基于GWO-LSSVM的直流故障电弧诊断方法

doi:10.16628/j.cnki.2095-8188.2025.01.003

摘要/Abstract

摘要：

针对直流故障电弧在不同工况下识别准确率不高的问题,提出基于灰狼优化算法的最小二乘支持向量机(GWO-LSSVM)对多负载工况下的直流电弧进行故障诊断。首先,应用改进的自适应噪声完备集合经验模态分解(ICEEMDAN),对参考高铁站混合负载得到的不同工况下直流电弧电流信号进行本征模态函数(IMF)分解。其次,进行筛选得到相关分量,结合多尺度排列熵(MPE)构造特征向量。最后,针对诊断模型的收敛速度较慢及模型倾向于陷入局部最优解的问题,应用GWO算法优化的LSSVM模型进行故障状态的识别。实验结果表明,准确率达到98.33%。通过与其他算法对比,证实所提方法的高效性。

关键词: 直流故障电弧, 多尺度排列熵, 灰狼优化算法, 故障诊断

Abstract:

In order to solve the problem that the identification accuracy of DC arc faults is not high under different working conditions,a gray wolf optimization least squares support vector machine (GWO-LSSVM) is proposed to diagnose DC arc under multi-load conditions.Firstly,the improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) is applied to perform the intrinsic mode function (IMF) decomposition on the DC arc current signals obtained from the mixed load of the reference high-speed railway station under the different operating conditions.Secondly,the relevant components are screened and combined with multi-scale permutation entropy (MPE) to construct the feature vectors.Finally,in response to the slow convergence speed of the diagnostic model and the tendency of the model to fall into the local optima,the LSSVM model optimized by GWO is applied for the fault state recognition.The experimental results show that the accuracy reaches 98.33%.By comparing with other algorithms,the efficiency of the proposed method has been confirmed.

Key words: DC fault arc, multi-scale permutation entropy, grey wolf optimization (GWO) algorithm, fault diagnosis

中图分类号:

TM501.2

刘树鑫, 刘丙泽, 邢朝建, 明欣, 周厚霖, 吕先锋. 基于GWO-LSSVM的直流故障电弧诊断方法[J]. 电器与能效管理技术, 2025, 0(1): 14-22.

LIU Shuxin, LIU Bingze, XING Zhaojian, MING Xin, ZHOU Houlin, LÜ Xianfeng. DC Fault Arc Diagnosis Method Based on GWO-LSSVM[J]. LOW VOLTAGE APPARATUS, 2025, 0(1): 14-22.

