登录    |    注册

您好,欢迎来到中国测试科技资讯平台!

在线投稿 专家审稿 编辑办公 咨询评价
首页> 《中国测试》期刊 >本期导读>基于PCA-KELM和AT的互感器故障诊断

基于PCA-KELM和AT的互感器故障诊断

77    2019-11-28

0.00

全文售价

作者:王昕1,2, 曹文彬3, 曹敏1,2, 赵旭1,2, 赵艳峰1,2, 李翔1,2, 蒋婷婷1,2, 田猛3, 王先培3

作者单位:1. 南方电网电能计量重点实验室, 云南 昆明 650217;
2. 云南电网有限责任公司电力科学研究院, 云南 昆明 650217;
3. 武汉大学电子信息学院, 湖北 武汉 430072


关键词:互感器;故障诊断;反正切变换;主元分析法;核极限学习机


摘要:

针对当前互感器故障诊断算法的准确率不高的问题,提出基于主元分析法核极限学习机(principal components analysis- kernel extreme learning machine,PCA-KELM)和反正切变换(arctangent transform,AT)的互感器故障诊断方法。AT可以改变互感器故障的数据结构,重新调节电压电流等数据的相关比例;PCA提取数据特征,不仅可以减小其维数,还可以保留所需的识别信息;KELM算法能够利用其结构参数来逼近非线性函数,且无需设定网络隐含层节点。通过构建互感器诊断模型,并给出互感器测量参数,将KELM和PCA相互结合对数据进行仿真。将所提出的方法分别与支持向量机(support vector machine,SVM)和原始的ELM的性能进行比较。仿真结果表明:所提出的互感器故障诊断方法在诊断精度上均优于SVM和ELM,且诊断速度同ELM相当,较SVM有明显提升,故障诊断准确率达到98.53%,可为故障信息有限情况下的互感器故障诊断提供参考。


Transformer fault diagnosis based on PCA-KELM and AT
WANG Xin1,2, CAO Wenbin3, CAO Min1,2, ZHAO Xu1,2, ZHAO Yanfeng1,2, LI Xiang1,2, JIANG Tingting1,2, TIAN Meng3, WANG Xianpei3
1. Key Laboratory of Energy Metering of Southern Power Grid, Kunming 650217, China;
2. Electric Power Research Institute, Yunnan Power Grid Co., Ltd., Kunming 650217, China;
3. School of Electronic Information, Wuhan University, Wuhan 430072, China
Abstract: The accuracy rate of the current transformer fault diagnosis algorithm is not high, a transformer fault based on principal component analysis-kernel extreme learning machine (PCA-KELM) and arctangent transform (AT, arctangent transform) diagnosis method is proposed. Here, AT can change the data structure of transformer faults, re-adjust the correlation ratio of voltage and current data, PCA can extract data features not only can reduce its dimension, but also retain the required identification information, KELM algorithm can use it Structural parameters to approximate nonlinear functions without the need to set network hidden layer nodes. By constructing a transformer diagnostic model and giving the transformer measurement parameters, the KELM and PCA are combined to simulate the data. The proposed method is compared with the performance of the support vector machine (SVM) and the original ELM. The simulation results show that the proposed transformer fault diagnosis method is superior to SVM and ELM in diagnosis accuracy, and the diagnostic speed is comparable to ELM, which is significantly improved compared with SVM. The accuracy of fault diagnosis reaches 98.53%, which can be limited for fault information. Provide a reference for transformer fault diagnosis.
Keywords: transformer;fault diagnosis;arctangent transform;principal components analysis;kernel extreme learning machine
2019, 45(11):72-78  收稿日期: 2018-09-27;收到修改稿日期: 2018-12-11
基金项目: 云南电网公司重点项目(YN2014-2-001)
作者简介: 王昕(1969-),女,云南昆明市人,高级工程师,从事电能计量等领域的研究
参考文献
[1] 肖燕彩, 陈秀海, 朱衡君. 遗传支持向量机在电力变压器故障诊断中的应用[J]. 上海交通大学学报, 2007, 41(11):1878-1886
[2] FEI S W, LIU C L, MIAO Y B. Support vector machine with genetic algorithm for forecasting of key-gas ratios in oil-immersed transformer[J]. Expert Systems with Applications, 2009, 36(4):6326-6331
[3] HUANG G B, CHEN L. Convex incremental extreme learning machine[J]. Neurocomputing, 2007, 70(16):3056-3062
[4] HUANG G B, WANG D H, LAN Y. Extrem learning machines:a survey[J]. International Journal of Machine Learning and Cybernetics, 2011, 2(2):107-122
[5] FENG G, HUANG G B, LIN Q, et al. Error minimized extreme learning machine with growth of hidden nodes and incremental learning[J]. IEEE Transactions on Neural Networks, 2009, 20(8):1352-1357
[6] 苑津莎, 张利伟, 王瑜, 等. 基于极限学习机的变压器故障诊断方法研究[J]. 电测与仪表, 2013, 50(12):21-26
[7] SCARDAPANE S, COMMINIELLO D, SCARPINITI M, et al. Online sequential extreme learning machine with kernels[J]. IEEE Transactions on Neural Networks and Learning Systems, 2015, 26(9):2214-2220
[8] ZHOU Y C, PENG J T, CHEN C L P. Extreme learning machine with composite kernels for hyperspectral image classification[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2015, 8(6):2351-2360
[9] 李军, 李大超. 基于优化核极限学习机的风电功率时间序列预测[J]. 物理学报, 2016, 65(13):39-48
[10] 杨锡运, 关文渊, 刘玉奇, 等. 基于粒子群优化的核极限学习机模型的风电功率区间预测方法[J]. 中国电机工程学报, 2015, 35(S1):146-153
[11] 陈绍炜, 柳光峰, 冶帅. 基于核极限学习机的模拟电路故障诊断研究[J]. 西北工业大学学报, 2015, 33(2):290-294
[12] 张英堂, 马超, 李志宁, 等. 基于快速留一交叉验证的核极限学习机在线建模[J]. 上海交通大学学报, 2014, 48(5):641-646
[13] 王杰, 郭晨龙. 小波核极限学习机分类器[J]. 微电子学与计算机, 2013, 30(10):73-76+80
[14] LI Y B, LIU C, ZHANG S Y, et al. Reproducing polynomial kernel extreme learning machine[J]. Journal of Advanced Computational Intelligence and Intelligent Informatics, 2017, 21(5):795-802
[15] LI P J, SHI Y, WANG H D, et al. Ordered Code-based Kernel Extreme Learning Machine for Ordinal Regression[J]. Journal of Electronics and Information Technology, 2018, 40(6):1287-1293
[16] CAO L L, SUN F C, LI H B, et al. Advancing the incremental fusion of robotic sensory features using online multi-kernel extreme learning machine[J]. Frontiers of Computer Science, 2017, 11(2):276-289
[17] 胡梦月, 胡志坚, 仉梦林, 等. 基于改进AdaBoost.RT和KELM的风功率预测方法研究[J]. 电网技术, 2017, 41(2):536-542
[18] 汤庆峰, 刘念, 张建华, 等. 基于EMD-KELM-EKF与参数优选的用户侧微电网短期负荷预测方法[J]. 电网技术, 2014, 38(10):2691-2699