Suzhou Electric Appliance Research Institute
期刊号: CN32-1800/TM| ISSN2097-6623

Article retrieval

文章检索

首页 >> 文章检索 >> 最新索引

基于多特征融合与极限学习机的低压串联电弧故障检测

来源:电工电气发布时间:2026-02-26 12:26 浏览次数:5

基于多特征融合与极限学习机的低压串联电弧故障检测

高启明,迟长春,赵路尧
(上海电机学院 电气学院,上海 201306)
 
    摘 要:针对低压串联电弧故障检测问题,提出了一种基于多特征融合的极限学习机(ELM)模型,构建包含时域特征、小波能量特征、AR模型预测误差比及变分模态分解(VMD)能量熵的多特征矩阵,以提高故障识别的准确性和鲁棒性,并采用优化的 ELM 模型进行故障分类,通过对比支持向量机(SVM)、随机森林和 BP 神经网络验证其性能优势。实验结果表明:所提方法在复杂负载环境下能有效降低误检率,准确率由单一特征的85%提升至96%,在混合负载场景下,误检率由12%降至4%;ELM 模型相较于 BP 神经网络,训练速度提升 5 倍,内存占用降低60%。该方法在噪声干扰和负载突变条件下仍具有良好鲁棒性,为智能电网安全运行提供了技术支持。
    关键词: 低压串联电弧故障;多特征融合;极限学习机;小波能量;VMD 能量熵
    中图分类号:TM501+.2     文献标识码:A     文章编号:2097-6623(2026)02-0054-05
 
Detection of Low-Voltage Series Arc Faults Based on Multi-Feature
Fusion and Extreme Learning Machine
 
GAO Qi-ming, CHI Chang-chun, ZHAO Lu-yao
(School of Electrical Engineering, Shanghai Dianji University, Shanghai 201306, China)
 
    Abstract: Aiming at the problem of low-voltage series arc fault detection, an extreme learning machine (ELM) model based on multi-feature fusion is proposed. A multi-feature matrix including time-domain features, wavelet energy features, AR model prediction error ratio and variational mode decomposition(VMD) energy entropy is constructed to improve the accuracy and robustness of fault identification. An optimized ELM model is adopted for fault classification, and its performance advantages are verified by comparison with support vector machine (SVM),Random Forest and BP neural network. The experimental results show that the proposed method can effectively reduce the misdetection rate in complex load environments, the accuracy is increased from 85% of single feature to 96%, and the misdetection rate is reduced from 12% to 4% in mixed load scenarios. Compared with the BP neural network, the ELM model has a 5-fold improvement in training speed and a 60% reduction in memory usage. This method still has good robustness under the conditions of noise interference and load mutation, providing technical support for the safe operation of smart grids.
    Key words: low-voltage series arc fault; multi-feature fusion; extreme learning machine; wavelet energy; variational mode decomposition energy entropy
 
参考文献
[1] 应急管理部消防救援局.中国消防救援年鉴(2020 年卷)[M]. 北京:应急管理部消防救援局,2020.
[2] BABRAUSKAS V.Fires originating in branch-circuit NM cables due to installation damage[J].Journal of Fire Sciences,2018,36(5) :438-450.
[3] JOHNSON J B.The arc-fault circuit interrupter:An emerging product[J].IEEE Transactions on Consumer Electronics,1993,39(2) :63-70.
[4] PARISE G, MARTIRANO L, LAURINI M.Simplified Arc-Fault Model:The Reduction Factor of the Arc Current[J].IEEE Transactions on Industry Applications,2013,49(4) :1703-1710.
[5] 贾科,张旸,温志文,等. 基于电流-温度动态映射的新能源直流汇集送出系统串联电弧故障检测[J]. 电网技术,2022,46(8) :3123-3131.
[6] 王炳和. 现代数字信号处理[M]. 西安:西安电子科技大学出版社,2011.
[7] 蓝会立,张认成. 基于混沌的故障电弧早期弧声检测方法研究[J] . 计算机测量与控制,2010,18(12) :2716-2717.
[8] 熊庆,肖戎,汲胜昌,等. 基于电磁辐射特性的直流电弧检测方法[J]. 高电压技术,2017,43(9) :2967-2975.
[9] 赵昕海, 张术臣, 李志深, 等. 基于 VMD 的故障特征信号提取方法[J] . 振动、测试与诊断,2018,38(1) :11-19.
[10] 李欣,李奎秀,李新宇,等. 变分模态分解融合特征熵的直流微电网电弧检测[J]. 电力系统及其自动化学报,2025,37(3) :47-58.
[11] 崔芮华,王传宇,王洋.VMD-ApEn 在航空交流串联型电弧故障检测中的应用[J] . 电机与控制学报,2020,24(8) :141-149.
[12] 中国电器工业协会. 电弧故障保护电器(AFDD)的一般要求:GB/T 31143—2014[S] . 北京:中国标准出版社,2014 :4-8.
[13] CUI Y,SHI J,WANG Z,et al.Series Arc Fault Detection Based on Variational Mode Decomposition and Support Vector Machine[J].IEEE Transactions on Instrumentation and Measurement,2022,71 :1-12.
[14] WU H R, LI X H, STADE D, et al.Arc fault model for low voltage AC systems[J].IEEE Transactions on Power Delivery,2005,20(2) :1204-1205.
[15] 薛雪,张政一,刘晓文,等. 时频信息融合的电弧故障识别及仿真实验[J] . 实验技术与管理,2022,39(1) :217-222.