基于改进支持向量机的有源配电网单相断线故障检测方法

doi:10.12096/j.2096-4528.pgt.260117

摘要/Abstract

摘要：

目的断线故障的及时、准确检测对保障配电网的正常安全运行至关重要，而目前的单相断线故障检测的传统判据在有源配电网中的应用存在一定的局限性，因此，提出了一种基于改进支持向量机对多种电气特征量进行融合的单相断线故障检测方法。方法首先，建立了兼具启动判据、传统判据、有源判据的电气特征量指标体系。其次，通过开关量化法对启动判据进行处理。然后，通过核主成分分析方法从启动判据以外的特征指标体系中筛除低贡献率的特征指标。最后，将降维后的数据输入支持向量机，通过麻雀搜索算法完成支持向量机参数优化，得到断线故障检测模型。结果在改进IEEE15节点模型上进行的仿真算例表明，所提方法可将有效实现特征量的降维，较单一判据提升了8.87%的检测准确率。结论该方法解决了单相断线故障检测的传统判据容易失效的问题，能有效完成不同场景下的故障检测。

关键词: 有源配电网, 单相断线, 支持向量机, 麻雀搜索算法, 核主成分分析

Abstract:

Objectives Timely and accurate detection of line-break faults is crucial for ensuring the normal and safe operation of distribution networks. However, the application of traditional criteria for single-phase line-break fault detection in active distribution networks has certain limitations. Therefore, a single-phase line-break fault detection method based on an improved support vector machine and integrating multiple electrical characteristic quantities is proposed. Methods First, an electrical characteristic index system incorporating startup criteria, traditional criteria, and active criteria is established. Second, the startup criteria are processed using the switching quantization method. Subsequently, kernel principal component analysis is employed to eliminate low-contribution feature indicators from the characteristic index system excluding the startup criteria. Finally, the dimensionality-reduced data are input into the support vector machine, and the sparrow search algorithm is utilized to optimize the parameters of the support vector machine, thereby obtaining the line-break fault detection model. Results Simulation case studies conducted on a modified IEEE 15-node model demonstrate that the proposed method effectively realizes dimensionality reduction of the characteristic quantities and improves the detection accuracy by 8.87% compared to using a single criterion. Conclusions The proposed method solves the problem that traditional criteria for single-phase line-break fault detection are prone to failure, and can effectively achieve fault detection under different scenarios.

Key words: active distribution network, single-phase line-break, support vector machine, sparrow search algorithm, kernel principal component analysis

中图分类号:

TK 01

吴宇轩, 欧阳森, 杨向宇, 陈汉栋, 黎人玮, 廖键. 基于改进支持向量机的有源配电网单相断线故障检测方法[J]. 发电技术, 2026, 47(1): 185-194.

Yuxuan WU, Sen OUYANG, Xiangyu YANG, Handong CHEN, Renwei LI, Jian LIAO. A Single-Phase Line-Break Fault Detection Method for Active Distribution Networks Based on Improved Support Vector Machine[J]. Power Generation Technology, 2026, 47(1): 185-194.

图/表 13

图1 正常运行时有源配电网的等效电路

Fig. 1 Equivalent circuit of active distribution network during normal operation

图2 单相断线故障下配电网的复合序网

Fig. 2 Composite sequence network of distribution network under single-phase line-break fault

图3 有源配电网断线故障特征指标体系

Fig. 3 Characteristic index system for line-break faults in active distribution network

图4 断线故障检测示意图

Fig. 4 Schematic diagram of line-break fault detection

图5 基于SVM的单相断线故障检测模型图

Fig.5 Model diagram of SVM-based single-phase line-break fault detection

图6 单相断线故障检测流程

Fig. 6 Flowchart for single-phase line-break fault detection

图7 基于改进IEEE 15节点的单相断线故障仿真拓扑

Fig. 7 Simulation topology of single-phase line-break fault based on improved IEEE 15-node system

表1 单一判据识别准确率

Tab. 1 Detection accuracy of single criterion

单一判据	识别率/%
负序电压	67.26
零序电压	56.25
正序电流	68.75
负序电流	68.75
相电压幅值	65.77
线电压幅值	39.29
负序正序电流比	72.02
DG电流变化率	85.71

图8 KPCA后各特征贡献率

Fig. 8 Contribution rate of each feature after KPCA

表2 KPCA和PCA主成分的累计贡献率对比

Tab. 2 Comparison of cumulative contribution rates between KPCA and PCA principal components

成分	KPCA累积贡献率/%	PCA累积贡献率/%
1	78.73	51.31
2	90.38	71.21
3	97.31	86.54
4	98.37	92.21
5	98.93	95.48

图9 GA-SVM、PSO-SVM、SSA-SVM的适应度曲线对比

Fig. 9 Comparison of fitness curves among GA-SVM, PSO-SVM, and SSA-SVM

表3 各机器学习算法的检测情况

Tab. 3 Detection performance of various machine learning algorithms

算法	正确检测次数	误判次数	漏判次数
SVM	80	2	14
RF	84	0	12
BP	81	0	15
PSO-SVM	86	4	6
GA-SVM	85	3	8
SSA-SVM	92	0	4

