《Thesis - Automated load curve data cleansingin power systems》由會(huì)員上傳分享，免費(fèi)在線閱讀，更多相關(guān)內(nèi)容在學(xué)術(shù)論文-天天文庫(kù)。

1、AutomatedLoadCurveDataCleansinginPowerSystemsbyJiyiChenB.Sc.PekingUniversity,China,2008ATHESISSUBMITTEDINPARTIALFULFILLMENTOFTHEREQUIREMENTSFORTHEDEGREEOFMASTEROFSCIENCEintheSchoolofComputingScience?JiyiChen2010SIMONFRASERUNIVERSITYSpring2010Allrightsreserved.However,inaccordancewit

2、htheCopyrightActofCanada,thisworkmaybereproduced,withoutauthorization,undertheconditionsforFairDealing.Therefore,limitedreproductionofthisworkforthepurposesofprivatestudy,research,criticism,reviewandnewsreportingislikelytobeinaccordancewiththelaw,particularlyifcitedappropriately.APP

3、ROVALName:JiyiChenDegree:MasterofScienceTitleofThesis:AutomatedLoadCurveDataCleansinginPowerSystemsExaminingCommittee:Chair:Dr.QianpingGu___________________________________________Dr.KeWangSeniorSupervisor___________________________________________Dr.JiguoCaoCo-Supervisor___________

4、________________________________Dr.FredPopowichSFUExaminerDateApproved:___________________________________________iiAbstractLoadcurvedatareferstopowerconsumptionrecordedbymetersatcertaintimeintervalsatdeliverypointsorenduserpoints,andcontainsvitalinformationforday-to-dayoperations,s

5、ystemanalysis,systemvisualization,systemreliabilityperformance,energysavingandadequacyinsystemplanning.Itisunavoidablethatloadcurvescontaincorrupteddataandmissingdataduetovariousrandomfailurefactorsinmetersandtransferprocesses.Inthisthesis,nonparametricsmoothingtechniquesareproposed

6、tomodeltheloadcurvedataanddetectcorrupteddata.Anadaptedmultiplicativemodelisbuilttocorrectcorrupteddataandfillinmissingdata.Inimplementation,anincrementaltrainingprocedureisproposedtoenhancetheperformance.TheexperimentresultsontherealBCTC(BritishColumbiaTransmissionCorporation)loadc

7、urvedatademonstratedtheeffectivenessofthepresentedsolution.Keywords:Timeseries,DynamicTimeWarping,Loadmanagement,Loadmodelling,Powersystems,Powerquality,Smoothingmethods.iiiAcknowledgementsIwouldliketothankmyseniorsupervisorDr.KeWangforhisinvaluableguidanceandsupport.Thisworkwouldno

8、thavebeenarealitywi