基于专家经验的电力系统无功优化进化规划算法研究

重庆大学硕士学位论文基于专家经验的电力系统无功优化进化规划算法研究姓名：程彬申请学位级别：硕士专业：电气工程指导教师：颜伟20060501IIIAbstractReactivepoweroptimizationwhichcanimprovethequalityofsystemvoltageandreducereactivepowerlosshasnotbeensolvedcompletelybecauseofthecomplexityofmixed-integernonlinearprogramming.Optimizedroutingalgorithmoffeasiblepowerflowisproposedinthispapercombiningwithevolutionaryprogrammingandthebasicprincipleofsystemvoltagereactivepoweroptimization.Thismethodisheuristicandgetsgreatefficiency.Itanalyzesthefactorssuchasthepartitionofallthesamelayer,netpowerflowanddistributionleveltooptimizedroutingadjustmentofreactivepowerandvoltageregulationofallpowerplantsandsubstations,includinglocalandglobe.Itmakespowerflowavailableandmeettheneedofthereactivepowerbalancingandthereversevoltageregulationbycombiningthereactivepowerbalancingprincipleofvoltage-grading&district-dividingandlocalreactivepowercompensationwithfeasiblepowerflowregulation,thereversevoltageregulationofcentralpointvoltage.Fewreactivepowervoltageregulationsandpowerflowcalculationswiththisalgorithmcanensuretheresultsofpowerflowandreductionloss.Theefficiencyofcontrollingactualsystemvoltageandreactivepoweroptimizedcontrolandoperationisprovedbysimulationexamplesinthispaper.Anewreactivepoweroptimizedevolutionaryprogrammingalgorithmhasbeenproposedthroughtheheuristicexperiencesofreactivepoweroptimizedvoltageregulationandadaptiveevolutionaryprogrammingalgorithm.Withtheseexperiencesthealgorithmcancontrolmutationdirectionofreactivepoweroptimizedmultivariableandadjustmentofrelativelocalandglobalcontrolmultivariablebyanalyzingthenon-feasiblereasonofindividualpowerflow.Finallythemutationefficiencyofnon-feasibleindividualcontrolmultivariablearebeingimproved.adaptivemutationalgorithmhasbeenadoptedtoensuretheglobalrandomsearchingspecialityforfeasibleindividual.Theglobalrandomsearchingabilityofadaptiveevolutionaryprogrammingalgorithmandhighevolutionaryabilitybythemethodinthispaperisillustratedbytheresultofactualpowersystems.Keyword:powersystem,reactivepoweoptimization,specialistexperiences,evolutionaryprogramming1111.1197812191019828419877232003814PRQXUU+∆=1.1(1.1)RjX+UU∆PQRUQPPj222+=∆1.21.2PQRP∆RRP∆RP∆PQP∆QP∆R12[14]131.21962J.CarpentierOptimalPowerFlow-OPF[1]OPFOPFOPFOPFOPFOPF1.2.1:min(,)..