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

Article retrieval

文章檢索

首頁 >> 文章檢索 >> 往年索引

基于VGG16圖像特征提取和SVM的電能質(zhì)量擾動分類

來源：電工電氣發(fā)布時間：2023-07-27 14:27 瀏覽次數(shù)：311

基于VGG16圖像特征提取和SVM的電能質(zhì)量擾動分類

童占北1，鐘建偉1，李禎維2，吳建軍2，李家俊2
(1 湖北民族大學 智能科學與工程學院，湖北 恩施 445000；
2 國網(wǎng)湖北省電力有限公司恩施供電公司，湖北 恩施 445000)
 
    摘 要：針對傳統(tǒng)電能質(zhì)量擾動分類方法需人工選取特征量，易受人為經(jīng)驗干擾的問題，提出一種基于 VGG16 圖像特征提取和支持向量機 (SVM) 結(jié)合的電能質(zhì)量擾動分類方法。通過格拉姆角場 (GAF) 將電能質(zhì)量擾動信號映射到極坐標系中，生成格拉姆矩陣，并轉(zhuǎn)換為二維擾動圖像；利用 VGG16 網(wǎng)絡(luò)自動提取圖像特征的特點，將擾動圖像輸入 VGG16 網(wǎng)絡(luò)中進行提?。粚⑻崛〉奶卣鲾?shù)據(jù)作為 SVM 分類器的輸入，并引入十折交叉驗證對SVM 進行訓練，以提升分類器的性能，最后對擾動信號進行準確分類。仿真結(jié)果表明，該方法對于電能質(zhì)量擾動的分類具有較高的準確率。
    關(guān)鍵詞： 電能質(zhì)量；擾動分類；格拉姆角場；VGG16 網(wǎng)絡(luò)；支持向量機；十折交叉驗證
    中圖分類號：TM712     文獻標識碼：A     文章編號：1007-3175(2023)07-0007-07
 
Power Quality Disturbance Classification Based on
VGG16 Image Feature Extraction and SVM
 
TONG Zhan-bei1, ZHONG Jian-wei1, LI Zhen-wei2, WU Jian-jun2, LI Jia-jun2
（1 College of Intelligent Systems Science and Engineering, Hubei Minzu University, Enshi 445000, China;
2 State Grid Hubei Electric Power Co., Ltd. Enshi Power Supply Company, Enshi 445000, China）
 
    Abstract: Traditional power quality disturbance classification methods need to manually select feature quantities, which are susceptible to human experience interference. Hence, the paper proposes a power quality disturbance classification method based on the combination of VGG16 image feature extraction and Support Vector Machine (SVM). It first maps power quality disturbance signals to the polar coordinate system through Gramian Angular Field (GAF) to generate the Gramian matrix which is transformed into a two-dimensional disturbance image. Second, the characteristic of VGG16 network to automatically extract image features is used to input disturbed images into VGG16 network for extraction. Third, the extracted feature data is used as the input of SVM classifier, ten-fold cross-validation is introduced to train SVM to improve the performance of the classifier, and then disturbance signals are classified in an accurate way. The simulation results show that this method has higher accuracy of power quality disturbances classification.
    Key words: power quality; disturbance classification; Gramian angular field; VGG16 network; support vector machine; ten-fold cross-validation
 
參考文獻
[1] WANG K, XI Y.A new method of power quality disturbance classification based on deep belief network[J].Journal of Physics Conference Series，2021，1827(1)：012021.
[2] HUANG J, QU H, LI X.Classification for hybrid power quality disturbance based on STFT and its spectral kurtosis[J].Power System Technology，2016，40(10)：3184-3191.
[3] 徐佳雄，張明，王陽，等. 基于改進 Hilbert-Huang 變換的電能質(zhì)量擾動定位與分類[J]. 現(xiàn)代電力，2021，38(4)：362-369.
[4] 程志友，楊猛. 基于二維離散余弦 S 變換的電能質(zhì)量擾動類型識別[J] . 電力系統(tǒng)保護與控制，2021，49(17)：85-92.
[5] 徐艷春，樊士榮，譚超，等. 基于改進 EWT-CMPE 的高滲透率主動配電網(wǎng)電能質(zhì)量擾動檢測與分類[J]. 電網(wǎng)技術(shù)，2020，44(10)：3991-4000.
[6] 楊劍鋒，姜爽，石戈戈. 基于分段改進 S 變換的復合電能質(zhì)量擾動識別[J] . 電力系統(tǒng)保護與控制，2019，47(9)：64-71.
[7] SUNDARAM P K, GIRISHKUMAR G.Power Quality Disturbance Classification Based on Kalman Filter and Adaptive Neural Fuzzy Inference System (ANFIS)[C]//International Conference on Robotics and Artificial Intelligence，2021.
[8] BHAGAT A, NIMKAR S, DONGRE K, et al.Power Quality Disturbance Detection and Classification Using Artificial Neural Network Based Wavelet[J].International Journal of Computational Intelligence Research，2017，13(8)：2043-2064.
[9] 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.
[10] MAHELA O P, SHAIK A G, KHAN B, et al.Recognition of Complex Power Quality Disturbances Using S-Transform Based Ruled Decision Tree[J].IEEE Access，2020，8：173530-173547.
[11] DAWOOD Z, BABULAL C K.Power quality disturbance classification based on efficient adaptive Arrhenius artificial bee colony feature selection[J].International Transactions on Electrical Energy Systems，2021，31(5)：e12868.
[12] MALIK H, ALMUTAIRI A, ALOTAIBI M A.Power quality disturbance analysis using data-driven EMD-SVM hybrid approach[J].Journal of Intelligent and Fuzzy Systems，2021，42(6)：1-10.
[13] 曹夢舟，張艷. 基于卷積- 長短期記憶網(wǎng)絡(luò)的電能質(zhì)量擾動分類[J] . 電力系統(tǒng)保護與控制，2020，48(2)：86-92.
[14] WANG Z, OATES T.Imaging time-series to improve classification and imputation[C]//Proceedings of the 24th International Conference on Artificial Intelligence，2015：3939-3945.
[15] SHUKLA J, PANIGRAHI B K, Ray P K.Power quality disturbances classification based on Gramian angular summation field method and convolutional neural networks[J].International Transactions on Electrical Energy Systems，2021，31(12)：e13222.
[16] SIMONYAN K , ZISSERMAN A . Very deep convolutional networks for large-scale image recognition[EB/OL].(2015-04-10)[2023-03-10].http://arxiv.org/abs/1409.1556.
[17] 邵凱旋，何怡剛，汪磊. 基于多尺度熵分析與改進 SVM 的變壓器故障識別[J] . 電子測量與儀器學報，2022，36(6)：161-168.
[18] MAMAT N, HAMZAH F M, JAAFAR O.Hybrid Support Vector Regression Model and K-Fold Cross Validation for Water Quality Index Prediction in Langat River, Malaysia[EB/OL].(2021-02-15)[2023-03-10].http://www.biorxiv.org/content/10.1101/2021.02.15.431242v1.abstract.
[19] IEEE Power & Energy Society.IEEE recommended practice for monitoring electric power quality:IEEE Std 1159-2019[S].Washington：IEEE，2019：34-79．