Suzhou Electric Appliance Research Institute
期刊號(hào): CN32-1800/TM| ISSN1007-3175

Article retrieval

文章檢索

首頁(yè) >> 文章檢索 >> 往年索引

電能質(zhì)量擾動(dòng)檢測(cè)與識(shí)別方法綜述

來(lái)源:電工電氣發(fā)布時(shí)間:2025-03-03 15:03 瀏覽次數(shù):31

電能質(zhì)量擾動(dòng)檢測(cè)與識(shí)別方法綜述

葉鵬1,2,宋弘3,吳浩1,2,邱函1,2
(1 四川輕化工大學(xué) 自動(dòng)化與信息工程學(xué)院,四川 宜賓 644000;
2 人工智能四川省重點(diǎn)實(shí)驗(yàn)室,四川 宜賓 644000;
3 阿壩師范學(xué)院,四川 阿壩 624000)
 
    摘 要:隨著新能源發(fā)電設(shè)施的快速發(fā)展,電能質(zhì)量擾動(dòng)(PQDs)問(wèn)題愈發(fā)嚴(yán)峻,對(duì)其高效檢測(cè)與準(zhǔn)確識(shí)別提出了更高要求。梳理了 PQDs 研究中包括信號(hào)特征檢測(cè)精度不足、特征選擇冗余及擾動(dòng)類型識(shí)別能力有限等關(guān)鍵問(wèn)題,對(duì)國(guó)內(nèi)外相關(guān)研究成果進(jìn)行歸納總結(jié),詳細(xì)闡述了電能質(zhì)量擾動(dòng)檢測(cè)與識(shí)別方法的最新研究進(jìn)展;探討了基于先進(jìn)信號(hào)處理技術(shù)的特征檢測(cè)方法和智能算法的特征提取策略,以及依托深度學(xué)習(xí)模型的分類識(shí)別技術(shù),分析了各類方法的優(yōu)勢(shì)與不足。指出在電能質(zhì)量擾動(dòng)檢測(cè)與識(shí)別方面存在的問(wèn)題,并對(duì)未來(lái)發(fā)展趨勢(shì)進(jìn)行了展望。
    關(guān)鍵詞: 電能質(zhì)量擾動(dòng);擾動(dòng)檢測(cè);擾動(dòng)信號(hào);特征選擇;擾動(dòng)識(shí)別;深度學(xué)習(xí)模型
    中圖分類號(hào):TM712     文獻(xiàn)標(biāo)識(shí)碼:A     文章編號(hào):1007-3175(2025)02-0001-09
 
A Review of Detection and Classification Methods for
Power Quality Disturbances
 
YE Peng1, 2, SONG Hong3, WU Hao1, 2, QIU Han1, 2
(1 School of Automation and Information Engineering, Sichuan University of Science and Engineering, Yibin 644000, China;
2 Key Laboratory of Artificial Intelligence in Sichuan Province, Yibin 644000, China;
3 Aba Teachers College, Aba 624000, China)
 
    Abstract: With the rapid development of renewable energy generation facilities, the issue of power quality disturbances (PQDs) has become increasingly severe, raising higher demands for efficient detection and accurate identification. This paper first identifies key issues in PQDs research, including inadequate detection accuracy of signal characteristics, redundancy in feature selection, and limited capability in identifying types of disturbances. It summarizes relevant research findings from both domestic and international sources and elaborates on the latest advancements in detection and identification methods for power quality disturbances. Next, it focuses on feature detection methods based on advanced signal processing techniques, feature extraction strategies using intelligent algorithms, and classification and recognition techniques relying on deep learning models, offering a comprehensive analysis of the strengths and weaknesses of various approaches. Finally,the problems in power quality disturbance detection and identification are pointed out, and the future development trend is outlooked.
    Key words: power quality disturbance; disturbance detection; disturbance signal; featur selection; disturbance identification; deep learning model
 
