基于改进麻雀算法-支持向量机的输电线路故障诊断Transmission Line Fault Diagnosis Based on Support Vector Machine Optimized by Improved Sparrow Search Algorithm
石万宇,魏军强,赵云灏
SHI Wanyu,WEI Junqiang,ZHAO Yunhao
摘要(Abstract):
建立了一种数据驱动的输电线路雷击故障分类诊断模型,提出了改进的麻雀搜索算法优化支持向量机参数的方法,并实现输电线路雷击故障分类诊断。利用自适应t分布算法改进麻雀搜索算法,克服了麻雀搜索算法易陷入局部最优的缺点,从而提高麻雀搜索方法优化能力。利用改进后的算法对SVM(支持向量机)参数进行寻优,提高分类准确率。利用输电线路故障分析与雷电定位系统数据,选取关键故障特征构成样本数据集,利用t SSA-SVM(自适应t分布改进的麻雀搜索算法-支持向量机)构建输电线路故障诊断分类模型。实例分析表明,该模型可有效实现基于数据分析的输电线路雷击故障诊断分类,所提算法应用于故障诊断分类的准确率优于经典SVM,而且算法效率显著优于PSO-SVM(粒子群算法-支持向量机)故障诊断分类算法。
This paper establishes a data-driven transmission line lightning fault classification and diagnosis model, proposes an improved sparrow search algorithm to optimize the parameters of the support vector machine, and realizes the transmission line lightning fault classification and diagnosis. First, the adaptive t-distribution algorithm is used to improve the sparrow search algorithm, which overcomes the shortcomings of the sparrow search algorithm that is easy to fall into the local optimum, thereby improving the optimization ability of the sparrow search method. Then the improved algorithm is used to optimize the SVM parameters to improve the classification accuracy. Finally, the transmission line fault analysis and lightning location system data are used, and the key fault characteristics are selected to form a sample data set, and the tSSA-SVM is used to construct the transmission line fault diagnosis and classification model. Case analysis shows that the model can effectively realize the lightning fault diagnosis and classification of transmission lines based on data analysis. The accuracy of the proposed algorithm in fault diagnosis classification is higher than that of the classic SVM, and the algorithm efficiency is superior to fault diagnosis and classification algorithm based on the PSO-SVM(particle swarm optimization-support vector machine).
关键词(KeyWords):
麻雀搜索算法;自适应t分布;支持向量机;输电线路;雷击故障
sparrow search algorithm;adaptive t-distribution;support vector machine;transmission line;lightning failure
基金项目(Foundation): 国家重点研发计划(2021YFE0102400)
作者(Author):
石万宇,魏军强,赵云灏
SHI Wanyu,WEI Junqiang,ZHAO Yunhao
DOI: 10.19585/j.zjdl.202111006
参考文献(References):
- [1]胡毅,刘凯,吴田,等.输电线路运行安全影响因素分析及防治措施[J].高电压技术,2014,40(11):3491-3499.
- [2]陆佳政,周特军,吴传平,等.某省级电网220 kV及以上输电线路故障统计与分析[J].高电压技术,2016,42(1):200-207.
- [3]刘道兵,顾雪平,梁海平,等.电网故障诊断完全解析模型的解集评价与最优解求取[J].中国电机工程学报,2014,34(31):5668-5676.
- [4]费思源.大数据技术在配电网中的应用综述[J].中国电机工程学报,2018,38(1):85-96.
- [5]宁琦,耿读艳,王晨旭,等.基于多尺度排列熵及PSOSVM的输电线路故障判别[J].电子测量与仪器学报,2019,33(7):173-180.
- [6]童晓阳,罗忠运.基于模糊支持向量机的输电线路故障模糊分类方法及其降维显示[J].高电压技术,2015,41(7):2276-2282.
- [7]安然,吴俊勇,邵美阳,等.基于随机矩阵和支持向量机的两阶段配电网综合优化[J].供用电,2019,36(10):2-8.
- [8]孙晓明,秦亮,刘涤尘.基于GMAPM和SOM-LVQ-ANN的输电线路故障综合识别方法[J].武汉大学学报(工学版),2019,52(12):1079-1090.
- [9]杨赛昭,向往,张峻榤,等.基于人工神经网络的架空柔性直流电网故障检测方法[J].中国电机工程学报,2019,39(15):4416-4430.
- [10]朱文博.基于改进决策树的轨道电路故障诊断方法研究[D].成都:西南交通大学,2017:33-61.
- [11]李兵,韩睿,何怡刚,等.改进随机森林算法在电机轴承故障诊断中的应用[J].中国电机工程学报,2020,40(4):1310-1319.
- [12]LUO X,KEZUNOVIC M.Implementing fuzzy reasoning Petri-nets for fault section estimation[J].IEEE Transactions on Power Delivery,2008,23(2):676-685.
- [13]毛彪,杨松,李英顺.改进动态因果图与模糊推理融合故障诊断方法[J].计算机工程与应用,2020,56(3):259-265.
- [14]周绍磊,廖剑,史贤俊.RBF-SVM的核参数选择方法及其在故障诊断中的应用[J].电子测量与仪器学报,2014,28(3):240-246.
- [15]陈歆技,丁同奎,张钊.蚁群算法在配电网故障定位中的应用[J].电力系统自动化,2006,30(5):74-77.
- [16]应林志,刘天琪,王建全.基于改造粒子群游的超高压故障限流器全局优化配置算法[J].电力自动化设备,2018,38(2):145-152.
- [17]ZAHARIS Z D,GRAVAS I P,YIOULTSIS T V,et al.Exponential log-periodic antenna design using improved particles warm optimization with velocity mutation[J].IEEE Transactions on Magnetics,2017,53(6):1-4.
- [18]梅飞,梅军,郑建勇,等.粒子群优化的KFCM及SVM诊断模型在断路器故障诊断中的应用[J].中国电机工程学报,2013,33(36):134-141.
- [19]董伟,杨强,葛磊蛟,等.基于混合智能模型的分布式风力发电预测方法[J].供用电,2019,36(1):34-39.
- [20]XUE J,SHEN B.A novel swarm intelligence optimization approach:sparrow search algorithm[J].Systems Science&Control Engine ering An Open Access Journal,2020,8(1):22-34.
- [21]JEONG W,KANG M,KIM S,et al.Full waveform inversion using student’s t distribution:A numerical study for elastic waveform inversion and simultaneous-source method[J].Pure and Applied Geophysice,2015,172(6):1491-1590.
- [22]吕鑫,慕晓冬,张钧,等.混沌麻雀搜索优化算法[J].北京航空航天大学学报,2021,47(8):1712-1720.
- [23]韩斐斐,刘升,赵齐辉.基于t分布的自适应花授粉算法[J].数学的实践与认识,2019,49(2):157-165.
- [24]孙瑶琴.基于改进FOA优化的SVM在故障诊断中的应用[J].机械强度,2017,39(2):285-290.