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 >> 文章瀏覽排名

一種基于SAE-RF算法的配電變壓器故障診斷方法

來(lái)源:電工電氣發(fā)布時(shí)間:2021-02-23 09:23 瀏覽次數(shù):756

一種基于SAE-RF算法的配電變壓器故障診斷方法

陳錦鋒1,張軍財(cái)1,盧思佳2,高偉2,范賢盛1,陳致遠(yuǎn)3
(1 國(guó)網(wǎng)福建南平供電公司,福建 南平 353000;2 福州大學(xué) 電氣工程與自動(dòng)化學(xué)院,福建 福州 350108;
3 上海宏力達(dá)信息技術(shù)股份有限公司,上海 200030)
 
    摘 要:為有效解決配電變壓器故障診斷中面臨的數(shù)據(jù)特征人工提取、機(jī)器學(xué)習(xí)調(diào)參困難等問(wèn)題,提出了一種基于堆棧自編碼器(SAE)和隨機(jī)森林(RF)組合的配電變壓器故障診斷方法。建立SAE配電變壓器故障特征自動(dòng)挖掘模型,利用大量的無(wú)標(biāo)簽數(shù)據(jù)對(duì)SAE模型中的每一個(gè)自編碼器進(jìn)行逐層無(wú)監(jiān)督訓(xùn)練,通過(guò)貝葉斯優(yōu)化算法自動(dòng)選擇模型的最優(yōu)參數(shù);通過(guò)有標(biāo)簽數(shù)據(jù)對(duì)模型參數(shù)進(jìn)行有監(jiān)督細(xì)調(diào),挖掘出能夠代表各種故障本質(zhì)屬性的特征量;創(chuàng)建一個(gè)RF分類器對(duì)故障類型進(jìn)行辨識(shí),調(diào)參過(guò)程同樣實(shí)現(xiàn)參數(shù)的自動(dòng)尋優(yōu)。試驗(yàn)結(jié)果表明,所提方法對(duì)配電變壓器故障診斷準(zhǔn)確率達(dá)96.67%,顯著優(yōu)于單獨(dú)使用SAE和RF的分類結(jié)果。
    關(guān)鍵詞:配電變壓器;故障診斷;堆棧自編碼器;隨機(jī)森林;貝葉斯優(yōu)化
    中圖分類號(hào):TM407;TM421     文獻(xiàn)標(biāo)識(shí)碼:A     文章編號(hào):1007-3175(2021)02-0017-07
 
A Novel Fault Diagnosis Method for Distribution Transformer Via Automatic
Feature Mining and Automatic Parameter Optimization
 
CHEN Jin-feng1, ZHANG Jun-cai1, LU Si-jia2, GAO Wei2, FAN Xian-sheng1, CHEN Zhi-yuan3
(1 State Grid Nanping Power Supply Company, Nanping 353000, China;
2 College of Electrical Engineering and Automation, Fuzhou University, Fuzhou 350108, China;
3 Shanghai Holystar Information Technology Co., Ltd, Shanghai 200030, China)
 
    Abstract: In order to effectively solve the problems of manual extraction of data features and difficulty of machine learning parameter adjustment in distribution transformer fault diagnosis, a fault diagnosis method for distribution transformer via the combination of stacked autoencoder (SAE) and random forest (RF) is proposed. First, a SAE model is established to realize automatic mining of distribution transformer fault characteristics, and a large number of unlabeled data is used to perform layer-by-layer unsupervised training of each auto-encoder in the model. After that, the optimal parameters of the model are automatically selected by Bayesian optimization algorithm. And then, fine-tune the model parameters through labeled data to mine features that can represent the essential attributes of various faults. Finally, an RF classifier is created to identify the fault type, and the parameter tuning process also realizes automatic parameter optimization. The test results show that the proposed method has an accuracy of 96.67% for distribution transformers fault diagnosis, which is significantly better than the results using SAE and RF alone.
    Key words: distribution transformer; fault diagnosis; stacked auto-encoder (SAE); random forest (RF); Bayesian optimization
 
參考文獻(xiàn)
[1] 趙莉華,豐遙,謝榮斌,等. 基于交叉小波的變壓器振動(dòng)信號(hào)幅頻特征量提取方法[J] . 高電壓技術(shù),2019,45(2):505-511.
[2] 王逸萍. 基于最小二乘支持向量機(jī)的變壓器故障診斷[J]. 電工電氣,2016(6):24-27.
[3] 孔德錢,張新燕,童濤,等. 基于差分進(jìn)化算法與BP 神經(jīng)網(wǎng)絡(luò)的變壓器故障診斷[J] . 電測(cè)與儀表,2020,57(5):57-61.
[4] 叢日立,趙明宇,周洋,等. 基于參數(shù)優(yōu)化的電力變壓器故障診斷模型[J] . 電測(cè)與儀表,2019,56(22):84-88.
[5] 齊波,王一鳴,張鵬,等. 基于自決策主動(dòng)糾偏的電力變壓器油色譜診斷模型[J] . 高電壓技術(shù),2020,46(1):23-32.
[6] 周光宇,馬松齡. 振動(dòng)法在線監(jiān)測(cè)變壓器的研究現(xiàn)狀與發(fā)展[J]. 電工電氣,2019(3):1-6.
[7] 楊毅,王豐華,段若晨,等. 基于自適應(yīng)篩選EMD和CFDC的變壓器繞組狀態(tài)檢測(cè)[J] . 振動(dòng)與沖擊,2017,36(19):106-111.
[8] 武立平,馬維青,程胤璋,等. 基于振動(dòng)信號(hào)指標(biāo)能量的變壓器機(jī)械故障檢測(cè)[J] . 電測(cè)與儀表,2020,57(6):126-131.
[9] ZHAO Miaoying, XU Gang.Feature extraction of power transformer vibration signals based on empirical wavelet transform and multiscale entropy [J]. IET Science , Measurement & Technology, 2018, 12(1):63-71.
[10] 趙妙穎,許剛. 基于經(jīng)驗(yàn)小波變換的變壓器振動(dòng)信號(hào)特征提取[J]. 電力系統(tǒng)自動(dòng)化,2017,41(20):63-69.
[11] 王杰,張曦煌. 基于圖卷積網(wǎng)絡(luò)和自編碼器的半監(jiān)督網(wǎng)絡(luò)表示學(xué)習(xí)模型[J] . 模式識(shí)別與人工智能,2019,32(4):317-325.
[12] 王懷遠(yuǎn),陳啟凡. 基于堆疊變分自動(dòng)編碼器的電力系統(tǒng)暫態(tài)穩(wěn)定評(píng)估方法[J] . 電力自動(dòng)化設(shè)備,2019,39(12):134-139.
[13] 崔廣新,李殿奎. 基于自編碼算法的深度學(xué)習(xí)綜述[J]. 計(jì)算機(jī)系統(tǒng)應(yīng)用,2018,27(9):47-51.
[14] 劉振興,陳震. 一種基于堆棧自編碼器的水聲信道均衡算法[J]. 通信技術(shù),2019,52(11):2605-2610.
[15] 徐鋒,方彥軍. 基于貝葉斯優(yōu)化XGBoost 的現(xiàn)場(chǎng)校驗(yàn)儀誤差預(yù)測(cè)[J] . 電測(cè)與儀表,2019,56(18):120-125.
[16] 葛陽(yáng),郭蘭中,牛曙光,等. 基于t-SNE和LSTM的旋轉(zhuǎn)機(jī)械剩余壽命預(yù)測(cè)[J] . 振動(dòng)與沖擊,2020,39(7):223-231.