Page 83 - 2023年第54卷第1期
P. 83
08 - 23.
[27] 许银山,梅亚东,杨娜,等.大规 模 混 联 水 库 群 长 期 优 化 调 度 [J].水 电 自 动 化 与 大 坝 监 测,2010,34
(4):58 - 63.
[28] 王金文,王仁权,张勇传,等.逐次逼近随机动态规划及库群优化调度 [J].人民长江,2002(11):45 -
47 ,54.
[29] 马立亚,雷晓辉,蒋云钟,等.基于 DPSA的梯级水库群优化调度 [J].中 国 水 利 水 电 科 学 研 究 院 学 报,
2012,10(2):140 - 145.
[30] 王丽萍,任宇,邱 启 仓,等.多 目 标 进 化 算 法 性 能 评 价 指 标 研 究 综 述 [J].计 算 机 学 报,2021,44(8):
1590 - 1619.
[31] ZITZLERE,THIELEL.Multi - objectiveOptimizationUsingEvolutionaryAlgorithms - A ComparativeCaseStudy
[C]??ProceedingsoftheInternationalConferenceonParallelProblemSolvingfromNature.Berlin,Germany.1998.
[32] 吴志远,黄显峰,李昌平,等.基于分段粒子群算法的梯级水库多目标优化调度模型研究[J].水资源与
水工程学报,2020,31(3):145 - 154.
Researchonmulti - objectiveoptimaloperationmodelofcascadedhydropower
system andCPF - DPSAalgorithm
4
1
3
1
1,2
1
AIXueshan ,GUOJiajun,MUZhenyu,CHENSenlin,YANGBaiyin,ZHOUPengcheng
(1.StateKeyLaboratoryofWaterResourcesandHydropowerEngineeringScience,WuhanUniversity,Wuhan 430072,China;
2.HubeiKeyLaboratoryofWaterSystemScienceforSpongeCityConstruction(WuhanUniversity),Wuhan 430072,China;
3.ChinaRenewableEnergyEngineeringInstitute,Beijing 100120,China;
4.KunmingInstitutionofSurveyingandMappingCo.,LtdofReconnaissanceofChinaElectricConstructionGroup,
Kunming 650051,China)
Abstract:Themulti - objectiveoptimaloperationofcascadedhydropowersystemisanimportantresearchcontentin
theefficientutilizationofhydropowerresources ,inwhichtheuniversalityoftheexistingmulti - objectiveoptimal
operationmodelanditssolvingalgorithmisurgenttoimprove.Onthebasisofsummarizingtheexistingthreetypes
ofobjectivefunction(cumulativevalue,extremevalueandpercentage),themulti - objectiveoptimaloperation
modelofcascadehydropowersystemwithuniversalapplicabilityisestablishedwiththemaximumofpowergenera
tion ,themaximumoftheminimumoutputpowerandthemaximumofwatersupplyguaranteerateofecologicalsec
tion ,andthedynamicprogrammingsuccessiveapproximationmethodbasedonthecombinationofpenaltyfactors
(CPF - DPSA)isproposedtosolvethemodel.Thisalgorithmexplorestherelationshipbetweeneachtargetandthe
correspondingfactor ,confirmsthescopeofinfluenceofeachpenaltycoefficient,andobtainsthenon - dominated
solutionsetwithrelativelyuniform distribution.TheapplicationofNam Oucascadedhydropowersystem inLaos
showsthatthismodelhasgoodapplicabilityandthenon - dominatedsolutionsetobtainedbyCPF - DPSAalgorithm
isdistributedwidelyanduniformly.Especiallyinthelongseriesproblem ,thealgorithm hasbetterperformance
thanthethirdgenerationoffastnon - dominatedsortinggeneticalgorithm(NSGA - Ⅲ)intermsofresultaccuracy,
non - inferiorsolutionqualityandParetofrontdistribution.
Keywords: multi -objective; cascadedhydropowersystem; CPF-DPSA; penaltyfactors; non -dominated
solutionset
(责任编辑:耿庆斋)
— 7 8 —