基于迁移学习的配电网内部过电压识别方法

doi:10.11930/j.issn.1004-9649.202006274

中国电力 ›› 2021, Vol. 54 ›› Issue (8): 52-59.DOI: 10.11930/j.issn.1004-9649.202006274

• 电能质量及其治理技术专栏 • 上一篇下一篇

基于迁移学习的配电网内部过电压识别方法

徐浩^1,2, 刘利强^1,2, 吕超³

1. 内蒙古工业大学电力学院，内蒙古呼和浩特 010080;
2. 内蒙古自治区电能变换传输与控制重点实验室，内蒙古呼和浩特 010080;
3. 内蒙古电力科学研究院，内蒙古呼和浩特 010020

收稿日期:2020-07-02 修回日期:2021-04-15 发布日期:2021-08-05
作者简介:徐浩(1994-),男,硕士研究生,从事电气设备状态检测与故障诊断方面的研究,E-mail:924326154@qq.com;刘利强(1975-),男,通信作者,博士,教授,从事电工新技术在电力系统中的应用研究,E-mail:llqiang@imut.edu.cn
基金资助:
内蒙古自治区自然科学基金面上项目（2020MS05029）

An Internal Overvoltage Identification Method for Distribution Network Based on Transfer Learning

XU Hao^1,2, LIU Liqiang^1,2, LV Chao³

1. College of Electric Power, Inner Mongolia University of Technology, Hohhot 010080, China;
2. Inner Mongolia Key Laboratory of Electrical Energy Conversion Transmission and Control, Hohhot 010080, China;
3. Inner Mongolia Electric Power Research Institute, Hohhot 010020, China

Received:2020-07-02 Revised:2021-04-15 Published:2021-08-05
Supported by:
This work is supported by Natural Science Foundation of Inner Mongolia Province in China (No.2020MS05029)

摘要/Abstract

摘要： 数据驱动方式作为解决配电网内部过电压识别的一种方法，因过电压样本数量较少而在实际应用中受到限制。为此，提出了一种基于迁移学习的深度卷积神经网络（D-CNN）配电网内部过电压识别算法。首先，通过仿真和连续小波变换（CWT）的方法构造了6种10 kV配电网内部过电压二维时频图。然后，分别利用Alexnet、Vgg-16、Googlenet、Resnet50等4种网络模型搭建了基于迁移学习的D-CNN网络模型。最后，将二维时频图带入改造后的D-CNN训练。经对实验结果比较分析发现，新搭建的VGG-16网络识别准确率最高且达到了99.07%，实现了在数据稀缺的情况下过电压故障的准确分类。

关键词: 配电网内部过电压, 连续小波变换, 迁移学习, 深度卷积神经网络, 模式识别

Abstract: As a measure for internal overvoltage identification of distribution network, the data driving method is limited in practical applications due to the small number of overvoltage samples. A transfer-learning-based deep convolutional neural network (D-CNN) algorithm is thus proposed to identify the internal overvoltage of distribution network. Firstly, 6 types of two-dimension time-frequency maps of 10 kV distribution network internal overvoltage are constructed by simulation and continuous wavelet transform (CWT). Then, the transfer-learning-based D-CNN network models are built using four network models, including Alexnet, Vgg-16, Googlenet and Resnet50. Finally, the two-dimension time-frequency maps are introduced into the transformed D-CNN for training. By comparing and analyzing the experimental results, it is found that the newly constructed VGG-16 network model has the highest identification accuracy, reaching 99.07%, which realizes the accurate classification of overvoltage faults in the case of scarce data.

Key words: internal overvoltage of distribution network, continuous wavelet transform, transfer learning, deep convolutional neural network, pattern recognition

徐浩, 刘利强, 吕超. 基于迁移学习的配电网内部过电压识别方法[J]. 中国电力, 2021, 54(8): 52-59.

XU Hao, LIU Liqiang, LV Chao. An Internal Overvoltage Identification Method for Distribution Network Based on Transfer Learning[J]. Electric Power, 2021, 54(8): 52-59.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.202006274

