logoGlobal Energy Interconnection

Identification of XLPE cable insulation defects based on deep learning

阅读原文 阅读原文

【论文推荐】北方工业大学万庆祝等:基于深度学习的XLPE电缆绝缘缺陷识别

摘要

电缆的绝缘老化是导致电缆寿命减少的主要原因,目前电力相关部门对电缆绝缘缺陷检测缺乏快速有效的方法。本文提出了一种基于深度学习的XLPE电缆绝缘缺陷识别方法。首先,介绍了谐波法检测电缆绝缘缺陷的原理;其次通过ANSYS软件模拟了电缆绝缘层含气泡、突起、水树枝缺陷,分析了各类缺陷对电缆绝缘层磁场强度和涡损电流的影响;然后,构建了总谐波含有率及2至10次谐波电流共10个特征量,建立了电缆绝缘缺陷数据库;最后,通过深度学习算法LSTM对电缆的绝缘缺陷类型进行了精确识别。结果表明,采用LSTM算法对电缆的绝缘缺陷能够有效诊断识别,其准确率达到95.83%。

Identification of XLPE cable insulation defects based on deep learning

基于深度学习的XLPE电缆绝缘缺陷识别

Tao Zhou 1, Xiaozhong Zhu 1, Haifei Yang 1, Xuyang Yan 2, Xuejun Jin 2, Qingzhu Wan 2

(1.State Grid Yangquan Power Supply Company, Yangquan Shanxi 045000, P. R. China

2.College of Electrical and Control Engineering, North China University of Technology, Beijing 100144, P. R. China)

收听作者1分钟语音介绍

Identification of XLPE cable insulation defects based on deep learning

Abstract

The insulation aging of cross-linked polyethylene (XLPE) cables is the main reason for the reduction in cable life. There is currently a lack of rapid and effective methods for detecting cable insulation defects in power-related sectors. To this end, this paper presents a method for identifying insulation defects in XLPE cables based on deep learning algorithms. First, the principle of the harmonic method for detecting cable insulation defects is introduced. Second, the ANSYS software is used to simulate the cable insulation layer containing bubbles, protrusions, and water tree defects, and the effects of each type of defect on the magnetic field strength and eddy loss current of the cable insulation layer are analyzed. Then, a total  of 10 characteristic quantities of the total harmonic content and 2nd to 10th harmonic currents are constructed to establish   a database of cable insulation defects. Finally, the deep learning algorithm, long short-term memory (LSTM), is used to accurately identify the types of insulation defects in cables. The results indicate that the LSTM algorithm can effectively diagnose and identify insulation defects in cables with an accuracy of 95.83%.

Keywords

Insulation defects, Deep learning, Database, Eddy loss current.

图片

Fig.1   Schematic diagram of eddy current in cable

图片

Fig.2   Common defects of cable insulation

图片

Fig.3   Schematic diagram of bubble defects in the insulating layer of XLPE cable

图片

Fig.4   Magnetic field intensity with changes in the bubble position

图片

Fig.5   Magnetic field intensity with changes in the bubble radius

图片

Fig.6   Magnetic field intensity with changes in the number of bubbles

图片

Fig.7   Trend of eddy loss currents as bubbles change

图片

Fig.8   Schematic diagram of protrusions defects in the insulating layer of XLPE cable

图片

Fig.9   Magnetic field intensity distribution of cable with different protrusion sizes

图片

Fig.10   Magnetic field intensity distribution of cable with different protrusions lengths

图片

Fig.11   Magnetic field intensity distribution of cable with the change of the number of protrusions

图片

Fig.12   Trend of eddy loss currents as the protrusion state changes

图片

Fig.13   Dielectric equivalent circuit diagram

图片

Fig.14   Vector diagram of a dielectric at AC voltage

图片

Fig.15  Equivalent circuit diagram for water tree defects

图片

Fig.16   Harmonic components of loss current in different development stages of water tree

图片

Fig.17   Basic unit structure of LSTM

图片

Fig.18   LSTM model diagnostic framework diagram

图片

Fig.19   Flow chart of cable insulation defect identification based on LSTM

图片

Fig.20   Accuracy and loss values the of LSTM algorithm

图片

Fig.21   Flow chart of cable insulation defect identification based on LSTM

本文引文信息

Zhou T, Zhu XZ, Yang HF, et al (2023) Identification of XLPE cable insulation defects based on deep learning. Global Energy Interconnection, 6(1): 36-50

周涛,朱晓中,杨海飞等 (2023) 基于深度学习的XLPE电缆绝缘缺陷识别. 全球能源互联网(英文), 6(1): 36-50

Biographies

图片

Tao Zhou 

Tao Zhou, senior engineer, is currently working in State Grid Shanxi Electric Power Company. His research interests are mainly in substation equipment operation and maintenance management, distribution equipment operation and maintenance management.

图片
图片

Xiaozhong Zhu

Xiaozhong Zhu, senior engineer, is currently working in State Grid Shanxi Electric Power Company. His research interests are mainly in power systems and fault diagnosis.

图片
图片

Haifei Yang

Haifei Yang, senior engineer, is currently working in State Grid Shanxi Electric Power Company. His research interests are mainly in power systems and fault diagnosis.

图片
图片

Xuyang Yan

Xuyang Yan is a graduate student at North China University of Technology. His research interests include cable fault diagnosis, and big data analysis.

图片
图片

Xuejun Jin

Xuejun Jin is a graduate student at North China University of Technology. Her research interests include cable fault diagnosis and smart grid.

图片
图片

Qingzhu Wan

Qingzhu Wan received Ph.D. degree at Tsinghua University. His research interests include cable fault diagnosis, big data analysis, and smart grids.

编辑:刘通明

审核:王   伟

郑重声明

根据国家版权局相关规定,纸媒、网站、微博、微信公众号转载、摘编本网站作品,需包含本网站名称、二维码等关键信息,并在文首注明《全球能源互联网》原创。 个人请按本网站原文转发、分享。