資源描述：
《基于改進(jìn)小波變換方法的電力系統(tǒng)》由會(huì)員上傳分享，免費(fèi)在線閱讀，更多相關(guān)內(nèi)容在工程資料-天天文庫(kù)。
1、基于改進(jìn)小波變換方法的電力系統(tǒng)低頻振蕩參數(shù)辨識(shí)*孫正龍1,蔡國(guó)偉1，王雨薇2，劉鋮1(1.東北電力大學(xué)電氣工程學(xué)院,吉林吉林132012；2.國(guó)網(wǎng)吉林省電力有限公司吉林供電公司，吉林吉林132012)摘要:基于多點(diǎn)量測(cè)數(shù)據(jù)的低頻振蕩模態(tài)參數(shù)辨識(shí)方法具有辨識(shí)精度高，覆蓋模態(tài)信息全的特點(diǎn)，但是該方法存在數(shù)據(jù)量增大，計(jì)算時(shí)間冗長(zhǎng)的問(wèn)題。針對(duì)上述問(wèn)題，本文將基于數(shù)據(jù)縮減技術(shù)的改進(jìn)小波變換參數(shù)識(shí)別方法應(yīng)用于電力系統(tǒng)低頻振蕩參數(shù)辨識(shí)中。該方法通過(guò)對(duì)發(fā)電機(jī)出口有功功率信號(hào)的正功率譜密度矩陣進(jìn)行奇異值分解，有效識(shí)別系統(tǒng)的模態(tài)階數(shù)。利用奇
2、異值分解將待辨識(shí)信號(hào)的協(xié)方差信號(hào)進(jìn)行數(shù)據(jù)縮減，充分保留信號(hào)的信息量，從而在保證計(jì)算合理及精度的前提有效地減少待辨識(shí)的數(shù)據(jù)量，進(jìn)而利用連續(xù)Morlet小波變換識(shí)別電力系統(tǒng)低頻振蕩參數(shù)。通過(guò)對(duì)4機(jī)2區(qū)域系統(tǒng)和EPRI-36節(jié)點(diǎn)系統(tǒng)進(jìn)行算例對(duì)分分析，結(jié)果表明改進(jìn)的小波變換方法能夠有在準(zhǔn)確提取電力系統(tǒng)低頻振蕩模態(tài)參數(shù)的前提下，有效減少計(jì)算所用數(shù)據(jù)量，提高計(jì)算效率。關(guān)鍵詞:低頻振蕩；模態(tài)辨識(shí)；Morlet小波變換；數(shù)據(jù)縮減中圖分類號(hào)：TM93文獻(xiàn)標(biāo)識(shí)碼：B文章編號(hào)：1001–1390（2016）00–0000–00Extracti
3、onandanalysisParameteridentificationofunderlowfrequencyoscillationusingbasedonimprovedwaveletalgorithmtransforminpowersystemSunZhenglong1,CaiGuowei1,WangYuwei2,LiuCheng1(1.SchoolofElectricalEngineering,（1.NortheastDianliUniversity,SchoolofElectricalEngineering,Ji
4、lin132012,Jilin,China.2.2.StateGridJilinShengElectricPowerSupplyCompany,Jilin132010,Jilin,Jilin,China）)Abstract:Theapproachofmodesextractingparameteridentificationunderoflowfrequencyoscillationmodesbasedonmulti-measurementdata,hastheadvantagesofhighprecisionandre
5、liability.However,thehugeamountofdataandlongcalculationtimehaveamountsofdataandlongcalculationtimehavelimitedtheapplicationoftheapproach.Inthispaper,arobustonlineapproachbasedonimprovedwavelettransformisproposedtoextracttheparametersofdominantoscillationmodefromw
6、ide-areameasurementsinpowersystemdominantoscillationmodefromwide-areameasurements.Thesingularvaluedecomposition(SVD)isusedtoanalyzethepositivepowerspectrummatrixofgeneratorelectromagnetismpowertodeterminetheordersofoscillationmodes.Toreducethecovariancesignals,th
7、eSVDisusedtodiminishtheamountofdatawhichisinvolvedintheextraction.Finally,themodalparametersareextractedfromeachmodeofreducedsignalsusingtheimprovedwaveletalgorithmtransforminthespecifiedfrequencyranges.Throughthestudyof4-generator2-areaandEPRI-36testsystems,itis
8、verifiedthattheproposedimprovedwaveletalgorithmtransformcouldextracttheaccurateoscillationparameterswithreducedcomputationdatasizetoincreaseimprovecalculatione
