Researchon air discharge fault identification method of indoor equipment under lowpressure based on small sample random forest model

LI Tianhui; DONG Chi; SHI Wentong; GU Chaomin; DENG Zhuoli; BO Kehan; CHU Jifeng; YANG Aijun

doi:null

您当前的位置：

首页 >

文章列表页 >

Researchon air discharge fault identification method of indoor equipment under lowpressure based on small sample random forest model

更新时间：2025-05-08

- Researchon air discharge fault identification method of indoor equipment under lowpressure based on small sample random forest model
- Moren Journal (2025)
- 作者机构：
  
  1.国网河北省电力有限公司电力科学研究院,河北省石家庄市050022
  2.西安交通大学电气工程学院电工材料电气绝缘全国重点实验室,陕西省西安市710049
- 作者简介：
- 基金信息：
  
  Science and Technology Project of State Grid Hebei Electric Power Co., Ltd(kj2024-032)
- DOI：
  CLC：
- Received：23 October 2024，
  
  Revised：21 November 2024，
  
  Accepted：31 December 2024，
- 稿件说明：
移动端阅览
LI Tianhui, DONG Chi, SHI Wentong, et al. Researchon air discharge fault identification method of indoor equipment under lowpressure based on small sample random forest model[J/OL]. Moren journal, 2025.
DOI：

LI Tianhui, DONG Chi, SHI Wentong, et al. Researchon air discharge fault identification method of indoor equipment under lowpressure based on small sample random forest model[J/OL]. Moren journal, 2025. DOI：

摘要

空气绝缘的室内电力设备发生放电故障时，利用半导体气体传感器检测空气放电分解物的成分和含量，能够判断放电故障的严重程度，因此研发智能、精准的机器嗅觉算法对监测室内电力设备的正常运行有着重要意义。本文搭建了空气放电实验平台，模拟了低气压高海拔条件下的火花和局部放电故障，通过基于半导体传感器阵列的空气放电分解物检测装置对不同放电故障的气体分解物进行特征提取，然后利用基于主成分分析的小样本扩充增强方法对数据集进行丰富和精简，最后采用随机森林算法对放电分解故障进行分类。结果表明随机森林算法平均识别准确度97.22%，并对低气压高海拔下不同放电故障的预测准确率能够达到98.24%。本文提出的基于小样本随机森林模型的空气放电分解物识别方法，适用于小样本数据集，具有识别精准等显著优势，在空气放电故障检测领域具有广阔的前景。

Abstract

When adischarge fault occurs in air-insulated power indoor electrical equipment

theseverity of the discharge fault can be determined by using semiconductor gassensors to detect the composition and content of air discharge decompositionproducts. Therefore

the development of intelligent and accurate machineolfactory algorithms is of great significance for monitoring the normaloperation of indoor power equipment. This study constructs an air dischargeexperimental platform

simulating spark and partial discharge faults in highaltitude conditions with low pressure

and employs an air dischargedecomposition product detection device based on a semiconductor sensor array toextract features of gases from different discharge faults. Subsequently

a dataset enrichment and reduction method based on Principal Component Analysis (PCA)for small sample expansion is applied to enrich and refine the dataset.Finally

a Random Forest algorithm is used for the classification of dischargedecomposition faults. The results indicate that the Random Forest algorithmachieves an average recognition accuracy of 97.22%

and the prediction accuracyfor different discharge faults in high altitude conditions with low pressure canreach 98.24%. The small sample Random Forest model-based air dischargedecomposition product recognition method proposed in this paper is applicableto small sample datasets and has significant advantages such as preciseidentification

holding broad prospects in the field of air discharge faultdetection.

关键词

Keywords

references

Chu , Jifeng , Weijuan Li , Xu Yang , Yue Wu , Dawei Wang , Aijun Yang , Huan Yuan , Xiaohua Wang , Yunjia Li , and Mingzhe Rong . 2021 . “ Identification of Gas Mixtures via Sensor Array Combining with Neural Networks .” Sensors and Actuators : B . Chemical 329 (February).

林涛，韩冬，钟海峰，靳骁，张国强 . 工频交流电晕放电下SF6气体分解物形成的影响因素 [J ] . 电工技术学报 , 2014 , 第29卷 ( 2 ): 219 - 225

李进，赵仁勇，杜伯学，郝留成，边凯 . 电工环氧绝缘件缺陷无损检测方法研究进展 [J ] . 电工技术学报 , 2021 , 第36卷 ( 21 ): 4598 - 4607

刘宝升，王涤，马爱军，归宇，张秋实，张军，龚立，田猛，王先培 . 封闭环境空气局部放电组分研究 [J ] . 电力系统保护与控制 , 2021 , 第49卷 ( 9 ): 21 - 28

Hassan Javed , Kang LI , Guoqiang Zhang , and Adrian Traian Plesca . 2018 . “ Experimental Study on Air Decomposition By-Product Under Creepage Discharge Fault and Their Impact on Insulating Materials .” Journal of Electrical Engineering & Technology 13 ( 6 ): 2392 – 2401 .

Li , Kang , Hassan Javed , Guoqiang Zhang , and Adrian Traian Plesca . 2017 . “ Analysis of Air Decomposition By-Products under Four Kinds of Partial Discharge Defects.” IEEE Transactions on Dielectrics and Electrical Insulation, Dielectrics and Electrical Insulation, IEEE Transactions on, IEEE Trans. Dielect. Electr . Insul 24 ( 6 ): 3713 – 21 .

