




版權(quán)說明:本文檔由用戶提供并上傳,收益歸屬內(nèi)容提供方,若內(nèi)容存在侵權(quán),請進行舉報或認領(lǐng)
文檔簡介
中英文資料PAGE29中英文資料外文翻譯文獻基于改進的灰色預(yù)測模型的電力負荷預(yù)測[摘要]盡管灰色預(yù)測模型已經(jīng)被成功地運用在很多領(lǐng)域,但是文獻顯示其性能仍能被提高。為此,本文為短期負荷預(yù)測提出了一個GM(1,1)—關(guān)于改進的遺傳算法(GM(1,1)-IGA)。由于傳統(tǒng)的GM(1,1)預(yù)測模型是不準確的而且參數(shù)的值是恒定的,為了解決這個問題并提高短期負荷預(yù)測的準確性,改進的十進制編碼遺傳算法(GA)適用于探求灰色模型GM(1,1)的最佳值。并且,本文還提出了單點線性算術(shù)交叉法,它能極大地改善交叉和變異的速度。最后,用一個日負荷預(yù)測的例子來比較GM(1,1)-IGA模型和傳統(tǒng)的GM(1,1)模型,結(jié)果顯示GM(1,1)-IGA擁有更好地準確性和實用性。關(guān)鍵詞:短期的負荷預(yù)測,灰色系統(tǒng),遺傳算法,單點線性算術(shù)交叉法第一章緒論日峰值負荷預(yù)測對電力系統(tǒng)的經(jīng)濟,可靠和安全戰(zhàn)略都起著非常重要的作用。特別是用于每日用電量的短期負荷預(yù)測(STLF)決定著發(fā)動機運行,維修,功率互換和發(fā)電和配電任務(wù)的調(diào)度。短期負荷預(yù)測(STLF)旨在預(yù)測數(shù)分鐘,數(shù)小時,數(shù)天或者數(shù)周時期內(nèi)的電力負荷。從一個小時到數(shù)天以上不等時間范圍的短期負荷預(yù)測的準確性對每一個電力單位的運行效率有著重要的影響,因為許多運行決策,比如:合理的發(fā)電量計劃,發(fā)動機運行,燃料采購計劃表,還有系統(tǒng)安全評估,都是依據(jù)這些預(yù)測。傳統(tǒng)的負荷預(yù)測模型被分為時間序列模型和回歸模型。通常,這些模型對于日常的短期負荷預(yù)測是有效的,但是對于那些特別的日子就會產(chǎn)生不準確的結(jié)果。此外,由于它們的復雜性,為了獲得比較滿意的結(jié)果需要大量的計算工作。灰色系統(tǒng)理論最早是由鄧聚龍?zhí)岢鰜淼?,主要是模型的不確定性和信息不完整的分析,對系統(tǒng)研究條件的分析,預(yù)測以及決策?;疑到y(tǒng)讓每一個隨機變量作為一個在某一特定范圍內(nèi)變化的灰色量。它不依賴于統(tǒng)計學方法來處理灰色量。它直接處理原始數(shù)據(jù),來尋找數(shù)據(jù)內(nèi)在的規(guī)律?;疑A(yù)測模型運用灰色系統(tǒng)理論的基本部分。此外,灰色預(yù)測可以說是利用介于白色系統(tǒng)和黑色系統(tǒng)之間的灰色系統(tǒng)來進行估計。信息完全已知的系統(tǒng)稱為白色系統(tǒng);相反地,信息完全未知的系統(tǒng)稱為黑色系統(tǒng)?;疑P虶M(1,1)(即一階單變量灰色模型)是灰色理論預(yù)測中主要的模型,由少量數(shù)據(jù)(4個或更多)建立,仍然可以得到很好地預(yù)測結(jié)果。灰色預(yù)測模型組成部分是灰色微分方程組——特性參數(shù)變化的非常態(tài)微分方程組,或者灰色差分方程組——結(jié)構(gòu)變化的非常態(tài)差分方程組,而不是一階微分方程組或者常規(guī)情況下的差分方程組。灰色模型GM(1,1)有一個參數(shù),它在很多文章里經(jīng)常被設(shè)為0.5,這個常數(shù)可能不是最理想的,因為不同的問題可能需要不同的值,否則可能產(chǎn)生錯誤的結(jié)果。為了修正前面提到的錯誤,本文嘗試用遺傳算法來估算值。JohnHolland首先描述了遺傳算法(GA),以一個抽象的生物進化來提出它們,并且給出了一個理論的數(shù)學框架作為歸化。一個遺傳算法相對于其他函數(shù)優(yōu)化方法的顯著特征是尋找一個最佳的解決方案來著手,此方案不是以一個單一逐次改變的結(jié)構(gòu),而是給出一組使用遺傳算子來建立新結(jié)構(gòu)的解決措施。通常,二進制表示法應(yīng)用于許多優(yōu)化問題,但是本文的遺傳算法(GA)采用改進的十進制編碼表示方案。本文打算用改進的遺傳算法(GM(1,1)-IGA)來解決電力系統(tǒng)中短期負荷預(yù)測(STLF)中遇到的問題。傳統(tǒng)的GM(1,1)預(yù)測模型經(jīng)常設(shè)定參數(shù)為0.5,因此背景值可能不準確。為了克服以上弊端,用改進的十進制編碼的遺傳算法來獲得理想的參數(shù)值,從而得到較準確的背景值。而且,提出了單點線性算術(shù)交叉法。它能極大地改善交叉和變異的速度,使提出的GM(1,1)-IGA能更準確地預(yù)測短期日負荷。本文結(jié)構(gòu)如下:第二章介紹灰色預(yù)測模型GM(1,1);第三章用改進的遺傳算法來估算;第四章提出了GM(1,1)-IGA來實現(xiàn)短期日負荷預(yù)測;最后,第五章得出結(jié)論。第二章灰色預(yù)測模型GM(1,1)本章重點介紹灰色預(yù)測的機理?;疑P虶M(1,1)是時間序列預(yù)測模型,它有3個基本步驟:(1)累加生成,(2)累減生成,(3)灰色建模?;疑A(yù)測模型利用累加的原理來創(chuàng)建微分方程。本質(zhì)上講,它的特點是需要很少的數(shù)據(jù)。灰色模型GM(1,1),例如:單變量一階灰色模型,總結(jié)如下:第一步:記原始數(shù)列:=是n階離散序列。是m次時間序列,但m必須大于等于4。在原始序列的基礎(chǔ)上,通過累加的過程形成了一個新的序列。而累加的目的是提供構(gòu)建模型的中間數(shù)據(jù)和減弱變化趨勢。定義如下:有,則是r次累加序列。第二步:設(shè)定值來預(yù)測通過GM(1,1),我們可以建立下面的一階灰色微分方程:它的差分方程是。a稱為發(fā)展系數(shù),b稱為控制變量。以微分的形式表示導數(shù)項,我們可以得到:在一個灰色GM(1,1)模型建立前,一個適當?shù)闹敌枰o出以得到一個好的背景值。背景值序列定義如下:其中,為方便起見,值一般被設(shè)為0.5,推導如下:然而,這個常量可能不是最理想的,因為不同的場合可能需要不同的值。而且,不管是發(fā)展系數(shù)a還是控制變量b都由值確定。由于系數(shù)是常量,原始灰色信息的白化過程可能被抑制。因此,GM(1,1)模型中預(yù)測值的準確性將會嚴重的降低。為了修正以上不足,系數(shù)必須是基于問題特征的變量,因此我們用遺傳算法來估算值。第三步:構(gòu)建累加矩陣B和系數(shù)向量。應(yīng)用普通最小二乘法(OLS)來獲得發(fā)展系數(shù)a,b。如下:于是有第四步:獲得一階灰色微分方程的離散形式,如下:解得為為第三章運用改進GA估算值為了預(yù)測出準確的灰色模型GM(1,1),殘差校驗是必不可少的。因此,本文中所提出的目標函數(shù)的方法可以確保預(yù)測值誤差是最小。目標函數(shù)定義為最小平均絕對百分比誤差,如下:且,為原始數(shù)據(jù),為預(yù)測值,n是該數(shù)列的維數(shù)。