https://www.electricpower.com.cn/CN/Y2021/V54/I8/52

参考文献

[1] 李泽文, 周卿松, 曾祥君, 等. 基于行波模量传输时差的配电网接地故障定位新方法[J]. 中国电力, 2015, 48(9):67-72
LI Zewen, ZHOU Qingsong, ZENG Xiangjun, et al. New fault location method based on time difference of traveling wave mode-component in distribution network[J]. Electric Power, 2015, 48(9):67-72
[2] 杨勇, 李立浧, 杜林, 等. 采用时频矩阵奇异值分解和多级支持向量机的雷电及操作过电压识别[J]. 电网技术, 2012, 36(8):31-37
YANG Yong, LI Licheng, DU Lin, et al. Lightning and switching overvoltage identification based on singular value decomposition of time-frequency matrix and multi-level support vector machine[J]. Power System Technology, 2012, 36(8):31-37
[3] GAO W, WAI R J, LIAO Y F, et al. Internal overvoltage identification of distribution network via time-frequency atomic decomposition[J]. IEEE Access, 2019, 7:85110-85122.
[4] 金涛, 许立彬, 高伟, 等. 一种基于LCD-Hilbert变换和奇异谱熵的配电网暂时过电压类型识别方法[J]. 电机与控制学报, 2018, 22(11):26-36
JIN Tao, XU Libin, GAO Wei, et al. Temporary overvoltage classification and recognition method of distribution network based on LCD-Hilbert transform and singular-spectrum entropy[J]. Electric Machines and Control, 2018, 22(11):26-36
[5] GUO Y M, LIU Y, OERLEMANS A, et al. Deep learning for visual understanding:a review[J]. Neurocomputing, 2016, 187:27-48.
[6] 高熠, 田联房, 杜启亮. 基于Mask R-CNN的复合绝缘子过热缺陷检测[J/OL]. 中国电力:1-8[2020-07-01]. http://kns.cnki.net/kcms/detail/11.3265.TM.20200424.0827.002.html.
GAO Yi, TIAN Lianfang, DU Qiliang. Mask overheating defect detection of composite insulator based on mask R-CNN[J/OL]. Electric Power:1-8[2020-07-01]. http://kns.cnki.net/kcms/detail/11.3265.TM.20200424.0827.002.html.
[7] 高伟, 杨耿杰, 郭谋发, 等. 基于DTCWT-DBN的配电网内部过电压类型识别[J]. 电力系统保护与控制, 2019, 47(9):80-89
GAO Wei, YANG Gengjie, GUO Moufa, et al. Internal overvoltage type identification for distribution network based on DTCWT-DBN algorithm[J]. Power System Protection and Control, 2019, 47(9):80-89
[8] GUO M F, YANG N C, CHEN W F. Deep-learning-based fault classification using Hilbert-Huang transform and convolutional neural network in power distribution systems[J]. IEEE Sensors Journal, 2019, 19(16):6905-6913.
[9] 郑智聪, 王红, 齐林海. 基于深度学习模型融合的电压暂降源识别方法[J]. 中国电机工程学报, 2019, 39(1):97-104, 324
ZHENG Zhicong, WANG Hong, QI Linhai. Recognition method of voltage sag sources based on deep learning models' fusion[J]. Proceedings of the CSEE, 2019, 39(1):97-104, 324
[10] 高伟, 郭谋发, 许立彬. 基于改进CWD-CNN的配电网内部过电压类型识别方法[J/OL]. 电机与控制学报:1-10[2020-06-13]. http://kns.cnki.net/kcms/detail/23.1408.TM.20200107.1031.006.html.
GAO Wei, GUO Moufa, XU Libin. A method for identifying internal overvoltage type for distribution network via improved CWD-CNN[J/OL]. Electric Machines and Control:1-10[2020-06-13]. http://kns.cnki.net/kcms/detail/23.1408.TM.20200107.1031.006.html.
[11] 廖宇飞, 杨耿杰, 高伟, 等. 基于AD-CNN算法的配电网内部过电压识别技术[J]. 高电压技术, 2019, 45(10):3182-3191
LIAO Yufei, YANG Gengjie, GAO Wei, et al. Recognition technology of internal overvoltage in distribution network based on AD-CNN algorithm[J]. High Voltage Engineering, 2019, 45(10):3182-3191
[12] 林刚, 王波, 彭辉, 等. 基于强泛化卷积神经网络的输电线路图像覆冰厚度辨识[J]. 中国电机工程学报, 2018, 38(11):3393-3401
LIN Gang, WANG Bo, PENG Hui, et al. Identification of icing thickness of transmission line based on strongly generalized convolutional neural network[J]. Proceedings of the CSEE, 2018, 38(11):3393-3401
[13] ZHANG R, TAO H Y, WU L F, et al. Transfer learning with neural networks for bearing fault diagnosis in changing working conditions[J]. IEEE Access, 2017, 5:14347-14357.
[14] 周自强, 纪扬, 苏烨, 等. 基于迁移学习卷积神经网络的电缆隧道锈蚀识别算法[J]. 中国电力, 2019, 52(4):104-110
ZHOU Ziqiang, JI Yang, SU Ye, et al. A hybrid transfer learning/CNN algorithm for cable tunnel rust recognition[J]. Electric Power, 2019, 52(4):104-110
[15] 孟子超, 杜文娟, 王海风. 基于迁移学习深度卷积神经网络的配电网故障区域定位[J]. 南方电网技术, 2019, 13(7):25-33
MENG Zichao, DU Wenjuan, WANG Haifeng. Distribution network fault area location based on deep convolution neural network with transfer learning[J]. Southern Power System Technology, 2019, 13(7):25-33
[16] KAUR T, GANDHI T K. Deep convolutional neural networks with transfer learning for automated brain image classification[J]. Machine Vision and Applications, 2020, 31(3):1-16.
[17] GUO M F, YANG N C. Features-clustering-based earth fault detection using singular-value decomposition and fuzzy c-means in resonant grounding distribution systems[J]. International Journal of Electrical Power & Energy Systems, 2017, 93:97-108.
[18] GUO M F, ZENG X D, CHEN D Y, et al. Deep-learning-based earth fault detection using continuous wavelet transform and convolutional neural network in resonant grounding distribution systems[J]. IEEE Sensors Journal, 2018, 18(3):1291-1300.
[19] 冯秋实, 陈剑云, 林鹏, 等. 基于连续小波变换的输电线路故障行波测距方法的研究[J]. 电测与仪表, 2016, 53(2):40-44
FENG Qiushi, CHEN Jianyun, LIN Peng, et al. The research of transmission line fault location method based on the continuous wavelet transform[J]. Electrical Measurement & Instrumentation, 2016, 53(2):40-44
[20] 钱国超, 赵仲勇, 邹德旭, 等. 基于连续小波变换的变压器绕组变形故障类型检测[J]. 高电压技术, 2017, 43(6):2016-2023
QIAN Guochao, ZHAO Zhongyong, ZOU Dexu, et al. Detection of transformer winding deformation fault types based on continuous wavelet transform[J]. High Voltage Technology, 2017, 43(6):2016-2023
[21] WANG J. Identification technology of substation lightning overvoltage based on wavelet transform research and application[C]//2020 5th Asia Conference on Power and Electrical Engineering (ACPEE). IEEE, 2020:1746-1750.
[22] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. Imagenet classification with deep convolutional neural networks[C]//Advances in neural information processing systems. 2012:1097-1105.
[23] SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[J]. arXiv preprint arXiv:1409.1556, 2014.
[24] SZEGEDY C, LIU W, JIA Y, et al. Going deeper with convolutions[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2015:1-9.
[25] HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas, NV, USA. IEEE, 2016:770-778.