张天东，王力农，宋斌，马国强 . 基于化学反应速率理论的空气开关柜局部放电特征气体研究 [J ] . 高压电器 , 2023 , 第59卷 ( 12 ): 19 - 26

卓然，关伟，陈林聪，裘吟君，张晓星，胡雄雄 . 气隙缺陷局部放电下空气分解组分特性研究 [J ] . 高压电器 , 2017 , 第53卷 ( 2 ): 139 - 146，153

Tang J. , F. Zeng , X. Zhang , J. Pan , Q. Yao , X. Hou , and Y. Tang . 2014 . “ Relationship between Decomposition Gas Ratios and Partial Discharge Energy in GIS, and the Influence of Residual Water and Oxygen.” IEEE Transactions on Dielectrics and Electrical Insulation, Dielectrics and Electrical Insulation, IEEE Transactions on, IEEE Trans. Dielect. Electr . Insul 21 ( 3 ): 1226 – 34 .

汪金刚，林伟，王志，李健，何为，王平 . 基于紫外检测的开关柜电弧在线检测装置 [J ] . 电力系统保护与控制 , 2011 ,( 5 ): 128 - 133，152

Zhu , Huichao , Yu Wu , Ge Yang , Ruijie Song , Jun Yu , and Jianwei Zhang . 2024 . “ Electronic Nose Drift Suppression Based on Smooth Conditional Domain Adversarial Networks .” Sensors ( 14248220 ) 24 ( 4 ): 1319 .

Zeng Y. , R. Ma , F. Wu , J. Zhuo , L. Wang , S. Duan , P. Feng , and X. Peng . 2024 . “ Design and Implementation of Portable Electronic Nose System Based on Lightweight Convolutional Neural Network .” IEEE Sensors Journal, Sensors Journal, IEEE, IEEE Sensors J 24 ( 12 ): 19770 – 79 .

刘红秀，骆德汉，张泽勇 . 机器嗅觉系统气味识别算法 [J ] . 传感技术学报 , 2006 ,( 6 ): 2518 - 2522

汪雨晴，赵庆贺 . 模拟退火优化的径向量核支持向量回归算法在人工嗅觉系统的应用 [J ] . 仪表技术与传感器 , 2022 ,( 7 ): 111 - 116

Yang , Xingan , Meng Li , Xiaohua Ji , Junqing Chang , Zanhong Deng , and Gang Meng . 2023 . “ Recognition Algorithms in E-Nose: A Review .” IEEE SENSORS JOURNAL 23 ( 18 ): 20460 – 72 .

张振宇，晋涛，王天正，邵云峰，刘永强，王鹏 . 基于气体检测的空气开关柜绝缘缺陷识别 [J ] . 武汉大学学报(工学版) , 2021 , 第54卷 ( 8 ): 761 - 768

Liu T. , W. Zhang , L. Wang , M. Ueland , S.L. Forbes , W.X. Zheng , and S.W. Su . 2024 . “ Numerical Differentiation From Noisy Signals: A Kernel Regularization Method to Improve Transient-State Features for the Electronic Nose.” IEEE Transactions on Systems, Man, and Cybernetics: Systems, Systems, Man, and Cybernetics: Systems, IEEE Transactions on, IEEE Trans. Syst . Man Cybern, Syst 54 ( 6 ): 3497 – 3511 .

Wang M. , D. Chen , and P. Jia . 2023 . “ Gas Sensitive Analysis of Composite MoS 2 ( 001 ) Surface to Air Discharge Products: A DFT Study.” IEEE Sensors Journal, Sensors Journal, IEEE, IEEE Sensors J 23 ( 19 ): 22188 – 95 .

Zhou , Qu , Changxiang Hong , Yao Yao , Shahid Hussain , Lingna Xu , Qingyan Zhang , Yinggang Gui , and Mingsong Wang . 2018 . “ Hierarchically MoS2 Nanospheres Assembled from Nanosheets for Superior CO Gas-Sensing Properties.” Materials Research Bulletin 101 (May) : 132 – 39 .

Zhang , Dongzhi , Yan’e Sun , Chuanxing Jiang , and Yong Zhang . 2017 . “ Room Temperature Hydrogen Gas Sensor Based on Palladium Decorated Tin Oxide/Molybdenum Disulfide Ternary Hybrid via Hydrothermal Route .” Sensors & Actuators : B . Chemical 242 (April): 15 – 24 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Intelligent Checking Method of Abnormal Electricity Data Based on Stacking Strategy and Deep Learning

Overview of Fault Prediction Technology of Power Grid Transmission Equipment Under Typhoon Disaster

Post-seismic State Assessment Method for ±800 kV Dry-type Smoothing Reactor Based on Machine Learning

Study on Characteristics of SF₆ Decomposition Products Based on Graphene/Metal Heterojunctions

Research on Screening and Identification Methods of Semiconductor Sensors for Detecting SF₆ Decomposition Products

Related Author

ZHU Jianquan

FANG Langsen

ZHENG Yingkai

WANG Jiang

YAO Liming

ZHANG Jie

DING Shengze

YANG Liu

Related Institution

School of Electric Power Engineering, South China University of Technology

Guangdong Electric Power Trading Center Co., Ltd.

Wuhan Nari Limited Liability Company of State Grid Electric Power Research Institute

Research Institute of State Grid Zhejiang Electric Power Company

College of Civil Engineering, Tongji University

⁰