Transformer Fault Diagnosis Based on Data Enhanced One-dimensional Improved Convolutional Neural Network

被引：0
|
作者：
Li P. [1 ]
Hu G. [1 ]
机构：
[1] School of Electrical and Information Engineering, Anhui University of Science and Technology, Anhui Province, Huainan
来源：
Dianwang Jishu/Power System Technology | 2023年 / 47卷 / 07期
关键词：
data enhancement; deep learning; one-dimensional convolutional neural network; transformer fault diagnosis; variational autoencoder;
D O I：
10.13335/j.1000-3673.pst.2022.1902
中图分类号：
学科分类号：
摘要：
In order to improve the accuracy of transformer fault diagnosis based on the deep learning, a new method of transformer fault diagnosis based on the data-enhanced one-dimensional improved convolutional neural network is proposed. Firstly, to deal with the blindness in the data generation process of the traditional variational auto-encoder, an improved variational auto-encoder (IVAE) is proposed to optimize the strategy of the transformer dissolved gas data generation. Secondly, in order to adapt the deep network structure for the dissolved gas data of the transformer, a one-dimensional improved convolutional neural network (1D-ICNN) is constructed as the classifier of the fault diagnosis. Finally, the feasibility and the adaptability of the proposed method are verified by simulation studies. The results show that the IVAE is able to effectively deal with the limitations of the scarce transformer fault data samples and the low diagnostic accuracy, and the data-enhanced 1D-ICNN is excellent in the classification accuracy with the accuracy increased by 13.49%. The research results may provide new ideas for the accurate diagnosis of the transformer faults. © 2023 Power System Technology Press. All rights reserved.
引用
收藏
页码：2957 / 2966
页数：9
相关论文
共 30 条
  • [1] LIU Yunpeng, HE Jiahui, XU Ziqiang, Equalization method of power transformer fault sample based on SVM SMOTE[J], High Voltage Engineering, 46, 7, pp. 2522-2529, (2020)
  • [2] Ke MENG, Zhaoyang DONG, WANG Dianhui, A self-adaptive RBF neural network classifier for transformer fault analysis[J], IEEE Transactions on Power Systems, 25, 3, pp. 1350-1360, (2010)
  • [3] Jinzhong LI, ZHANG Qiaogen, WANG Ke, Optimal dissolved gas ratios selected by genetic algorithm for power transformer fault diagnosis based on support vector machine[J], IEEE Transactions on Dielectrics and Electrical Insulation, 23, 2, pp. 1198-1206, (2016)
  • [4] ZHANG Yujie, FENG Jian, LI Dianyang, New feature selection method for transformer fault diagnosis based on DGA data[J], Power System Technology, 45, 8, pp. 3324-3332, (2021)
  • [5] (2015)
  • [6] TAHA I B M，, HOBALLAH A, GHONEIM S S M．, Optimal ratio limits of rogers’ four-ratios and IEC 60599 code methods using particle swarm optimization fuzzy-logic approach[J], IEEE Transactions on Dielectrics and Electrical Insulation, 27, 1, pp. 222-230, (2020)
  • [7] SOUAHLIA S, BACHA K, CHAARI A．, MLP neural network-based decision for power transformers fault diagnosis using an improved combination of Rogers and Doernenburg ratios DGA[J], International Journal of Electrical Power & Energy Systems, 43, 1, pp. 1346-1353, (2012)
  • [8] CHEN Xin'gang, LI Taifu, Research on expert system of transformer insulation fault diagnosis based on DGA characteristic parameters[J], Transformer, 42, 1, pp. 33-36, (2005)
  • [9] SARKAR S, SHARMA T, BARAL A, An expert system approach for transformer insulation diagnosis combining conventional diagnostic tests and PDC，RVM data[J], IEEE Transactions on Dielectrics and Electrical Insulation, 21, 2, pp. 882-891, (2014)
  • [10] HINTON G E, SALAKHUTDINOV R R, Reducing the dimensionality of data with neural networks[J], Science, 313, 5786, pp. 504-507, (2006)
123