基于迁移学习的配电网内部过电压识别方法

An Internal Overvoltage Identification Method for Distribution Network Based on Transfer Learning

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 7

编辑推荐

Metrics

[1]	任明, 李乾宇, 夏昌杰, 董明. 基于稀疏光谱成像的线路复合绝缘子积污状态可视化评估[J]. 中国电力, 2025, 58(2): 203-215.
[2]	赵晨浩, 焦在滨, 李程昊, 张迪, 张鹏辉. 基于主动迁移学习的电力系统暂态稳定自适应评估[J]. 中国电力, 2025, 58(1): 70-77.
[3]	黄雪莜, 熊俊, 张宇, 刘辉, 陈鹭, 孟祥麟, 江秀臣. 基于残差卷积神经网络的开关柜局部放电模式识别[J]. 中国电力, 2021, 54(2): 44-51.
[4]	高熠, 田联房, 杜启亮. 基于Mask R-CNN的复合绝缘子过热缺陷检测[J]. 中国电力, 2021, 54(1): 135-141.
[5]	周自强, 纪扬, 苏烨, 蔡钧宇. 基于迁移学习卷积神经网络的电缆隧道锈蚀识别算法[J]. 中国电力, 2019, 52(4): 104-110.
[6]	田双双, 靳一林, 舒乃秋, 邵翠玲, 张扬. 绝缘子污秽放电声发射信号特征量提取的研究[J]. 中国电力, 2016, 49(7): 15-19.
[7]	李培江, 朱晓锦, 尤婷. 基于神经网络GIS局部放电模式的识别[J]. 中国电力, 2013, 46(10): 60-66.

模态框（Modal）标题

基于迁移学习的配电网内部过电压识别方法

An Internal Overvoltage Identification Method for Distribution Network Based on Transfer Learning

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 7

编辑推荐

Metrics