表4 融合判据测试集识别正确率

Tab. 4 Detection accuracy of fused-criteria test set

融合判据	平均识别率/%
SVM	83.18
RF	87.34
BP	84.74
PSO-SVM	88.85
GA-SVM	88.96
SSA-SVM	94.58

参考文献 26

[1]	陈锦荣，李赫，谭振鹏，等．基于电压相序突变信息的配电线路单相断线故障方向识别方法[J]．供用电，2022，39(9)：35-41．
	CHEN J R， LI H， TAN Z P，et al ．Identification method of single-phase disconnection fault direction for distribution lines based on voltage phase sequence mutation information[J]．Distribution & Utilization，2022，39(9)：35-41．
[2]	张明泽，栾文鹏，艾欣，等．基于边缘计算的低压配电网健康状态评价方法[J]．电测与仪表，2024，61(5)：91-97．
	ZHANG M Z， LUAN W P， AI X，et al ．A health status assessment method for low-voltage distribution network based on edge computing[J]．Electrical Measurement & Instrumentation，2024，61(5)：91-97．
[3]	陈春，阳汉琨，肖轩怡，等．面向配电网低压用户的停电区域研判方法[J]．电力系统保护与控制，2025，53(16)：28-38．
	CHEN C， YANG H K， XIAO X Y，et al ．A method for identifying the outage area of low-voltage distribution network users[J]．Power System Protection and Control，2025，53(16)：28-38．
[4]	徐玉韬，冯起辉，谈竹奎，等．考虑转供与重构协同的多端柔性互联配电网供电恢复策略[J]．电工技术学报，2024，39(9)：2696-2709．
	XU Y T， FENG Q H， TAN Z K，et al ．Optimal power restoration strategy for multi-terminal flexible interconnected distribution networks based on flexible interconnection device and network reconfiguration[J]．Transactions of China Electrotechnical Society，2024，39(9)：2696-2709．
[5]	郑凯元，于希娟，杨禹锡，等．城市安全可靠供电需求下的韧性配电网实践[J]．电力建设，2025，46(2)：35-48．
	ZHENG K Y， YU X J， YANG Y X，et al ．Resilient distribution network practice of cities under the demand for safe and reliable power supply[J]．Electric Power Construction，2025，46(2)：35-48．
[6]	胡满琳，李楠，李一鸣，等．基于负序分量的含光伏电源配电网故障区段定位方法[J]．中国电力，2024，57(5)：188-199．
	HU M L， LI N， LI Y M，et al ．Fault location method for distribution network with photovoltaic power based on negative sequence component[J]．Electric Power，2024，57(5)：188-199．
[7]	李成钢，刘亚东，杨雪凤，等．基于变分模态分解与空洞卷积神经网络的配电网故障选线方法[J]．电网与清洁能源，2024，40(2)：110-118．
	LI C G， LIU Y D， YANG X F，et al ．A method of fault line selection for distribution networks based on variational mode decomposition and dilated convolutional neural network[J]．Power System and Clean Energy，2024，40(2)：110-118．
[8]	王可淇，赵子涵，钟俊，等．基于改进模拟退火遗传算法的配电网动态故障恢复策略[J]．智慧电力，2024，52(6)：16-22．
	WANG K Q， ZHAO Z H， ZHONG J，et al ．Dynamic fault recovery strategy for distribution network based on improved simulated annealing genetic algorithm[J]．Smart Power，2024，52(6)：16-22．
[9]	陈雅芳．低压系统断零与缺相故障诊断技术及解决方案研究[D]．广州：广东工业大学，2016．
	CHEN Y F ．Research on fault diagnosis technology and solution of zero break and phase loss in low voltage system[D]．Guangzhou：Guangdong University of Technology，2016．
[10]	吴照裕．小电阻接地配电网断线与高阻接地故障保护策略研究[D]．广州：华南理工大学，2021．
	WU Z Y ．Study on protection strategy of disconnection and high resistance grounding fault in distribution network with small resistance grounding[D]．Guangzhou：South China University of Technology，2021．
[11]	秦龙宇．有源配电网单相断线故障检测与区段定位方法研究[D]．济南：山东大学，2022． doi:10.1109/cieec50170.2021.9510437
	QIN L Y ．Research on fault detection and section location method of single-phase disconnection in active distribution network[D]．Jinan：Shandong University，2022． doi:10.1109/cieec50170.2021.9510437
[12]	薛永端，陈梦琦，曹丽丽，等．不接地系统单相断线故障电压分析[J]．中国电机工程学报，2021，41(4)：1322-1333．
	XUE Y D， CHEN M Q， CAO L L，et al ．Analysis of voltage characteristics of single-phase disconnection fault in ungrounded distribution system[J]．Proceedings of the CSEE，2021，41(4)：1322-1333．
[13]	常仲学，宋国兵，张维．配电网单相断线故障的负序电压电流特征分析及区段定位[J]．电网技术，2020，44(8)：3065-3072．
	CHANG Z X， SONG G B， ZHANG W ．Characteristic analysis and fault segment location on negative sequence voltage and current of single phase line breakage fault in distribution network[J]．Power System Technology，2020，44(8)：3065-3072．
