資源描述：
《求解全局優(yōu)化問題的混合自適應(yīng)正交遺傳算法》由會員上傳分享，免費(fèi)在線閱讀，更多相關(guān)內(nèi)容在行業(yè)資料-天天文庫。
1、ISSN1000-9825,CODENRUXUEWE-mail:jos@iscas.ac.cnJournalofSoftware,Vol.21,No.6,June2010,pp.1296?1307http://www.jos.org.cndoi:10.3724/SP.J.1001.2010.03592Tel/Fax:+86-10-62562563?byInstituteofSoftware,theChineseAcademyofSciences.Allrightsreserved.?求解全局優(yōu)化問題的混
2、合自適應(yīng)正交遺傳算法+江中央,蔡自興,王勇(中南大學(xué)信息科學(xué)與工程學(xué)院,湖南長沙410083)HybridSelf-AdaptiveOrthogonalGeneticAlgorithmforSolvingGlobalOptimizationProblems+JIANGZhong-Yang,CAIZi-Xing,WANGYong(SchoolofInformationScienceandEngineering,CentralSouthUniversity,Changsha410083,China)+Cor
3、respondingauthor:E-mail:ywang@csu.edu.cnJiangZY,CaiZX,WangY.Hybridself-adaptiveorthogonalgeneticalgorithmforsolvingglobalopti-mizationproblems.JournalofSoftware,2010,21(6):1296?1307.http://www.jos.org.cn/1000-9825/3592.htmAbstract:Thispaperpresentsahybri
4、dself-adaptiveorthogonalgeneticalgorithm(HSOGA)basedonorthogonalexperimentaldesignmethodforsolvingglobaloptimizationproblems.InHSOGA,theorthogonalexperimentaldesignmethodisutilizedtodesigncrossoveroperator,andasaresult,aself-adaptiveorthogonalcrossoverop
5、eratorisproposed.Theself-adaptiveorthogonalcrossoveroperatorself-adaptivelyadjuststhenumberoforthogonalarray’sfactorsandthelocationfordividingtheparentsintoseveralsub-vectorsaccordingtothesimilarityofthetwoparents,inordertoproduceasmallbutrepresentatives
6、etofpointsasthepotentialoffspring.Inaddition,inHSOGAtheself-adaptiveorthogonalcrossoveroperatorisalsoadoptedtogenerateaninitialpopulationthatisscattereduniformlyoverthefeasiblesolutionspaceinordertomaintainthediversity.Moreover,alocalsearchschemeisincorp
7、oratedintoHSOGAinthepurposeofenhancingthelocalsearchabilityandspeedinguptheconvergenceofHSOGA.HSOGAistestedwithfourteenbenchmarkfunctions.TheexperimentalresultssuggestthatHSOGAisgenericandeffective.Keywords:orthogonalgeneticalgorithm;localsearch;globalop
8、timization;orthogonalexperimentaldesign摘要:提出了一種基于正交實驗設(shè)計的混合自適應(yīng)正交遺傳算法(hybridself-adaptiveorthogonalgeneticalgorithm,簡稱HSOGA)以求解全局優(yōu)化問題,此算法利用正交實驗設(shè)計方法設(shè)計交叉算子,并提出一種自適應(yīng)正交交叉算子.該自適應(yīng)正交交叉算子根據(jù)父代個體的相似度自適應(yīng)地調(diào)整正交表的因素個數(shù)和對父代個體進(jìn)行因素分割的位置,生成具有代