從上面描述構(gòu)建的GM(1,1),我們可以得到:在GM(1,1)中參數(shù)的值能夠決定的值;不管是發(fā)展系數(shù)a還是控制變量b都由值確定。更重要的是,的結(jié)果由a,b決定,因此整個模型選擇過程最重要的部分就是的值。在和殘差之間有著某些復雜的非線性關(guān)系,這些非線性是很難通過解析來解決的,因此選擇最理想的值是GM(1,1)的難點。遺傳算法是一個隨機搜索算法,模擬自然選擇與演化。它能廣泛應(yīng)用正是基于后面兩個基本方面:計算代碼非常簡單并且還提供了一個強大的搜索機制。它們函數(shù)相對獨立,意味著它們不會被函數(shù)的屬性所限制,例如:連續(xù)性,導數(shù)的存在,等等。盡管二進制法經(jīng)常應(yīng)用于許多優(yōu)化問題,但是在本文我們采用改進十進制編碼法方案來解決。在數(shù)值函數(shù)優(yōu)化方面,改進的十進制編碼法相對于二進制編碼法擁有很大的優(yōu)勢。這些優(yōu)勢簡要的敘述如下:第一步:GA的效率提高了,因此,沒有必要將染色體轉(zhuǎn)換為二進制類型。第二步:由于有效的內(nèi)部電腦浮點表示,需要較少的內(nèi)存。第三步:甄別二進制或其它值不會使精度降低,并且有更大的自由來使用不同的遺傳算子。我們利用改進的十進制碼代表性方法來尋找在灰色GM(1,1)模型中最佳系數(shù)的值。本文中,我們提出單點線性算術(shù)交叉法,并且利用它來獲得值;它能極大地提高交叉和變異的速度。改進的十進制碼代表性方法的步驟如下:(1)編碼:假設(shè)是二進制字符串的C位,然后由右至左每隔n位轉(zhuǎn)換為十進制。(n<C,n和C的值要確保精度)(2)隨機化種群:選擇一個整數(shù)M作為種族的大小,然后隨機地從集合選擇M點,如,這些點組成個體的原始種群,該序列被定義為:(3)評估適應(yīng)度:在選擇的過程中,個體被選擇參與新個體的繁殖。擁有高度地適應(yīng)度F()的個體逐代衍化和發(fā)展。適應(yīng)度函數(shù)是是從個體獲得的預(yù)測值。是迭代最小二乘總和的最大值。第四步:選擇:在本文中,我們根據(jù)它們的適應(yīng)度函數(shù)分別地計算出個體選定的概率,然后我們通過輪盤選擇法,使繁殖的各自概率是,最后我們拿原始的個體來生成下一代的。第五步:交叉和變異:編碼和交叉是相關(guān)的;我們利用了十進制碼表示法,因此我們提出了一種新的交叉算子“單點線性算術(shù)交叉”。1)選擇合適的兩個有交叉概率的個體。2)為這兩個選擇的個體,我們?nèi)匀徊捎秒S機抽樣方法以得到交叉算子。例如:3)交叉①互相交換它們的正確的字符串。②位在左側(cè)的交叉可以通過以下計算算法:a:基因分析:b:交換后基因:稱為交叉系數(shù),每次根據(jù)隨機的交叉系統(tǒng)來選擇。4)變異:下面是一個新的變異方案:當變異算子被選擇,新的基因值是一個在域權(quán)重的隨機數(shù),它是用原始基因值得到的加權(quán)總和。如果變異算子的值是,變異值是:是變異系數(shù),。r是一個隨機數(shù),。每當進行變異操作時,r會被隨機的挑選。因此,新的后代可以通過交叉和變異操作來創(chuàng)建。第六步:推出原則:選擇當前的一代個體來繁殖下一代個體,然后求出適應(yīng)度值并判斷算法是否符合退出條件。如果符合條件,這個值就是最佳的,否則回到第四步,直到種群內(nèi)所有個體達到統(tǒng)一標準或幾代個體的數(shù)量超過最大值100。第四章.負荷預(yù)測案例在本章,我們試著對GM(1,1)-關(guān)于改進的遺傳算法進行性能評估。第一步:m天的日負荷數(shù)據(jù)序列定義為,我們測量了每個小時的電力負荷,于是負荷序列向量就是一個24維數(shù)據(jù)。1點:2點:j點:24點:式中m是所建模型的天數(shù),是日負荷數(shù)據(jù)序列的第j點。圖1.原始數(shù)據(jù)和預(yù)測值第二步:我們利用改進的遺傳算法為各自的負荷數(shù)據(jù)序列來選擇值。接著,我們可以算出a和b,然后我們利用GM(1,1)-IGA來預(yù)測第m+1天中的第j點的負荷,于是我們可以得到,最后第m+1天地24個預(yù)測值構(gòu)成了這個負荷數(shù)據(jù)序列。這有一個GM(1,1)-關(guān)于改進的遺傳算法(GM(1,1)-IGA)的例子,兩種預(yù)測日負荷數(shù)據(jù)曲線(7月26號)和原始的日負荷曲線同時在圖1中畫出。第三步:我們可以利用GM(1,1)-遺傳算法的四個指標來檢驗精度,包括相對誤差,均方差率,小誤差概率和關(guān)聯(lián)度誤差。如果相對誤差和均方差率較低,或者小誤差概率和關(guān)聯(lián)度誤差較大,GM(1,1)-GA的準確性檢驗是較好的。設(shè)置模擬殘差為,k=1,2,…,n設(shè)置模擬的相對剩余為k=1,2,…,n設(shè)置平均值為設(shè)置的方差為設(shè)置殘差平均值為設(shè)置殘差方差為因此,GM(1,1)-IGA的校驗值如下:1).平均相對誤差為2).均方差率為3).小誤差概率為4).關(guān)聯(lián)度為其中,根據(jù)上述公式,GM(1,1)-IGA的指標的校驗值見表1。表1GM-IGA和GM的四個指標GM-GAGM平均相對誤差0.0000900.0001均方差率0.00390.0073小誤差概率10.92關(guān)聯(lián)度0.980.90通過表1可以看出,GM-GA所以指標的精確度都是一級的,因此這個GM(1,1)-IGA可以被用來預(yù)測短期負荷。第四步:在圖1中,我們可以得到GM(1,1)-IGA的預(yù)測負荷數(shù)據(jù)曲線比GM(1,1)的曲線更接近于原始的日負荷數(shù)據(jù)曲線。進一步分析,本文選擇相對誤差作為標準來評價兩種模式。兩種模型的偏差值如下,GM(1,1)的平均誤差為2.285%,然而,GM(1,1)-IGA的平均誤差為0.914%。第五章.結(jié)論本文提出了GM(1,1)-關(guān)于改進的遺傳算法(GM(1,1)-IGA)來進行短期負荷預(yù)測。采用十進制編碼代表性方案,改進的遺傳算法用于獲得GM(1,1)模型中的最優(yōu)值。本文也提出了單點線性算術(shù)交叉法,它能極大地提高交叉和變異的速度,因此GM(1,1)-IGA可以準確地預(yù)測短期日負荷。GM(1,1)-IGA的特點是簡單、易于開發(fā),因此,它在電力系統(tǒng)中作為一個輔助工具來解決預(yù)測問題是適宜的。圖2.GM(1,1)的偏差值圖3.GM(1,1)-IGA的偏差值致謝這項工作是由國家自然科學基金部分支持。(70671039)參考文獻[1]P.GuptaandK.Yamada,“AdaptiveShort-TermLoadForecastingofHourlyLoadsUsingWeatherInformation,”IEEETr.OnPowerApparatusandSystems.VolPas-91,pp2085-2094,1972.[2]D.W.Bunn,E.D.Farmer,“ComparativeModelsforElectricalLoadForecasting”.JohnWiley&Son,1985,NewYork.