[14]	王开科，熊小伏，肖扬，等．基于负序电流的主动配电网单相断线故障保护方法[J]．电力系统保护与控制，2021，49(6)：10-18．
	WANG K K， XIONG X F， XIAO Y，et al ．Single-phase break fault protection method for an active distribution network based on negative sequence current[J]．Power System Protection and Control，2021，49(6)：10-18．
[15]	杜雪，欧阳金鑫，龙晓轩，等．基于分布式电源电流变化率的主动配电网单相断线保护方法[J]．电力系统保护与控制，2020，48(22)：41-48．
	DU X， OUYANG J X， LONG X X，et al ．A protection method of a single-phase break fault for an active distribution network based on current change rates of distributed generation[J]．Power System Protection and Control，2020，48(22)：41-48．
[16]	YANG F， LI H， HU W，et al ．Identification of single-phase line break fault direction based on local voltage information in small current grounding system considering the impact of DG[J]．IEEE Access，2023，11：120754-120765． doi:10.1109/access.2023.3327337
[17]	ZHANG X， ZHANG S， DONG W，et al ．Method for locating single-phase disconnection faults in a smart distribution system based on refined composite multiscale permutation q-complexity entropy and random forest[J]．IEEE Sensors Journal，23(11)：11925-11935． doi:10.1109/jsen.2023.3267066
[18]	慕静茹，喻锟，曾祥君，等．考虑多扰动因子的含光伏电源低压台区漏电故障检测[J]．南方电网技术，2024，18(10)：130-141．
	MU J， YU K， ZENG X，et al ．Leakage fault detection in low-voltage distribution areas with photovoltaic power supply considering multiple disturbance factors[J]．Southern Power System Technology， 2024， 18(10)： 130-141．
[19]	门茂琛，吴焕昭，徐铭铭，等．基于变分模态分解-支持向量机混合算法的TN系统大阻抗接地故障配电线路保护[J]．科学技术与工程，2023，23(21)：9244-9250．
	MEN M C， WU H Z， XU M M，et al ．Protection of distribution lines for large impedance grounding faults in TN systems based on a hybrid algorithm combining variational mode decomposition and support vector machine[J]．Science Technology and Engineering，2023，23(21)：9244-9250．
[20]	林师远，黄雄，吴天杰，等．基于TNN-BL模型的低压配电网断零与缺相故障检测方法研究[J]．电机与控制应用，2024，51(10)：40-50．
	LIN S Y， HUANG X， WU T J，et al ．Research on broken neutral wire and phase loss fault detection method in low-voltage distribution network based on TNN-BL model[J]．Electric Machines & Control Application，2024，51(10)：40-50．
[21]	曾祥君，陈磊，喻锟，等．基于配电网双端信息融合的单相断线故障实时监测方法[J]．电力科学与技术学报，2020，35(3)：12-18．
	ZENG X J， CHEN L， YU K，et al ．A real time monitoring method for the single-line break fault based on dual-terminal information in the distribution network[J]．Journal of Electric Power Science and Technology，2020，35(3)：12-18．
[22]	杜雪，肖扬，付滨，等．基于负序电流比的主动配电网断线故障保护方法[J]．电测与仪表，2021，58(7)：57-66．
	DU X， XIAO Y， FU B，et al ．Break fault protection method of active distribution network based on negative-sequence current ratio[J]．Electrical Measurement & Instrumentation，2021，58(7)：57-66．
[23]	曹现刚，叶煜，赵友军，等．基于KPCA-LSTM的旋转机械剩余使用寿命预测[J]．振动与冲击，2023，42(24)：81-91．
	CAO X G， YE Y， ZHAO Y J，et al ．Remaining useful life prediction of rotating machinery based on KPCA-LSTM[J]．Journal of Vibration and Shock，2023，42(24)：81-91．
[24]	赵国钦，蓝茂蔚，李杨，等．基于最小二乘支持向量机的火电厂烟气含氧量预测模型优化研究[J]．发电技术，2023，44(4)：534-542． doi:10.12096/j.2096-4528.pgt.21088
	ZHAO G Q， LAN M W， LI Y，et al ．Study on optimization of prediction model of flue gas oxygen content in thermal power plant based on least squares support vector machine[J]．Power Generation Technology，2023，44(4)：534-542． doi:10.12096/j.2096-4528.pgt.21088
[25]	XUE J， SHEN B ．A novel swarm intelligence optimization approach：sparrow search algorithm[J]．Systems Science & Control Engineering，2020，8(1)：22-34． doi:10.1080/21642583.2019.1708830
[26]	汪自虎，王文天，惠慧，等．基于近邻成分分析与优化核极限学习机的光伏接入配电网漏电识别[J]．高压电器，2024，60(6)：203-211．
	WANG Z H， WANG W T， HUI H，et al ．Leakage identification in photovoltaic-integrated distribution networks based on nearest component analysis and optimized kernel extreme learning machine[J]．High Voltage Apparatus，2024，60(6)：203-211．