图/表 17

图1

图2

图3

图4

图5

表1

图6

图7

图8

图9

图10

图11

图12

图13

图15

表4

参考文献 24

[1]	牛壮. 光伏系统中直流故障电弧识别方法研究[D]. 沈阳: 沈阳工业大学, 2023.
[2]	汪倩, 陈思磊, 孟羽, 等. 交直流串联故障电弧实验平台开发[J]. 实验室研究与探索, 2023, 42(11):72-75,123.
[3]	赵宇初. 基于机器学习的光伏直流串联电弧故障识别方法研究[D]. 天津: 河北工业大学, 2022.
[4]	王尧, 盛德杰, 李腾, 等. 低压直流串联故障电弧研究综述[J]. 电器与能效管理技术, 2024(8):11-20.
[5]	李兴文, 张照资, 杨皓文, 等. 低压直流电弧检测与分断技术研究综述[J]. 电器与能效管理技术, 2024(8):1-10.
[6]	周学, 代文鑫. 基于时频域参数分析的直流电弧故障检测装置[J]. 电器与能效管理技术, 2024 (8):86-90,99.
[7]	高汉林, 许峰, 张涵. 光伏系统直流故障电弧特性与检测装置研究[J]. 电器与能效管理技术, 2022 (11):46-53.
[8]	陈欣凯, 鲍光海. 基于改进奇异值分解去噪算法的光伏系统串联电弧故障检测方法[J]. 电器与能效管理技术, 2023 (10):61-69,75.
[9]	朱海. 基于鲸鱼优化算法改进随机森林的电弧故障检测方法[J]. 电器与能效管理技术, 2024(2):21-27,55.
[10]	吴春华, 胡雅, 李智华, 等. 基于SSTDR的光伏系统直流母线电弧故障在线检测与定位[J]. 中国电机工程学报, 2020, 40(8):2725-2734.
[11]	熊庆, 刘小军, 郭自清, 等. 基于电流频谱积分差值的光伏系统电弧故障检测和定位[J]. 高电压技术, 2021, 47(5):1625-1633.
[12]	谢振华, 刘玉莹, 侯林明, 等. 基于注意力机制优化电弧特征的光伏直流电弧故障检测方法[J]. 电器与能效管理技术, 2024 (8):50-56,68.
[13]	郑昕, 单潇洁. 低压交流电弧电流零区电压波形特征分析与应用[J]. 电工技术学报, 2020, 35(22):4717-4725.
[14]	卢其威, 王涛, 李宗睿, 等. 基于小波变换和奇异值分解的串联电弧故障检测方法[J]. 电工技术学报, 2017, 32(17):208-217.
[15]	陈烜, 冷继伟, 李海峰. 基于全相位谱和深度学习的串联故障电弧识别方法[J]. 电力系统保护与控制, 2020, 48(17):1-8.
[16]	龙官微, 穆海宝, 张大宁, 等. 基于多特征融合神经网络的串联电弧故障识别技术[J]. 高电压技术, 2021, 47(2):463-471.
[17]	SAIBABU B B, KIRAN T, VISHALTEJA K. Hybrid wind speed forecasting using ICEEMDAN and transformer model with novel loss function[J]. Energy, 2023,265:126383.
[18]	HUANGY T, BAI Y L, YU Q H, et al. Application of a hybrid model based on the Prophet model,ICEEMDAN and multi-model optimization error correction in metal price prediction[J]. Resources Policy, 2022,79:102969.
[19]	XUE S, TAN J P, SHI X, et al. Rope tension fault diagnosis in hoisting systems based on vibration signals using EEMD,improved permutation entropy,and PSO-SVM[J]. Entropy, 2020, 22(2):209.
[20]	MIRJALILI S, MIRJALILI SM, LEWIS A. Grey wolf optimizer[J]. Advances in Engineering Software, 2014,69:46-61.
[21]	唐子辉, 蒋学君, 孙晓丽. 小波包分解结合LSSVM的永磁同步电机偏心故障诊断[J]. 机械设计与制造工程, 2024, 53(1):65-70.
[22]	JIANG J, ZHANG X, YUAN Z. Feature selection for classification with Spearman’s rank correlation coefficient-based self-information in divergence-based fuzzy rough sets[J]. Expert Systems with Applications, 2024,249:123633.
[23]	WANG Z Y, YAO L G, CHEN G, et al. Modified multiscale weighted permutation entropy and optimized support vector machine method for rolling bearing fault diagnosis with complex signals[J]. ISA Transactions, 2021,114:470-484.
[24]	李瑞, 范玉刚. 基于CEEMDAN多尺度排列熵和SORELM的高压隔膜泵单向阀故障诊断[J]. 振动与冲击, 2023, 42(5):127-135.

负载	数值
电阻/Ω	10、50、80、100
直流电灯/W	350
直流电机/W	375
电感/mH	15

模型	准确率/%	时间/s
ICEEMDAN-MPE-GWO-LSSVM	98.33	4.61
ICEEMDAN-MPE-WOA-LSSVM	92.50	12.57
ICEEMDAN-MPE-PSO-LSSVM	95.86	9.32
CEEMDAN-MPE-GWO-LSSVM	92.50	14.32
CEEMDAN-MPE-WOA-LSSVM	91.25	30.14
CEEMDAN-MPE-PSO-LSSVM	88.51	22.75
CEEMDAN - MPE -ELM	83.28	10.41
CEEMDAN - MPE -SVM	80.75	8.66