[3]AbdolhosienS.Dehdashti,JamesRTudor,MichaelC.Smith,“ForecastingOfHourlyLoadByPatternRecognition-ADeterministicApproach,”IEEETr.OnPowerApparatusandSystems,Vol.AS-101,No.9Sept1982.[4]S.RahrnanandRBhamagar,“AnexpertSystemBasedAlgorithmforShort-TermLoadForecast,”IEEETr.OnPowerSystems,Vol.AS-101,No.9Sept.1982[5]M.T.Hagan,andS.M.Behr,“TimeSeriesApproachtoShort-TermLoadForecasting,”IEEETrans.onPowerSystem,Vol.2,No.3,pp.785-791,1987.[6]XieNaiming,LiuSifeng.“ResearchonDiscreteGreyModelandItsMechanism”.IEEETr.System,ManandCybernetics,Vol1,2005,pp:606-610[7]J.L.Deng,“Controlproblemsofgreysystems,”SystemsandControlLetters,vol.1,no.5,pp.288-294,1982.[8]J.L.Deng,Introductiontogreysystemtheory,J.GreySyst.1(1)(1989)1–24[9]J.L.Deng,PropertiesofmultivariablegreymodelGM(1N),J.GreySyst.1(1)(1989)125–141.[10]J.L.Deng,Controlproblemsofgreysystems,Syst.ControlLett.1(1)(1989)288–294.[11]Y.P.Huang,C.C.Huang,C.H.Hung,Determinationofthepreferredfuzzyvariablesandapplicationstothepredictioncontrolbythegreymodelling,TheSecondNationalConferenceonFuzzyTheoryandApplication,Taiwan,1994,pp.406–409.[12]S0aeroandMRIrving,“AGeneticAlgorithmForGeneratorSchedulingInPowerSystems,”IEEETr.ElectricalPower&EnergySystems,Vol18.No1,pp19-261996.[13]Edmund,T.H.HengDiptiSrinivasanA.C.Liew.“ShortTermLoadForecastingUsingGeneticAlgorithmAndNeuralNetworks”.IEEECatalogueNo:98EX137pp576-581[14]Chew,J.M.,Lin,Y.H.,andChen,J.Y.,"TheGreyPredictorControlinInvertedPendulumSystem",JournalofChinaInstituteofTechnologyandCommerce,Vol.11,pp.17-26,1995[15]J.GreySyst.,“Introductiontogreysystemtheory,”vol.1,no.1,pp.1–24,1989ApplicationofImprovedGreyPredictionModelforPowerLoadForecasting[Abstract]Althoughthegreyforecastingmodelhasbeensuccessfullyutilizedinmanyfields,literaturesshowitsperformancestillcouldbeimproved.Forthispurpose,thispaperputforwardaGM(1,1)-connectionimprovedgeneticalgorithm(GM(1,1)-IGA)forshort-termloadforecasting(STLF).WhileTraditionalGM(1,1)forecastingmodelisnotaccurateandthevalueofparameterisconstant,inordertosolvethisproblemandenhancetheaccuracyofshort-termloadforecasting(STLF),theimproveddecimal-codegeneticalgorithm(GA)isappliedtosearchtheoptimalvalueofgreymodelGM(1,1).What’smore,thispaperalsoproposestheone-pointlinearityarithmeticalcrossover,whichcangreatlyimprovethespeedofcrossoverandmutation.Finally,adailyloadforecastingexampleisusedtotesttheGM(1,1)-IGAmodelandtraditionalGM(1,1)model,resultsshowthattheGM(1,1)-IGAhadbetteraccuracyandpracticality.Keywords:Short-termLoadForecasting,GreySystem,GeneticAlgorithm,One-pointLinearityArithmeticalCrossover.IntroductionDailypeakloadforecastingplaysanimportantroleinallaspectsofeconomic,reliable,andsecurestrategiesforpowersystem.Specifically,theshort-termloadforecasting(STLF)ofdailyelectricityusageiscrucialinunitcommitment,maintenance,powerinterchangeandtaskschedulingofbothpowergenerationanddistributionfacilities.Short-termloadforecasting(STLF)aimsatpredictingelectricloadsforaperiodofminutes,hours,daysorweeks.Thequalityoftheshort-termloadforecastswithleadtimesrangingfromonehourtoseveraldaysaheadhasasignificantimpactontheefficiencyofoperationofanypowerutility,becausemanyoperationaldecisions,suchaseconomicdispatchschedulingofthegeneratingcapacity,unitcommitment,schedulingoffuelpurchaseaswellassystemsecurityassessmentarebasedonsuchforecasts[1].Traditionalshort-termloadforecastingmodelscanbeclassifiedastimeseriesmodelsorregressionmodels[2,3,4].Usually,thesetechniquesareeffectivefortheforecastingofshort-termloadonnormaldaysbutfailtoyieldgoodresultsonthosedayswithspecialevents[5,6,7].Furthermore,becauseoftheircomplexities,enormouscomputationaleffortsarerequiredtoproduceacceptableresults.Thegreysystemtheory,originallypresentedbyDeng[8,9,10],focusesonmodeluncertaintyandinformationinsufficiencyinanalyzingandunderstandingsystemsviaresearchonconditionalanalysis,forecastinganddecisionmaking.Thegreysystemputseachstochasticvariableasagreyquantitythatchangeswithinagivenrange.Itdoesnotrelyonstatisticalmethodtodealwiththegreyquantity.Itdealsdirectlywiththeoriginaldata,andsearchestheintrinsicregularityofdata[11].Thegreyforecastingmodelutilisestheessentialpartofthegreysystemtheory.Therewith,greyforecastingcanbesaidtodefinetheestimationdonebytheuseofagreysystem,whichisinbetweenawhitesystemandablack-boxsystem.Asystemisdefinedasawhiteoneiftheinformationinitisknown;otherwise,asystemwillbeablackboxifnothinginitisclear.ThegreymodelGM(1,1)isthemainmodelofgreytheoryofprediction,i.e.asinglevariablefirstordergreymodel,whichiscreatedwithfewdata(fourormore)andstillwecangetfineforecastingresult[12].Thegreyforecastingmodelsaregivenbygreydifferentialequations,whicharegroupsofabnormaldifferentialequationswithvariationsinbehaviorparameters,orgreydifferenceequationswhicharegroupsofabnormaldifferenceequationswithvariationsinstructure,ratherthanthefirst-orderdifferentialequationsorthedifferenceequationsinconventionalcases[13].ThegreymodelGM(1,l)hasparameterwhichwasoftensetto0.5inmanyarticles,andthisconstantmightnotbeoptimal,becausedifferentquestionsmightneeddifferentvalue,whichproduceswrongresults.Inordertocorrecttheabove-mentioneddefect,thispaperattemptstoestimatebygeneticalgorithms.Geneticalgorithms(GA)werefirstlydescribedbyJohnHolland,whopresentedthemasanabstractionofbiologicalevolutionandgaveatheoreticalmathematicalframeworkforadaptation[14].ThedistinguishingfeatureofaGAwithrespecttootherfunctionoptimizationtechniquesisthatthesearchtowardsanoptimumsolutionproceedsnotbyincrementalchangestoasinglestructurebutbymaintainingapopulationofsolutionsfromwhichnewstructuresarecreatedusinggeneticoperators[15].Usually,thebinaryrepresentationwasappliedtomanyoptimizationproblems,butinthispapergeneticalgorithms(GA)adoptedimproveddecimal-coderepresentationscheme.ThispaperproposedGM(1,1)-improvedgeneticalgorithm(GM(1,1)-IGA)tosolveshort-termloadforecasting(STLF)problemsinpowersystem.ThetraditionalGM(1,1)forecastingmodeloftensetsthecoefficientto0.5,whichisthereasonwhythebackgroundvaluez(1)(k)maybeunsuitable.Inordertoovercometheabove-mentioneddrawbacks,theimproveddecimal-codegeneticalgorithmwasusedtoobtaintheoptimalcoefficientvaluetosetproperbackgroundvaluez(1)(k).Whatismore,theone-pointlinearityarithmeticalcrossoverwasputforward,whichcangreatlyimprovethespeedofcrossoverandmutationsothattheproposedGM(1,1)-IGAcanforecasttheshort-termdailyloadsuccessfully.Thepaperisorganizedasfollows:section2proposesthegreyforecastingmodelGM(1,1):section3presentsEstimatewithimprovedgeneticalgorithm:section4putsforwardashort-termdailyloadforecastingrealizedbyGM(1,1)-IGAandfinally,aconclusionisdrawninsection5.2.GreypredictionmodelGM(1,1)Thissectionreviewstheoperationofgreyforecastingindetails.ThegreymodelGM(1,1)isatimeseriesforecastingmodel.Ithasthreebasicoperations:(1)accumulatedgeneration,(2)inverseaccumulatedgeneration,and(3)greymodeling.Thegreyforecastingmodelusestheoperationsofaccumulatedtoconstructdifferentialequations.Intrinsicallyspeaking,ithasthecharacteristicsofrequiringlessdata.ThegreymodelGM(1,1),i.e.,asinglevariablefirst-ordergreymodel,issummarizedasfollows:Step1:Denotetheinitialtimesequenceby=x(0)isthegivendiscreten-th-dimensionalsequence.x(0)(m)isthetimeseriesdataattimem,nmustbeequaltoorlargerthan4.Onthebasisoftheinitialsequencex(0),anewsequencex(1)issetupthroughtheaccumulatedgeneratingoperationinordertoprovidethemiddlemessageofbuildingamodelandtoweakenthevariationtendency,sox(1)isdefinedas:Where,andandisthertimesaccumulatedseries.Step2:Tosetthevaluetofinez(1)(k)AccordingtoGM(1,1),wecanformthefollowingfirst-ordergreydifferentialequation:Anditsdifferenceequationis.WhereawascalledthedevelopingcoefficientofGM,andbwascalledthecontrolvariable.Denotingthedifferentialcoefficientsubentryintheformofdifference,wecanget:BeforeagreyGM(1,1)modelwassetup,apropervalueneededtobeassignedforabetterbackgroundvaluez(1)(k).Thesequenceofbackgroundvalueswasdefinedas:AmongthemForconvenience,thevaluewasoftensetto0.5,thez(1)(k)wasderivedas:However,thisconstantmightnotbeoptimalbecausethedifferentquestionsmightneeddifferentvalue.And,bothdevelopingcoefficientaandcontrolvariablebweredeterminedbythez(1)(k).Theprocessoftheoriginalgreyinformationforwhiteningmaybesuppressedresultedfromthecoefficientwasconstant.Hence,theaccuracyofpredictionvaluex?(0)(k)inGM(1,1)modelwouldseriouslybedecreased.Inordertocorrectthedefect,thecoefficientmustbeavariablebasedonthefeatureofproblems,soweestimatebygeneticalgorithms.Step3:ToconstructaccumulatedmatrixBandcoefficientvectorXn.ApplyingtheOrdinaryLeastSquare(OLS)methodobtainsthedevelopingcoefficienta,bwasasfollows:andSoStep4:Toobtainthediscreteformoffirst-ordergreydifferentialequation,asfollows:Thesolutionofx(1)isAndthesolutionofx(0)is3.EstimatewithimprovedGAInordertoestimatetheaccuracyofgreymodeGM(1,1),theresidualerrortestwasessential.Therefore,theobjectivefunctionoftheproposedmethodinthispaperwastoensurethattheforecastingvalueerrorswereminimum.Theobjectivefunctionwasdefinedasmeanabsolutepercentageerror(MAPE)minimizationasfollows:Where,x(0)(k)isoriginaldata,isforecastingvalue,nisthenumberofsequencedata.However,fromtheabovedescriptionoftheestablishmentofGM(1,1),wecanget:InGM(1,1),thevalueofparametercandeterminez(1),and,bothdevelopingcoefficientaandcontrolvariablebweredeterminedbythez(1)(k).Whatismore,thesolutionofx(0)wasdeterminedbyaandb,sothekeypartofthewholemodelselectingprocesswasthevalueof.Thereiskindofcomplicatednonlinearrelationshipbetweenandresidualerrors,andthisnonlinearitywashardtosolvebyresolution,sotheoptimalselectionofwasthedifficultpointofGM(1,1).Geneticalgorithmisarandomsearchalgorithmthatsimulatesnaturalselectionandevolution.Itisfindingwidespreadapplicationasaconsequenceoftwofundamentalaspects:thecomputationalcodeisverysimpleandyetprovidesapowerfulsearchmechanism.Theyarefunctionindependentwhichmeanstheyarenotlimitedbythepropertiesofthefunctionsuchascontinuity,existenceofderivatives,etc.Althoughthebinaryrepresentationwasusuallyappliedtomanyoptimizationproblems,inthispaper,weusedtheimproveddecimal-coderepresentationschemeforsolution.Theimproveddecimal-coderepresentationintheGAoffersanumberofadvantagesinnumericalfunctionoptimizationoverbinaryencoding.Theadvantagescanbebrieflydescribedasfollows:Step1:EfficiencyofGAisincreasedasthereisnoneedtoconvertchromosomestothebinarytype,Step2:Lessmemoryisrequiredasefficientfloating-pointinternalcomputerrepresentationscanbeuseddirectly,Step3:Thereisnolossinprecisionbydiscriminationtobinaryorothervalues,andthereisgreaterfreedomtousedifferentgeneticoperators.Weutilizedtheimproveddecimal-coderepresentationschemeforsearchingoptimalcoefficientvalueingreyGM(1,1)model.Inthispaper,weproposedone-pointlinearityarithmeticalcrossoverandutilizedittoselectthevalueof;itcangreatlyimprovethespeedofcrossoverandmutation.Thestepsoftheimproveddecimal-coderepresentationschemeareasfollows:(1)Coding:SupposeisabinarystringofCbits,thenleteverynbitstransformadecimalfromrighttoleft.(n<C,thevaluesofnandCareensuredbyprecision)(2)Randomizepopulation:SelectoneintegerMasthesizeofthepopulation,andthenselectMpointsstochasticallyfromtheset,as,thesepointscomposetheindividualsoftheoriginalpopulation,thesequenceisdefinedas:(3)Evaluatethefitness:Intheselectionstep,individualsarechosentoparticipateinthereproductionofnewindividuals.TheindividualwiththehighestfitnessF()hasthepriorityandadvancestothenextgeneration.Thefitnessfunctionisandisthevalueofforecastingwhichisgainedbytheindividual.isthemaximumofthesumofiterativesquares.Step4:Selection:Inthispaper,wecalculateindividualselectedprobabilityrespectivelyaccordingtotheirfitnessfunctions,thenweadopttheroulettewheelselectionscheme,sothatthepropagatedprobabilityofrespectiveindividualisp(k),afterthatwetaketheinbornindividualtocomposethenextgenerationp(k+1).Step5:CrossoverandMutation:Codingandcrossoverarecorrelative;weutilizedthedecimal-coderepresentation,soweproposeanewcrossoveroperator“one-pointlinearityarithmeticalcrossover”1)Selectthefittwoindividualswithprobabilityofcrossover.2)Forthetwoselectedindividuals,westilladopttherandomselectionmeanstoensurethecrossoveroperator.Forexample:3)crossover:①Weexchangetheirrightstringseachother.②Thebitontheleftofcrossovercanbecalculatedthroughthefollowingalgorithm:a:Geneanalysis:b:Exchangethebackgene:Theiscalledcrossovercoefficient,itischoseneachtimebyrandomcrossoveroperation.4)Mutation:Thereisanewmutationoperation:whenthemutationoperatorwaschosen,thenewgenevalueisthatarandomnumberwithinthedomainofweight,whichisoperatedintoaweightedsumwithoriginalgenevalue.IfthevalueofmutationoperatorisZi,themutationvalueis:Andisthemutationcoefficient,.risarandomnumber,.Itisselectedrandomlyeverytimewhenmutationoperationishappening.Therefore,thenewoffspringcanbecreatedthroughcrossoverandmutationoperations.Step6:Quitprinciple:Selecttheremainingindividualsinthecurrentgenerationtoreproducetheindividualsinthenextgeneration,thenevaluatethefitnessvalueandjudgewhetherthealgorithmfulfilsthequitcondition.Ifitiscertifiable,inthiscasethevalueisoptimalsolution,elserepeatfromStep4untilallindividualsinpopulationmeettheconvergencecriteriaorthenumberofgenerationsexceedsthemaximumof100.4.LoadpredictionexampleInthissection,wetrytoevaluatetheperformanceofGM(1,1)-connectionimprovedgeneticalgorithm.First:Thedailyloaddatasequencesofmdaysaredefinedas,wemeasuredthepowerloadeachhour,andtheloadsequencevectorisatwenty-four-dimensionaldata.01thetimeofday:02thetimeofday:jthetimeofday:24thetimeofday:Wheremisthenumberofmodelingdays,Xjisthedailyloaddatasequenceofthej-thtimeofday.Fig1.OriginaldataandforecastingvalueSecond:WeutilizeimprovedgeneticalgorithmtoselectthevalueofforrespectiveloaddatasequenceXj.Afterthat,wecancalculateaandb,thenweutilizeGM(1,1)-IGAtopredicttheloadforecastingofthej-thtimeofthe(m+1)-thday,sowecouldgetXj(m+1),andthetwenty-fourforecastingvaluesofthe(m+1)-thdaystructuretheloaddatasequence.TherewasanexampleofGM(1,1)-connectionimprovedgeneticalgorithm(GM(1,1)-IGA),boththetwoforecastingdailyloaddatacurves(July26)andtheoriginaldailyloaddatacurveweredrawnsimultaneouslyonFig1.Thirdly:WecanusefourindexesofthisGM(1,1)-GAtoverifytheprecise,includingoftherelativeerror,theratioofmeansquareerror,themicroerrorprobabilityandtherelevancedegree.TheaccuracyverificationofGM(1,1)-GAisbetteriftherelativeerrorandtheratioofmeansquareerrorislower,orthemicroerrorprobabilityandtherelevancedegreeislarger[16].Setthesimulatedresidualofx(0)(k)isk=1,2,…,nSetthesimulatedrelativeresidualisk=1,2,…,nSetthemeanofx(0)isSetthevarianceofx(0)is1SetthemeansofresidualerrorisSetthevarianceofresidualerrorisSothecheckvalueofthisGM(1,1)-GAisasfollowed:1).themeanrelativeerroris2).theratioofmeansquareerroris3.)themicroerrorprobabilityis4).therelevancedegreeisThereamong,Onthebasisofaboveformula,theindexesofverificationofGM(1,1)-GAandGMisinTable1.Accordingtotable1,theallprecisionindexesofGM-GAarefirstdegree,sothisGM(1,1)-GAcanbeusedtopredicttheshort-termload.Fourth:AtFig1,wecangetthattheforecastingloaddatacurveofGM(1,1)-GAwasmoreclosedtotheoriginaldailyloaddatacurvethanGM(1,1)’s.Forfurtheranalysis,thispaperselectsrelativeerrorsasacriteriontoevaluatethetwomodels.Theerrorfiguresoftwomodelsareasfollows,andtheaverageerrorofGM(1,1)was2.285%,otherwise,theaverageerrorofGM(1,1)-IGAwas0.914%.5.ConclusionThispaperproposesGM(1,1)–connectionimprovedgeneticalgorithm(GM(1,1)-IGA)forshort-termloadforecasting.Adoptingdecimal-coderepresentationscheme,theimprovedgeneticalgorithmisusedtoselecttheoptimizevalueofGM(1,1)model.Thepaperalsoputsforwardtheone-pointlinearityarithmeticalcrossoverwhichcangreatlyimprovethespeedofcrossoverandmutation,sothattheGM(1,1)-IGAcanforecasttheshort-termdailyloadsuccessfully.TheGM(1,1)-IGAischaracteristicofbeingsimpleandeasytodevelop,therefore,itisappropriateasanaidtooltosolvetheforecastingproblemsinpowersystem.Fig2.TheerrorsofGM(1,1)Fig3.TheerrorsofGM(1,1)-IGAAcknowledgementThisworkispartiallysupportedbytheNationalNaturalScienceFoundation(70671039)參考文獻[1]P.GuptaandK.Yamada,“AdaptiveShort-TermLoadForecastingofHourlyLoadsUsingWeatherInformation,”IEEETr.OnPowerApparatusandSystems.VolPas-91,pp2085-2094,1972.