Least Squares Support Vector Machine Based Incipient Fault Identification in Non-solidly Grounding System

doi:10.11985/2023.03.032

Abstract

Abstract:

The incipient fault is the manifestation before a permanent fault. Timely and accurate identification of incipient fault is of great significance for realizing fault early warning and reducing the occurrence of permanent faults. Incipient fault characteristics in non-solidly grounding system are weak, and it’s quite difficult to be detected and identified. Therefore, an incipient fault identification algorithm is proposed based on least squares support vector machine(LS-SVM). Firstly, the shallow feature sets of disturbances are extracted based on physical characteristics and statistical characteristics respectively, and the energy entropy and singular entropy of different frequency bands are obtained based on S-transformation. Then, the maximum relevance and minimum redundancy method is adopted to reduce the data dimension of the feature set while retaining strongly correlated features, and the optimal feature set is constructed. Finally, PSCAD/EMTDC simulation system is used to obtain various disturbance samples, and the incipient fault identification model is obtained based on LS-SVM, then the particle swarm optimization algorithm is used to optimize LS-SVM parameters to improve the efficiency. Simulation examples show that the proposed algorithm can accurately identify incipient faults in non-solidly grounding system, which verifies the correctness and effectiveness of the proposed method.

Key words: Incipient fault, least squares support vector machine, low current grounding system, maximum relevance and minimum redundancy

CLC Number:

TM71

LI Hua, ZHU Yimin, MA Haijun, DING Jibo, CHU Tianxin, ZHANG Wenhai. Least Squares Support Vector Machine Based Incipient Fault Identification in Non-solidly Grounding System[J]. Journal of Electrical Engineering, 2023, 18(3): 297-306.

