江亞群等
摘要:自適應(yīng)重合閘的功能是快速、準(zhǔn)確地辨識(shí)故障性質(zhì)及捕捉電弧熄滅時(shí)刻.在分析瞬時(shí)性故障和永久性故障斷路器跳閘后的端電壓波形復(fù)雜性的基礎(chǔ)上,提出了局部均值分解(LMD)、近似熵和線性支持向量機(jī)(SVM)相結(jié)合的自適應(yīng)重合閘整體實(shí)現(xiàn)方案.利用LMD分解故障信號(hào)得到若干個(gè)PF分量,選取前3個(gè)PF分量算出其近似熵值構(gòu)成三維特征向量,將三維特征向量作為SVM的輸入量來(lái)區(qū)分故障性質(zhì)和捕捉電弧熄滅時(shí)刻.線路故障仿真結(jié)果表明,該方案可智能識(shí)別故障性質(zhì)和捕捉電弧熄滅時(shí)刻且具有一定的抗噪能力.
關(guān)鍵詞:自適應(yīng)重合閘;局部均值分解;近似熵;支持向量機(jī)
中圖分類號(hào):TM76 文獻(xiàn)標(biāo)識(shí)碼:A
Adaptive Reclosure Method Based
on LMDapproximate Entropy and SVM
JIANG Yaqun,LENG Chongfu,HUANG Chun,DAI Xusheng
(College of Electrical and Information Engineering, Hunan Univ, Changsha, Hunan410082,China)
Abstract:The key function of adaptive reclosing is to correctly identify fault nature and quickly capture the transient fault arc extinction time. Based on the analysis of the waveform complexity of fault terminal voltage after circuit breaker tripping under transient fault and permanent fault, this paper presented an adaptive reclosing overall implementation by combining local mean decomposition (LMD), approximate entropy and support vector machine (SVM). After getting the PF components of fault signal by using LMD decomposition, the approximate entropy of the first three PF components is calculated, which constitutes a threedimensional feature vector as the input of SVM to identify fault nature and to capture arc extinguishing moment. Simulation results verify that this method can intelligently distinguish the transient fault from permanent fault, and capture transient fault extinction time with a strong antinoise ability.
Key words:adaptive reclosure; local mean decomposition; approximate entropy; support vector machine
超高壓輸電線路由于重負(fù)荷、長(zhǎng)距離輸電以及布局在室外,致使其易受自然因素的影響而發(fā)生瞬時(shí)性故障和永久性故障.運(yùn)行數(shù)據(jù)表明,大多數(shù)故障為單相接地瞬時(shí)性故障.因此,超高壓輸電線路上普遍安裝單相重合閘裝置來(lái)提高輸電線路供電的連續(xù)性及可靠性.單相自動(dòng)重合閘裝置在故障跳閘后延時(shí)一段時(shí)間就進(jìn)行重合,當(dāng)重合于永久性故障或未熄弧的瞬時(shí)性故障時(shí),將造成再次沖擊,可能破壞電力系統(tǒng)的穩(wěn)定性和損壞昂貴的電氣設(shè)備.為避免盲目重合帶來(lái)的不利影響,自適應(yīng)重合閘技術(shù)得到了廣泛的研究.
區(qū)分故障性質(zhì)和捕捉瞬時(shí)性故障熄弧時(shí)間是自適應(yīng)重合閘實(shí)現(xiàn)的關(guān)鍵.目前故障狀態(tài)區(qū)分的方法有多種.文獻(xiàn)[1]使用離散傅里葉變換(DFT)進(jìn)行諧波分析,并以奇次諧波含量大小來(lái)區(qū)分故障性質(zhì),計(jì)算量小,物理意義明顯,但DFT主要適應(yīng)穩(wěn)態(tài)信號(hào),計(jì)算暫態(tài)信號(hào)時(shí)誤差較大且該方法易受諧波變化的影響.文獻(xiàn)[2]利用小波能量譜區(qū)分故障性質(zhì),方法簡(jiǎn)單易于實(shí)現(xiàn),但小波變換不能自適應(yīng)的分解信號(hào),且小波基的優(yōu)化選取比較復(fù)雜.文獻(xiàn)[3]利用經(jīng)驗(yàn)?zāi)B(tài)分解(EMD)提取特征量來(lái)區(qū)分故障性質(zhì),能自適應(yīng)的分解故障信號(hào)且計(jì)算速度快,但EMD分解得到的分量缺乏實(shí)際的物理意義及存在模態(tài)混淆.文獻(xiàn)[4~5]采用神經(jīng)網(wǎng)絡(luò)對(duì)故障性質(zhì)進(jìn)行識(shí)別,需要大量的樣本,計(jì)算速度及精度不能滿足實(shí)際要求.文獻(xiàn)[6]基于電流差動(dòng)原理可實(shí)現(xiàn)瞬時(shí)性故障電弧熄滅判別,由于電力系統(tǒng)運(yùn)行的復(fù)雜性,導(dǎo)致該方法不易實(shí)現(xiàn).文獻(xiàn)[7]采用電壓幅值法區(qū)分故障性質(zhì),但易受故障地點(diǎn)的影響,且當(dāng)輸電線路上的電容耦合電壓較低時(shí)存在誤判的可能.文獻(xiàn)[8~9]利用恢復(fù)電壓的拍頻特性識(shí)別瞬時(shí)性故障,當(dāng)并聯(lián)電抗器的補(bǔ)償度較大時(shí)將導(dǎo)致判據(jù)的區(qū)分度不明顯,且該方法只適應(yīng)于帶并聯(lián)電抗器的輸電線路.
本文根據(jù)不同故障性質(zhì)在斷路器跳閘后電弧階段故障相端電壓波形復(fù)雜度的不同進(jìn)行故障性質(zhì)的識(shí)別,依據(jù)瞬時(shí)性故障的電弧狀態(tài)和電弧熄滅狀態(tài)的故障相端電壓波形復(fù)雜性的不同來(lái)捕捉電弧熄滅時(shí)刻.利用局部均值分解(Local Mean Decomposition,LMD)將故障信號(hào)分解成具有物理意義的PF分量,使用近似熵計(jì)算出PF的復(fù)雜度構(gòu)成特征向量,以所得特征向量作為線性支持向量機(jī)(Support Vector Machine,SVM)的輸入量來(lái)區(qū)分故障性質(zhì)和捕捉熄弧時(shí)刻,提出了自適應(yīng)重合閘的整體實(shí)現(xiàn)方案.