[2]D.W.Bunn,E.D.Farmer,“ComparativeModelsforElectricalLoadForecasting”.JohnWiley&Son,1985,NewYork.[3]AbdolhosienS.Dehdashti,JamesRTudor,MichaelC.Smith,“ForecastingOfHourlyLoadByPatternRecognition-ADeterministicApproach,”IEEETr.OnPowerApparatusandSystems,Vol.AS-101,No.9Sept1982.[4]S.RahrnanandRBhamagar,“AnexpertSystemBasedAlgorithmforShort-TermLoadForecast,”IEEETr.OnPowerSystems,Vol.AS-101,No.9Sept.1982[5]M.T.Hagan,andS.M.Behr,“TimeSeriesApproachtoShort-TermLoadForecasting,”IEEETrans.onPowerSystem,Vol.2,No.3,pp.785-791,1987.[6]XieNaiming,LiuSifeng.“ResearchonDiscreteGreyModelandItsMechanism”.IEEETr.System,ManandCybernetics,Vol1,2005,pp:606-610[7]J.L.Deng,“Controlproblemsofgreysystems,”SystemsandControlLetters,vol.1,no.5,pp.288-294,1982.[8]J.L.Deng,Introductiontogreysystemtheory,J.GreySyst.1(1)(1989)1–24[9]J.L.Deng,PropertiesofmultivariablegreymodelGM(1N),J.GreySyst.1(1)(1989)125–141.[10]J.L.Deng,Controlproblemsofgreysystems,Syst.ControlLett.1(1)(1989)288–294.[11]Y.P.Huang,C.C.Huang,C.H.Hung,Determinationofthepreferredfuzzyvariablesandapplicationstothepredictioncontrolbythegreymodelling,TheSecondNationalConferenceonFuzzyTheoryandApplication,Taiwan,1994,pp.406–409.[12]S0aeroandMRIrving,“AGeneticAlgorithmForGeneratorSchedulingInPowerSystems,”IEEETr.ElectricalPower&EnergySystems,Vol18.No1,pp19-261996.[13]Edmund,T.H.HengDiptiSrinivasanA.C.Liew.“ShortTermLoadForecastingUsingGeneticAlgorithmAndNeuralNetworks”.IEEECatalogueNo:98EX137pp576-581[14]Chew,J.M.,Lin,Y.H.,andChen,J.Y.,"TheGreyPredictorControlinInvertedPendulumSystem",JournalofChinaInstituteofTechnologyandCommerce,Vol.11,pp.17-26,1995[15]J.GreySyst.,“Introductiontogreysystemtheory,”vol.1,no.1,pp.1–24,1989基于C8051F單片機直流電動機反饋控制系統(tǒng)的設(shè)計與研究基于單片機的嵌入式Web服務(wù)器的研究MOTOROLA單片機MC68HC(8)05PV8/A內(nèi)嵌EEPROM的工藝和制程方法及對良率的影響研究基于模糊控制的電阻釬焊單片機溫度控制系統(tǒng)的研制基于MCS-51系列單片機的通用控制模塊的研究基于單片機實現(xiàn)的供暖系統(tǒng)最佳啟停自校正(STR)調(diào)節(jié)器單片機控制的二級倒立擺系統(tǒng)的研究基于增強型51系列單片機的TCP/IP協(xié)議棧的實現(xiàn)基于單片機的蓄電池自動監(jiān)測系統(tǒng)HYPERLINK"/detail.htm
溫馨提示
- 1. 本站所有資源如無特殊說明,都需要本地電腦安裝OFFICE2007和PDF閱讀器。圖紙軟件為CAD,CAXA,PROE,UG,SolidWorks等.壓縮文件請下載最新的WinRAR軟件解壓。
- 2. 本站的文檔不包含任何第三方提供的附件圖紙等,如果需要附件,請聯(lián)系上傳者。文件的所有權(quán)益歸上傳用戶所有。
- 3. 本站RAR壓縮包中若帶圖紙,網(wǎng)頁內(nèi)容里面會有圖紙預(yù)覽,若沒有圖紙預(yù)覽就沒有圖紙。
- 4. 未經(jīng)權(quán)益所有人同意不得將文件中的內(nèi)容挪作商業(yè)或盈利用途。
- 5. 人人文庫網(wǎng)僅提供信息存儲空間,僅對用戶上傳內(nèi)容的表現(xiàn)方式做保護處理,對用戶上傳分享的文檔內(nèi)容本身不做任何修改或編輯,并不能對任何下載內(nèi)容負責。
- 6. 下載文件中如有侵權(quán)或不適當內(nèi)容,請與我們聯(lián)系,我們立即糾正。
- 7. 本站不保證下載資源的準確性、安全性和完整性, 同時也不承擔用戶因使用這些下載資源對自己和他人造成任何形式的傷害或損失。
最新文檔
- 聘請顧問協(xié)議書
- 彩鋼瓦修復安全協(xié)議書
- 液化氣購銷合同協(xié)議書
- 現(xiàn)場建筑體變更協(xié)議書
- 學生碰牙齒調(diào)節(jié)協(xié)議書
- 理發(fā)店門店合同協(xié)議書
- 移動代理協(xié)議書
- 維修補漏協(xié)議書
- 電瓶購置協(xié)議書
- 資助建房協(xié)議書
- 期末易錯題型創(chuàng)新改編練習(專項練習)六年級下冊數(shù)學人教版
- 《橋梁工程概況介紹》課件
- 2025年四川成都道德與法制中考試卷(無)
- 2024年不動產(chǎn)登記代理人《地籍調(diào)查》考試題庫大全(含真題、典型題)
- 中醫(yī)基礎(chǔ)學題庫(附答案)
- 大學美育知到智慧樹章節(jié)測試課后答案2024年秋長春工業(yè)大學
- 2024年秋《MySQL數(shù)據(jù)庫應(yīng)用》形考 實驗訓練1 在MySQL中創(chuàng)建數(shù)據(jù)庫和表答案
- 《數(shù)據(jù)資產(chǎn)會計》 課件 第五章 數(shù)據(jù)資產(chǎn)的價值評估
- 合同到期不續(xù)簽的模板
- 北京市2018年中考歷史真題試卷(含答案)
- (完整版)新概念英語第一冊單詞表(打印版)
評論
0/150
提交評論