区域电力市场负荷预测的研究

上海交通大学硕士学位论文区域电力市场负荷预测的研究姓名：周健申请学位级别：硕士专业：电气工程指导教师：蒋传文;陶卫国20060914(EMS)WEBTHERESEARCHOFPOWERLOADFORECASTINGINREGIONALPOWERMARKETABSTRACTLoadForecastingplaysanimportantroleintheEnergyManagementSystem(EMS),whichhasgreatinfluenceontheoperating,controllingandplanningofelectricpowersystem.Itisoneoftheimportanttasksoftheelectricdepartment.Withtheconditionoftheregionalpowermarket,loadforecastingisbecomingoneoftechnologicmeansthatsupportingtheregionalpowermarketduetoitsresponsetopowerprice,andhassignificancetoallpartsthattakepartintheregionalpowermarket.Therefore,enhancingtheloadforecastinglevelwillbepropitiousforthemanagementofelectricityplanning,willbepropitiousfortheformulationofreasonableelectricitysourceconstruction,willbeadvantageousinenhancingtheeconomicandsocialefficiencyofthepowersystemaswellas.ThisthesistakeHuadongregionfourprovincesandonecityforexampletostudiedproblemofloadforecastinginregionalpowermarket.Firstly,thisthesisanalyzedloadforecastinginconditionofregionalpowermarket.Thethesisdiscussedproblemsofregionalpowermarket,andinvestigatedtechnologiesanditsfeaturesofloadforecasting.Secondly,thisthesisanalyzedcharacteristicofHuadongregionalpowermarket,speciallystudiedrulesofloadanddiscussedrelationbetweentemperatureandload.Thirdly,thisthesisintroducesparticleswarmoptimizeralgorithmtoconfirmthemostoptimizedweightcoefficient,consequently,thecombinationmodeofloadforecastingissetup.Finally,thecombinationmodelofloadforecastingistestedwithhistoricalloaddataofHuadongPowerGridinthethesis,andkindofthesinglemodelswasalsousetoloadforecast,wecomparetheforecastingresultsafterusingthedataofthesameperiod,testresultsshowgoodpredictionaccuracy,thecombinationmodelhasmadeverygoodpredictionresult.TheanalysisofHuadongregionalpowermarketandthearchitecturedesignofHuadongloadforecastingsystembasedonthewebsiteinthethesiscanbebenefitedforthosewhotakepartinandserveHuadongregionalpowermarket.Keywords:RegionalPowerMarket,HuadongPowerGrid,LoadForecasting,ParticleSwarmOptimization(PSO),thecombinationmodel200511192005111920069141601.1[1],[2],[3]2002[2002]520021229251120035()1.1.1[4],[5](1)()1-11-21-3()()260(2)()1-11-23601-31.1.2[6](1)();(2);(3)()():(4)();(5)();(6)();(7)();(8)1.1.3[7],[8][10],[11]1460280%~90%;5601.1.4[12]L.V.Bertalanffy1234567891011121314151.2[13][14]1.2.11.2.2660(1)(2)(3)1.2.3(1)(2)(3)1.2.4(1)760(2)(3)(4)860(5)(6)(7)(8)9601.31.3.1[15],[16],[17],[18]1106023116041.3.2[18],[19],[20],[21];12601.3.312331.4ARARMA1360,WEB14602.1[22](1)(2)(ArtificialNeuralNetworkANN)15602.2[23](1)yp1ppxxx21ypxxx21+++++=),0(~222110σεεNxbxbxbbypp2-1210σpbbbpxxx21εppxxx21ynn2-1+++++=+++++=+++++=nnppnnppppxbxbxbbyxbxbxbbyxbxbxbbyεεε#221101222222110211122111012-2ε+=XBY2-3pbbb21∑=−−−−=nippixbxbbyQ12110)(2-4Qpbbb211660YXBXX′=′ˆ2-5Bˆ2-52-1(2)(3)2.3(1)XY2-5Bt-F-XY2-11760BoxJenkinsYt2-2etARMAARMAARIMAARMAARMAAR,MA,ARMA∇BttttYBYYY)1(1−=−=∇−dtdtdYBY)1(−=∇ARMAARMAdARMApqteBYBtd)()(θ=∇ΦARIMABTTetYt2-21860∇T∇TTB−=∇1ARIMAiTiDTTeBYB)()(θ=∇Φ)(TBΦiTeB)(θpqTpTq(2)(3)2-319602.4[24],[25](1),,,,ErrorBackPropagationBPNLM2-4BPBP2-4BPBPk2060ANN(2)(3)BP2-521602.5(1)19823hGMn,h)1(1)1(21)1(12)1(1221)1(111)1(1hnnnnnnnnxbxbxadtxdadtxdadtxd−−−−−++=++++2-6GM11GM11uaxdtdx=+)1()1(2-7x1x(0)au2260aueauxxakK+−=−+)(()0()1()1()1k=0,1,22-82-8GM1,1x0akakeauxex−−+−−=))(1()0()1()0()1(2-9(2)(3)GM11YNaˆGM(1,n)?C(8)C2-623602.6(Logistic)(Gompertz)2.719822460FoxBASETurbo-Basic1FoxBASETurbo-BasicTurbo-BasicFoxBASE2VP-EXPERTFoxBASEVP-EXPERT2.82.8.11MAD11MAD||NtteN==∑2-102MSE()211MSENtteN==∑2-113RMSE()211RMSENtteN==∑2-124MAPE256011MAPE100%NttteNy==×∑2-13ˆttteyy=−ˆtytyN2.8.21F-1-10:210====pbbbH∑=−=niiEyyQ12)ˆ(∑=−=niiieyyQ12)ˆ()1,(~)1(−−−−=pnpFpnQpQFeE2-14αF)1,(−−pnpFα)(21iipiiyxxxF)1,(−−≥pnpFFα0H)1,(−−pnpFFα0H2t-1-1),2,1(0:0pjbHj==jjc1)(−′=XXC)1(~)1(ˆ−−−−=pntpnQcbtejjjj2-15α)1(2−−≥pnttjαjt)1(2−−≥pnttjα0Hjbjxy266031ˆ−−=pnQeσ∑∑==−−++=pipjjjiijjxxxxcnd11000))((11y)1(α−−−±σαˆ)1(ˆ020dpnty2-162.990,2760[26]3.1200063.23.2.120031997GDP3.7112.5%200210.4:49.3:40.39.1:51.2:39.7GDP159871933200215%3.2.23.2.398.9200428609331t2003120t3.2.43.33.3.120034527kwh19.1%20043.3.213-13-12003kw1109.261109.260002238.272224.5013.78002382.141532.12605.45240.603.97993.40928.4964.91001386.58709.28676.1101.208109.656503.651360.25240.65.17220036530kw1358kw9.7%2217kw12%1568kw296023.2%782kw11%951kw22.6%20037200kw700kw20051103-23-22005110(MW%)123456789106703265989684006706266794795568366585818837096942788.384.189.490.090.289.289.389.589.990.1378981219201819200020023-320023-13-320002002%12345678910111220000.90.830.860.830.840.910.981.000.960.870.880.9420010.7890.7890.7740.7490.8010.8961.000.9490.8450.7980.8160.90920020.7470.6990.7430.7150.7260.8190.9731.0000.9300.8000.8160.90030603-1200224h200387.3%20021.63.3.32003137.75kwh104kwh98.21kwh39.54kwh330kw2003199.1kwh60%200331kwh200360kwh200200.20.40.60.811.2135791113151719212331603.43.4.13.4.230%matlab3-22008001400200032603-2-+−=+−=−−=21875.96875.209375.0375.1208333.10416666.0777777.35.00555555.0222XXYXXYXXY3.4.33.4.42004200312633603.51234563.6346