(,)0(,)0fuxstguxhux=≤1.31.3uxuPQPVx[2]()()22,1minmin2cosniijijijkijkfGuuuuδδ==+−−∑1.4[2]:1minminspecnjjspecjjuufu=−=∆∑1.514..st11minmaxminmaxminminminmaxminmax(cossin)0(sincos)0niijijijijijjniijijijijijjGiGiGiCiCiCiiiiijijijiiiPUUGBQUUGBQQQQQQUUUTTTδδδδδδδ==−+=−+=≤≤≤≤≤≤≤≤≤≤∑∑1.6iPiQimaxGiQminGiQmaxCiQminCiQiimaxiUminiUimaxiTminiTi:PQ−[3]1min()GSlosslossQkGkkfPfQλ==+∑1.7lossλlossP()QkGkfQk[4]min()()()GCgpiGigqiGicjCjiNjNfCPCQCQ∈∈=++∑∑1.8GNCN()gpiGiCPi()gqiGiCQi()cjCjCQj1.2.2151)1968DommelTinneyRG[9]RGCarpentierAbadieGRG[9]2)[5]80[6][7][8]ab163)1973Sasson[1]1984D.I.SunLagrangeOPF1987MariaFindlayLP[4]1994[5]1996[6]OPF4)Karmarkar1984[13]-Primal-DualAffineScalingAlgorithm--[10][14]NP-hardNP-hardNP-hard1)[19]17GeneticAlgorithmGA(EvolutionaryProgrammingEP)(EvolutionarystrartgiesES)aGABagleyJ.D1967,1975MichiganJ.H.Holland[15]HollandSimpleGeneticAlgorithm,SGA[16]bEP[17]FogelLJ2060EPGAEPEP[1853~54]cESRechenbergI2060ES2)ACOPSOAFSAa[26]M.Dorigopheromone18b[27]KennedyEberhart1995PSOPSOPSO[28]CECc[29]3)(SA)(SimulatedAnnealingSA)Metropolis19531983KirkpatrickSASAabc19[20]SA4)(TS)[21]F.Glover60TSTLTSabcTLd[22]TSIEEE30125SGASAGATS5)(CO)[23][24]Logistic[25]1101.321122.1//2.22122.3110kVβ2.1∑∑==∆∆=NLiCiNLiLiQQ11β2.1NLLiQ∆iCiQ∆iLiQ∆CiQ∆2132.42.4.1220kV220kV500kV220kVrefUmaxminmin1.21+0.051.211.211refrefrefrefUifUUifUifββββ≥−=×−≤2.2maxrefUminrefU[38][39]500kV330kV220kV1~1.1pu110kV35kV0.97~1.07puUmaxUminmaxrefUminrefUmaxmaxmaxminminminmaxmin()/4()/4refrefUUUUUUUU=−−=+−2.3110kV220kV100%1.2β≥80~100%1.21β≥80%2141β22.1UG0U01k2.1Fig2.1theelectricdiagramofpowerplantsUG0U0kUref0()/GrefUUUk∆=−2.42.22.3UHUMULkHkMkLUMref300maxmaxminminmaxmin()MMrefHHMHMMrefHHHHHHHHHHHHHHHHHHUUkkroundkUUUkkkifkkkkkkifkkkkifkkkkk−∆=∆×∆−∆−∆=−∆−∆−∆−2.50Hk∆round.UHUMULQCPD+jQDkH11kM1kLZHZMZLY0UHUMUL2.22.3Fig2.2theelectricdiagramofsubstationFig2.3theequivalentcircuitofsubstation21500maxmaxminminmaxmin()MrefMMMHHMMMMMMMMMMMMMMMMMMUUkkroundkUkkkifkkkkkkifkkkkifkkkkk−∆=∆×∆−∆−∆=−∆−∆−∆−2.60Mk∆2.4.2[38][39]2.12.1Table2.1TherangeofpowerfactorfortheprimaryandsecondarysideofsubstationkV500330220110350.98~10.98~10.9~10.98~10.95~10.9~1220kV500330kV0.98220kV12.12.7216max1maxmin1maxminmin1minminmax1&max((),)1&max((),)1&0min((),)ccDDcDccrefDCcDCDCcQifQQRDQQifQQQRDQQifQRDQQifβββ≥≤≥=−−1minminmin0000min1max1&0()()()()t()DCDCDCCCDDCCDCDDQQRDQQroundQQRDQQroundQQPgQβϕ≤=∆×∆=∆×∆=−2.7maxminccQQDPDQ1maxϕ2.10CQ∆min(,)ABmax(,)ABAB1β2.711β≥11βQDCmin00CQ2.8CcrefCQQQ∆=−2.8CrefQ22max1222max11max1211max12(())1&1(()())1&1(())1&TiTiTiTiTiCrefTiTiRDPtgQifRDPtgQPtgifQRDQPtgifϕββϕϕββϕββ×+∆≥≥×+∆−×≥=∆−×121()1&1TiRDQifββ∆≥2.92.91maxϕ2maxϕ2.1TiQ∆1TiP2TiP1β2β(.)RD2.7CQ217maxmaxminminmaxminCCCrefCCCCCrefCCrefCCCrefCQQifQQQQQifQQQQifQQQ−≥