Figures/Tables 16

References 32

[1]	李宇, 杨柳林. 基于卷积神经网络的配电网单相接地故障识别[J]. 电气工程学报, 2020, 15(3):22-30.
	LI Yu, YANG Liulin. Identification of single phase to earth fault in distribution network based on convolution neural network[J]. Journal of Electrical Engineering, 2020, 15(3):22-30.
[2]	STRINGER N T, KOJOVIC L A. Prevention of underground cable splice failures[J]. IEEE Transactions on Industry Applications, 2001, 37(1):230-239. doi: 10.1109/28.903154
[3]	KULKARNI S, SANTOSO S, SHORT T A. Incipient fault location algorithm for underground cables[J]. IEEE Transactions on Smart Grid, 2014, 5(3):1165-1174. doi: 10.1109/TSG.2014.2303483
[4]	SHI S, HU Z, MA Z, et al. Travelling waves-based identification of sub-health condition of feeders in power distribution system[J]. IET Generation Transmission & Distribution, 2018, 12(5):1067-1073. doi: 10.1049/gtd2.v12.5
[5]	JIANG J, CHEN R, CHEN M, et al. Dynamic fault prediction of power transformers based on hidden Markov model of dissolved Gases analysis[J]. IEEE Transactions on Power Delivery, 2019, 34(4):1393-1400. doi: 10.1109/TPWRD.61
[6]	刘亚东, 丛子涵, 严英杰, 等. 配电设备早期故障检测基本原理、关键技术与发展趋势[J]. 供用电, 2020, 37(4):10-16,32.
	LIU Yadong, CONG Zihan, YAN Yingjie, et al. Basic principles,key technologies and development trends of incipient fault detection for power distribution equipments[J]. Distribution & Utilization, 2020, 37(4):10-16,32.
[7]	戴铭. 10 kV地下电缆早期故障检测与识别方法探讨[D]. 成都: 西南交通大学, 2015.
	DAI Ming. Discussion methods of incipient fault detection and identification in 10 kV underground cables[D]. Chengdu: Southwest Jiaotong University, 2015.
[8]	KASZTENY B, VOLOH I, JONES C G, et al. Detection of incipient faults in underground medium voltage cables[C]// IEEE Protective Relay Engineers 61st Annual Conference, April 1-3,2008,Texas A&M University,College Station,Texas. Piscataway:IEEE, 2008:349-366.
[9]	SAMET H, TAJDINIAN M, KHALEGHIAN S, et al. A statistical-based criterion for incipient fault detection in underground power cables established on voltage waveform characteristics[J]. Electric Power Systems Research, 2021, 197:107303. doi: 10.1016/j.epsr.2021.107303
[10]	GHANBARI T. Kalman filter based incipient fault detection method for underground cables[J]. IET Generation,Transmission & Distribution, 2015, 9(14):1988-1997. doi: 10.1049/gtd2.v9.14
[11]	ZHANG W, XIAO X, ZHOU K, et al. Multi-cycle incipient fault detection and location for medium voltage underground cable[J]. IEEE Transactions on Power Delivery, 2017, 32(3):1450-1459. doi: 10.1109/TPWRD.2016.2615886
[12]	熊思衡, 刘亚东, 方健, 等. 配电线路早期故障辨识方法[J]. 高电压技术, 2020, 46(22):3970-3976.
	XIONG Siheng, LIU Yadong, FANG Jian, et al. Detection method of incipient faults of power system lines[J]. High Voltage Engineering, 2020, 46(22):3970-3976.
[13]	汪颖, 卢宏, 杨晓梅, 等. 堆叠自动编码器与S变换相结合的电缆早期故障识别方法[J]. 电力自动化设备, 2018, 38(8):117-124.
	WANG Ying, LU Hong, YANG Xiaomei, et al. Cable incipient fault identification based on stacked autoencoder and S-transform[J]. Electric Power Automation Equipment, 2018, 38(8):117-124.
[14]	汪颖, 孙建风, 肖先勇, 等. 基于优化卷积神经网络的电缆早期故障分类识别[J]. 电力系统保护与控制, 2020, 48(7):10-18.
	WANG Ying, SUN Jianfeng, XIAO Xianyong, et al. Cable incipient fault classification and identification based on optimized convolution neural network[J]. Power System Protection and Control, 2020, 48(7):10-18.
[15]	WENG Y, CUI Q, GUO M. Transform waveforms into signature vectors for general-purpose incipient fault detection[J]. IEEE Transactions on Power Delivery, 2022, 37(6):4559-4569. doi: 10.1109/TPWRD.2022.3151110
[16]	刘健, 张小庆, 张志华, 等. 提升小电流接地系统单相接地故障处理能力[J]. 供用电, 2021, 38(10):52-56.
	LIU Jian, ZHANG Xiaoqing, ZHANG Zhihua, et al. Solutions to improve the single-phase grounding fault management of neural non-effective grounded systems[J]. Distribution & Utilization, 2021, 38(10):52-56.
[17]	楚恬歆, 张文海, 瞿科, 等. 小电流接地系统接地型早期故障扰动特征分析[J]. 电力系统保护与控制, 2021, 49(18):52-61.
	CHU Tianxin, ZHANG Wenhai, QU Ke, et al. Grounded incipient fault analysis in a non-solidly grounding system[J]. Power System Protection and Control, 2021, 49(18):52-61.
[18]	楚恬歆, 张文海, 瞿科, 等. 基于复合判据的小电流接地系统接地型早期故障检测[J]. 高电压技术, 2022, 48(3):1022-1031.
	CHU Tianxin, ZHANG Wenhai, QU Ke, et al. The grounded incipient fault detection based on compound criterion in non-solidly grounding system[J]. High Voltage Engineering, 2022, 48(3):1022-1031.
[19]	陶维青, 夏熠, 陆鼎堃. S变换熵理论及其在电力系统故障检测中的应用研究[J]. 合肥工业大学学报, 2016, 39(1):40-45.
	TAO Weiqing, XIA Yi, LU Dingkun. Study of S-transform entropy theory and its application in fault detection of electric power system[J]. Journal of Hefei University of Technology, 2016, 39(1):40-45.
[20]	程玉胜, 宋帆, 王一宾, 等. 基于专家特征的条件互信息多标记特征选择算法[J]. 计算机应用, 2020, 40(2):503-509. doi: 10.11772/j.issn.1001-9081.2019091626
	CHENG Yusheng, SONG Fan, WANG Yibin, et al. Multi-label feature selection algorithm based on conditional mutual information of expert feature[J]. Journal of Computer Applications, 2020, 40(2):503-509. doi: 10.11772/j.issn.1001-9081.2019091626
[21]	毛莺池, 曹海, 平萍, 等. 基于最大联合条件互信息的特征选择[J]. 计算机应用, 2019, 39(3):734-741. doi: 10.11772/j.issn.1001-9081.2018081694
	MAO Yingchi, CAO Hai, PING Ping, et al. Feature selection based on maximum conditional and joint mutual information[J]. Journal of Computer Application, 2019, 39(3):734-741.
[22]	盖晓平, 王冬青, 赵喜兰, 等. 利用概率统计特性的保护告警信息特征降维方法[J]. 电网技术, 2021, 45(5):2017-2024.
	GAI Xiaoping, WANG Dongqing, ZHAO Xilan, et al. Feature reduction method for alarm information protection with probability statistical characteristics[J]. Power System Technology, 2021, 45(5):2017-2024.
[23]	李扬, 顾雪平. 基于改进最大相关最小冗余判据的暂态稳定评估特征选择[J]. 中国电机工程学报, 2013, 33(34):179-186.
	LI Yang, GU Xueping. Feature selection for transient stability assessment based on improved maximal relevance and minimal redundancy criterion[J]. Proceedings of the CSEE, 2013, 33(34):179-186.
[24]	孙志鹏, 孙志龙, 魏建. 基于决策树支持向量机算法的电力变压器故障诊断研究[J]. 电气工程学报, 2019, 14(4):42-45.
	SUN Zhipeng, SUN Zhilong, WEI Jian. Research on power transformer fault diagnosis based on decision tree support vector machine[J]. Journal of Electrical Engineering, 2019, 14(4):42-45.
[25]	徐世晖. 基于改进在线最小二乘支持向量机电池故障诊断方法研究[D]. 哈尔滨: 哈尔滨理工大学, 2020.
	XU Shihui. Research on battery fault diagnosis method based on improved online least square support vector machine[D]. Harbin:Harbin University of Science and Technology, 2021.
[26]	邓乃扬, 田英杰. 数据挖掘中的新方法——支持向量机[M]. 北京: 科学出版社, 2004.
	DENG Naiyang, TIAN Yingjie. A new method of data mining:Support vector machines[M]. Beijing: China Science Publishing & Media Ltd., 2004.
[27]	郑含博, 王伟, 李晓纲, 等. 基于多分类最小二乘支持向量机和改进粒子群优化算法的电力变压器故障诊断方法[J]. 高电压技术, 2014, 40(11):3424-3429.
	ZHENG Hanbo, WANG Wei, LI Xiaogang, et al. Fault diagnosis method of power transformers using multi-class LS-SVM and improved PSO[J]. High Voltage Engineering, 2014, 40(11):3424-3429.
[28]	LIN S, YING K, CHEN S, et al. Particle swarm optimization for parameter determination and feature selection of support vector machines[J]. Expert Systems with Applications, 2008, 35(4):1817-1824. doi: 10.1016/j.eswa.2007.08.088
[29]	冯茜, 李擎, 全威, 等. 多目标粒子群优化算法研究综述[J]. 工程科学学报, 2021, 43(6):745-753.
	FENG Qian, LI Qing, QUAN Wei, et al. Overview of multiobjective particle swarm optimization algorithm[J]. Chinese Journal of Engineering, 2021, 43(6):745-753.
[30]	王文凯, 邓斌. 基于DAE-IPSO-SVM的电缆早期故障识别方法[J]. 国外电子测量技术, 2021, 40(8):29-35.
	WANG Wenkai, DENG Bin. Cable incipient fault identification method based on DAE-IPSO-SVM[J]. Foreign Electronic Measurement Technology, 2021, 40(8):29-35.
[31]	CHANG G, HONG Y, LI G. A hybrid intelligent approach for classification of incipient faults in transmission network[J]. IEEE Transactions on Power Delivery, 2019, 34(4):1785-1794. doi: 10.1109/TPWRD.61
[32]	杨杏丽. 分类学习算法的性能度量指标综述[J]. 计算机科学, 2021, 48(8):209-219. doi: 10.11896/jsjkx.200900216
	YANG Xingli. Survey of performance measure index of classification learning algorithm[J]. Computer Science, 2021, 48(8):209-219. doi: 10.11896/jsjkx.200900216