參考文獻(xiàn)
[1] 齊霽. 淺析能源轉(zhuǎn)型助力實(shí)現(xiàn)“雙碳”目標(biāo)[J]. 環(huán)渤海經(jīng)濟(jì)瞭望,2024(2) :36-40.
[2] 吳榆俊,鐘森. 風(fēng)光儲(chǔ)微電網(wǎng)容量配置優(yōu)化綜述[ J ] .電氣技術(shù)與經(jīng)濟(jì),2022(4) :23-25.
[3] STET D, CZUMBIL L, MICU D D, et al.Power factor correction using EMTP-RV for engineering education[C]//2019 54th International Universities Power Engineering Conference(UPEC), 2019 :1-5.
[4] SHAO H, HENRIQUES R, MORAIS H, et al.Power quality monitoring in electric grid integrating offshore wind energy: A review[J].Renewable and Sustainable Energy Reviews,2024,191 :114094.
[5] 王燕. 電能質(zhì)量擾動(dòng)檢測(cè)的研究綜述[J]. 電力系統(tǒng)保護(hù)與控制,2021,49(13) :174-186.
[6] OLIVEIRA R A, BOLLEN M H J.Deep learning for power quality[J].Electric Power Systems Research,2023,214 :108887.
[7] RAY P K, MOHANTY S R, KISHOR N.Classification of power quality disturbances due to environmental characteristics in distributed generationsystem[J].IEEE Transactions on Sustainable Energy,2012,4(2) :302-313.
[8] ROBERTSON D C, CAMPS O I, MAYER J S, et al.Wavelets and electromagnetic power system transients[J].IEEE Transactions on Power Delivery,1996,11(2) :1050-1058.
[9] WRIGHT P S.Short-Time Fourier Transforms and Wigner-Ville Distributions Applied to the Calibration of Power Frequency Harmonic Analyzers[J].IEEE Transactions on Instrumentation and Measurement,1999,48(2) :475-478.
[10] GU Y H, BOLLEN M H J.Time-frequency and timescale domain analysis of voltage disturbances[J].IEEE Transactions on Power Delivery,2000,15(4) :1279-1284.
[11] SEVGI L.Numerical Fourier Transforms: DFT and FFT[J].IEEE Antennas and Propagation Magazine,2007,49(3) :238-243.
[12] LAI L L, CHAN W L, TSE C T, et al.Real-time frequency and harmonic evaluation using artificial neural networks[J].IEEE Transactions on Power Delivery,1999,14(1) :52-59.
[13] 張明銳,孫佳秀,周霖. 基于小波變換和 FFT 的電能質(zhì)量擾動(dòng)分類[J]. 機(jī)電一體化,2010,16(12) :29-34.
[14] KARIMI M, MOKHTARI H, IRAVANI M R.Wavelet based on-line disturbance detection for power quality applications[J].IEEE Transactions on Power Delivery,2000,15(4) :1212-1220.
[15] HE H , SHEN X , STARZYK J A . Power quality disturbances analysis based on EDMRA method[J].International Journal of Electrical Power & Energy Systems, 2009,31(6) :258-268.
[16] THIRUMALA K, PAL S, JAIN T, et al.A classification method for multiple power quality disturbances using EWT based adaptive filtering and multiclass SVM[J].Neurocomputing, 2019, 334 :265-274.
[17] LEE I W, DASH P K.S-transform-based intelligent system for classification of power quality disturbance signals[J].IEEE Transactions on Industrial Electronics, 2003, 50(4) :800-805.
[18] CHILUKURI M V , DASH P K . Multiresolution S-transform-based fuzzy recognition system for power quality events[J].IEEE Transactions on Power Delivery, 2004, 19(1) :323-330.
[19] 朱勇,陶用偉,李澤群. 基于 S 變換與特征優(yōu)選的電能質(zhì)量擾動(dòng)識(shí)別[J]. 電工技術(shù),2023(21) :97-100.
[20] BISWAL B, BISWAL M, MISHRA S, et al.Automatic classification of power quality events using balanced neural tree[J].IEEE Transactions on Industrial Electronics,2013,61(1) :521-530.
