資源描述:
《基于改進(jìn)粒子群算法的無(wú)功優(yōu)化研究》由會(huì)員上傳分享,免費(fèi)在線閱讀,更多相關(guān)內(nèi)容在教育資源-天天文庫(kù)。
1、東北電力大學(xué)碩士學(xué)位論文基于改進(jìn)粒子群算法的無(wú)功優(yōu)化研究姓名:張冰冰申請(qǐng)學(xué)位級(jí)別:碩士專業(yè):電力系統(tǒng)及其自動(dòng)化指導(dǎo)教師:祝洪博20070301摘要隨著電力工業(yè)向“大系統(tǒng)、超高壓、遠(yuǎn)距離、大容量”發(fā)展,電網(wǎng)結(jié)構(gòu)日趨復(fù)雜,系統(tǒng)穩(wěn)定性問(wèn)題日益突出。通過(guò)無(wú)功優(yōu)化來(lái)改善電壓質(zhì)量、降低運(yùn)行網(wǎng)損以及提高系統(tǒng)運(yùn)行的安全穩(wěn)定性顯得尤為必要。我國(guó)電力市場(chǎng)的逐步建立與完善,傳統(tǒng)的集發(fā)電、輸電和配電為一體的服務(wù)結(jié)構(gòu)必將被打破。因此,對(duì)電力市場(chǎng)環(huán)境下無(wú)功優(yōu)化的研究勢(shì)在必行。本文對(duì)無(wú)功優(yōu)化模型及相關(guān)求解算法進(jìn)行了深入研究,分析了各種優(yōu)化算法的優(yōu)缺點(diǎn)和適用范圍。在廠
2、網(wǎng)分開(kāi),競(jìng)價(jià)上網(wǎng)的市場(chǎng)模式下,綜合考慮電力系統(tǒng)的安全約束,以電網(wǎng)公司向發(fā)電廠購(gòu)買有功費(fèi)用、無(wú)功費(fèi)用最小為目標(biāo)函數(shù)建立了電網(wǎng)無(wú)功優(yōu)化的數(shù)學(xué)模型。本文采用了VB6.0程序設(shè)計(jì)語(yǔ)言,應(yīng)用改進(jìn)粒子群優(yōu)化MPSO(ModifiedParticleSwarmOptimization)算法求解所建無(wú)功優(yōu)化模型,并結(jié)合動(dòng)態(tài)調(diào)整罰函數(shù)法,將無(wú)功優(yōu)化問(wèn)題轉(zhuǎn)化成一個(gè)無(wú)約束求極值問(wèn)題,有效提高了MPSO算法的全局收斂能力和計(jì)算精度。以IEEE30節(jié)點(diǎn)和遼源電網(wǎng)為例進(jìn)行了仿真計(jì)算,取得了較為滿意的結(jié)果,從而驗(yàn)證了本文模型和算法的正確性、適用性和經(jīng)濟(jì)性。將研究結(jié)果
3、應(yīng)用到遼源電網(wǎng),表明該模型既克服了現(xiàn)行無(wú)功優(yōu)化在電力市場(chǎng)條件下的不足,又使電網(wǎng)公司為購(gòu)買有功和無(wú)功的費(fèi)用支出最小,從而獲得最大的經(jīng)濟(jì)效益。為將來(lái)進(jìn)一步在現(xiàn)場(chǎng)實(shí)施電力市場(chǎng)下的無(wú)功優(yōu)化控制方案提供了重要的依據(jù),因此具有重要意義。關(guān)鍵詞電力市場(chǎng):無(wú)功優(yōu)化:改進(jìn)粒子群算法;罰函數(shù)AbstractWiththedevelopmentofpowerindustry,themodempowersystemhasbecomemoretremendous,itsframeworkandmodesofoperationgettobemorecomplexi
4、ntheextreme,andthesecurityandstabilityofpowersystemhavebecomemof營(yíng)tremendous.Reactivepoweroptimizationfro'o)inpowersystemisoneofthemostimportantmeasurementstoimprovethevoltagequality,reducethenetreallossandensurepowersystemsecurelyandstably.Astheestablishmentandimprovemen
5、tofelectricpowermarketofOurcountry,theelectricpowerenterprisemustbreaktheconventionalelectricpowerserviceconstructionintegratedwithgenerateelectricity,powertransmissionandpowerdistribution.SothestudyonRPOisquiteimperativeunderthepowermarketenvironment.Inthispaper,themath
6、ematicalmodelsofVaroptimizationproblemarefullystudiedaswellasitsalgorithm.a(chǎn)ndallkindsofmethodsforsolvingtheproblemaresummarized,What’smore,thedifferencesandfeaturesofthesemethodsaregiveninthepaper.Underthebackgroundofputtingforward‘‘separationofpowerplantfromelectricnetw
7、ork,connectiontothenationalnetworkbypricecompetition'’,andconsideringthesecureandpracticaloperationofpowersystem,amathematicmodelisdesignedinwhichthecostofactivepowerlossandcostofreactivepoweraretakenaSobjectivefunctionwithitsallkindsofconstrains.Inthispaper,MPSOalgorith
8、misappliedtoreactivepoweroptimizationmodelanditsprogramisdesignedwithVB6.0language.Incorporationofnon-s