSFLAs,thispaperhybridizestheSFLAandthe DifferentialEvolution(DE)algorithmtoformahybridoptimizationalgorithm,namelySFL-DE,whichborrowstheideafrom DE/best/1/binstrategythathastheadvantagesofstrongglobalsearchabilityandbetterpopulationdiversity.Comparisonsare presentedtotestperformancesofthenewalgorithmemploying6benchmark30-dimensionalfunctions.ComparedwithSFLA andstandardDE(i.e.,DE/best/1/binandDE/rand/1/binschemes)algorithms,theexperimentalresultsintermsoftheglobal optimizationefficiency,thesolutionaccuracyandthecomputationrobustnessdemonstratethattheSFL-DEalgorithmisabetter toolforsolvingsomebenchmarkoptimizationproblemswithinafewfixedgenerations,buttakesalongerruntime. Keywords:ShuffledFrogLeapingAlgorithm(SFLA);DifferentialEvolution(DE)algorithm;hybridoptimization;continuous optimizationproblem 摘要:混洗蛙跳算法(SFLA)具有算法简单、控制参数少、易于实现等优点,但在高维难优化问题中算法容易早熟收敛且求解精 度不高。导致该缺陷的主要原因是在进化后期种群多样性迅速下降,且缺乏局部细化搜索能力。借鉴差分进化算法(DE)中DE/ best/1/bin版本具有全局搜索能力较强、种群多样性较好的优点,将SFLA与DE有机融合,形成混合优化算法(SFL-DE),以克服 SFLA容易早熟收敛的缺陷。给出了6个30维benchmark问题数值对比实验,结果表明,在给定的较小进化代数内,SFL-DE的寻 优效率、计算精度、鲁棒性