[21] JAYASREE T, DEVARAJ D, SUKANESH R.Power quality disturbance classification using Hilbert transform and RBF networks[J].Neurocomputing,2010,73(7/9) :1451-1456.
[22] 蘭名揚(yáng),劉宇龍,金濤,等. 基于可視化軌跡圓和 ResNet18 的復(fù)合電能質(zhì)量擾動(dòng)類型識(shí)別[J] . 中國(guó)電機(jī)工程學(xué)報(bào),2022,42(17) :6274-6285.
[23] 徐佳雄,張明,王陽(yáng),等. 基于改進(jìn) Hilbert-Huang 變換的電能質(zhì)量擾動(dòng)定位與分類[J] . 現(xiàn)代電力,2021,38(4) :362-369.
[24] 張小東. 基于自適應(yīng)卡爾曼濾波算法在電能質(zhì)量檢測(cè)中的應(yīng)用[D]. 長(zhǎng)沙:長(zhǎng)沙理工大學(xué),2019.
[25] BOLLEN M H J, GU I Y H.Signal processing of power quality disturbances[M].New York :John Wiley & Sons, 2006.
[26] 陳子璇,席燕輝,沈銀. 基于卡爾曼濾波和深度置信網(wǎng)絡(luò)的復(fù)合電能質(zhì)量擾動(dòng)分類[J]. 電力系統(tǒng)保護(hù)與控制,2022,50(7) :81-90.
[27] REDDY J B V, DASH P K, SAMANTARAY R, et al.Fast tracking of power quality disturbance signals using an optimized unscented filter[J].IEEE Transactions on Instrumentation and Measurement,2009,58(12) :3943-3952.
[28] HUANG S J, HUANG C L, HSIEH C T.Application of Gabor transform technique to supervise power system transient harmonics[J].IEE Proceedings-Generation, Transmission and Distribution,1996,143(5) :461-466.
[29] KAWADY T A, ELKALASHY N I, IBRAHIM A E, et al.Arcing fault identification using combined Gabor transform-neural network for transmission lines[J].International Journal of Electrical Power & Energy Systems,2014,61 :248-258.
[30] CHO S H, JANG G, KWON S H.Time-frequency analysis of power-quality disturbances via the Gabor-Wigner transform[J].IEEE Transactions on Power Delivery,2010,25(1) :494-499.
[31] 商立群,李朝彪,鄧力文,等. 基于 ISSA-XGBoost 的電能質(zhì)量擾動(dòng)識(shí)別方法研究[J] . 電力系統(tǒng)保護(hù)與控制,2024,52(13) :115-124.
[32] 李天云,陳昌雷,周博,等. 奇異值分解和最小二乘支持向量機(jī)在電能質(zhì)量擾動(dòng)識(shí)別中的應(yīng)用[J]. 中國(guó)電機(jī)工程學(xué)報(bào),2008,28(34) :124-128.
[33] 瞿合祚,劉恒,李曉明,等. 一種電能質(zhì)量多擾動(dòng)分類中特征組合優(yōu)化方法[J] . 電力自動(dòng)化設(shè)備,2017,37(3) :146-152.
[34] AHILA R, SADASIVAM V, MANIMALA K.An integrated PSO for parameter determination and feature selection of ELM and its application inclassification of power system disturbances[J].Applied Soft Computing,2015,32 :23-37.
[35] BISWAL B, DASH P K, MISHRA S.A hybrid ant colony optimization technique for power signal pattern classification[J].Expert Systems with Applications,2011,38(5) :6368-6375.
[36] 阮梓航,肖先勇,胡文曦,等. 基于多粒度特征選擇和模型融合的復(fù)合電能質(zhì)量擾動(dòng)分類特征優(yōu)化[J]. 電力系統(tǒng)保護(hù)與控制,2022,50(14) :1-10.
[37] HUANG N, LU G, CAI G, et al.Feature selection of power quality disturbance signals with an entropy-importance-based random forest[J].Entropy,2016,18(2) :44.
[38] CAMARILLO-PE~NARANDA J R , RAMOS G . Fault classification and voltage sag parameters computation using voltage ellipses[C]//IEEE 2018 IEEE/IAS 54th Industrial and Commercial Power Systems Technical Conference(I&CPS),2018 :1-6.
[39] LUO Y, LI K, LI Y, et al.Three-layer Bayesian network for classification of complex power quality disturbances[J].IEEE Transactions on Industrial Informatics,2017,14(9) :3997-4006.