算法参数	数值
粒子群大小	10
加权因子取值范围	[0.5, 0.9]
学习因子c₁、c₂	1.5、1.7
最大迭代次数	100
核函数宽度取值范围	[2^-5, 2²]
惩罚系数取值范围	[1, 100]
粒子位置取值范围	[-10, 10]
粒子速度取值范围	[-10, 10]

线路	相序	电阻/(Ω/km)	电感/(mH/km)	电容/(μF/km)
架空线	正序	0.080	2.425	0.009
架空线	零序	0.240	0.803	0.006
电缆	正序	0.099	0.968	0.339
电缆	零序	0.970	0.225	0.280

参数	数值
故障距离D/m	100~14 600 (步长500 m)
接入电容器值C/mF	1~7 000 mF(步长100 mF)
接入变压器S/(kV·A)	1~7 000 (步长100 kV·A)
接入恒定阻抗Z/Ω	1~7 000 (步长100 Ω)
电弧时间常数τ/ms	0.2~0.4 (步长0.2 ms)
电弧特征电压u₀/V	100~1 300(步长200 V)
电弧特征电阻r₀/Ω	0.001~0.013 (步长0.003 Ω)

测试集	训练集	P₁(%)	P₂(%)	P₃(%)	P₄(%)
1	2~5	0.979 2	1	0.958 3	0.978 7
2	1,3~5	0.965 3	0.985 5	0.944 4	0.964 5
3	1~2,4~5	0.965 3	0.971 8	0.958 3	0.965
4	1~3,5	0.958 3	1	0.916 7	0.956 5
5	1~4	0.979 2	0.985 9	0.972 2	0.979 0
平均值		0.969 46	0.988 64	0.949 98	0.968 74

特征集	P₁	P₂	P₃	P₄	处理时间/s
最优特征	0.969 46	0.988 64	0.949 98	0.968 74	129.244
原始数据	0.904 16	0.934 66	0.868 54	0.900 62	1 322.359