[40] 聶曉華. 一種基于卡爾曼濾波的電能質(zhì)量擾動(dòng)檢測(cè)新方法[J] . 中國(guó)電機(jī)工程學(xué)報(bào),2017,37(22) :6649-6658.
[41] DENG L, YU D.Deep learning: Methods and applications[J].Foundations and Trends® in Signal Processing,2013,7(3/4) :197-387.
[42] WANG X.Deep learning in object recognition,detection, and segmentation[J].Foundations and Trends® in Signal Processing,2016,8(4):217-382.
[43] 龔正,鄒陽(yáng),金濤,等. 基于特征融合并行優(yōu)化模型的電能質(zhì)量擾動(dòng)分類方法[J] . 中國(guó)電機(jī)工程學(xué)報(bào),2023,43(3) :1017-1027.
[44] 鄭煒,林瑞全,王俊,等. 基于 GAF 與卷積神經(jīng)網(wǎng)絡(luò)的電能質(zhì)量擾動(dòng)分類[J] . 電力系統(tǒng)保護(hù)與控制,2021,49(11) :97-104.
[45] 瞿合祚,李曉明,陳陳,等. 基于卷積神經(jīng)網(wǎng)絡(luò)的電能質(zhì)量擾動(dòng)分類[J] . 武漢大學(xué)學(xué)報(bào)(工學(xué)版),2018,51(6) :534-539.
[46] 陳偉,何家歡,裴喜平. 基于相空間重構(gòu)和卷積神經(jīng)網(wǎng)絡(luò)的電能質(zhì)量擾動(dòng)分類[J] . 電力系統(tǒng)保護(hù)與控制,2018,46(14) :87-93.
[47] 賀才郡,李開(kāi)成,董宇飛,等. 基于知識(shí)蒸餾與 RP-MobileNetV3 的電能質(zhì)量復(fù)合擾動(dòng)識(shí)別[J]. 電力系統(tǒng)保護(hù)與控制,2023,51(14) :75-84.
[48] 張立鵬,鄭巖,秦剛,等. 一種實(shí)時(shí)電能質(zhì)量擾動(dòng)識(shí)別分類方法[J]. 河北工業(yè)科技,2019,36(1) :50-54.
[49] 錢(qián)倍奇,陳謙,李宗源,等. 基于馬爾可夫轉(zhuǎn)換場(chǎng)與多頭注意力機(jī)制的電能質(zhì)量擾動(dòng)分類方法[J]. 電網(wǎng)技術(shù),2024,48(2) :721-729.
[50] 張逸,歐杰宇,金濤,等. 基于特征圖像組合與改進(jìn) ResNet-18 的電能質(zhì)量擾動(dòng)識(shí)別方法[J]. 中國(guó)電機(jī)工程學(xué)報(bào),2024,44(7) :2531-2544.
[51] 王維博,張斌,曾文入,等. 基于特征融合一維卷積神經(jīng)網(wǎng)絡(luò)的電能質(zhì)量擾動(dòng)分類[J]. 電力系統(tǒng)保護(hù)與控制,2020,48(6) :53-60.
[52] 王偉,李開(kāi)成,許立武,等. 基于一維卷積神經(jīng)網(wǎng)絡(luò)多任務(wù)學(xué)習(xí)的電能質(zhì)量擾動(dòng)識(shí)別方法[J] . 電測(cè)與儀表,2022,59(3) :18-25.
[53] 王繼東,張迪. 基于側(cè)輸出融合卷積神經(jīng)網(wǎng)絡(luò)的電能質(zhì)量擾動(dòng)分類方法[J] . 電力自動(dòng)化設(shè)備,2021,41(11) :107-112.
[54] 朱瑞金,郭威麟,龔雪嬌. 基于自編碼器和卷積神經(jīng)網(wǎng)絡(luò)的電能質(zhì)量擾動(dòng)分類[J]. 電力系統(tǒng)及其自動(dòng)化學(xué)報(bào),2019,31(7) :70-75.
[55] 曹夢(mèng)舟, 張艷. 基于卷積-長(zhǎng)短期記憶網(wǎng)絡(luò)的電能質(zhì)量擾動(dòng)分類[J] . 電力系統(tǒng)保護(hù)與控制,2020,48(2) :86-92.
[56] 劉佳翰,陳克緒,馬建,等. 基于卷積神經(jīng)網(wǎng)絡(luò)和隨機(jī)森林的三相電壓暫降分類[J] . 電力系統(tǒng)保護(hù)與控制,2019,47(20) :112-118.
[57] CHEN Z, LI M, JI T, et al.Real-time recognition of power quality disturbance-based deep belief network using embedded parallel computing platform[J].IEEJ Transactions on Electrical and Electronic Engineering,2020,15(4) :519-526.
[58] 胡婧, 周洋, 何志強(qiáng), 等. 基于深度置信網(wǎng)絡(luò)和隨機(jī)森林的電力擾動(dòng)檢測(cè)方法[J]. 供用電,2020,37(9) :17-22.
[59] 武昭旭,楊岸,祝龍記. 基于循環(huán)神經(jīng)網(wǎng)絡(luò)的電能質(zhì)量擾動(dòng)識(shí)別[J] . 電力系統(tǒng)保護(hù)與控制,2020,48(18) :88-94.
[60] 王以忠,欒振國(guó),郭肖勇,等. 基于注意力機(jī)制和雙向長(zhǎng)短期記憶網(wǎng)絡(luò)的電能質(zhì)量擾動(dòng)識(shí)別[J]. 天津科技大學(xué)學(xué)報(bào),2021,36(4) :51-56.
[61] KHETARPAL P, NAGPAL N, SIANO P, et al.Power quality disturbance signal segmentation and classification based on modified BI-LSTM with double attention mechanism[J].IET Generation,Transmission & Distribution,2024,18(1) :50-62.
[62] 簡(jiǎn)獻(xiàn)忠,王緒濤,王如志. 基于生成對(duì)抗網(wǎng)絡(luò)的電能質(zhì)量信號(hào)壓縮重構(gòu)方法[J] . 控制工程,2021,28(8) :1654-1661.
[63] 屈相帥,段斌,尹橋宣,等. 基于稀疏自動(dòng)編碼器深度神經(jīng)網(wǎng)絡(luò)的電能質(zhì)量擾動(dòng)分類方法[J]. 電力自動(dòng)化設(shè)備,2019,39(5) :157-162.