不完备故障类别下基于Multi-SVDD的高压隔离开关故障诊断方法

doi:10.19595/j.cnki.1000-6753.tces.180026

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (23829 KB)
输出: BibTeX | EndNote (RIS)

摘要

针对高压隔离开关故障诊断时特征库中故障类别不完备的问题,提出了基于多重支持向量域描述（Multi-SVDD）的故障诊断方法。首先通过主成分分析将正常和已知故障样本特征量按贡献度进行排序作为新的特征向量,并以特征量贡献度构造加权高斯核函数,提高对类间特征差异的辨识能力。然后利用粒子群算法对核参数进行优化,提高模型的推广能力和对样本类别识别的正确率。其次对正常和已知故障样本集进行训练,建立描述隔离开关不同工作状态的超球体作为预测模型。最后利用Multi-SVDD对样本空间进行划分并计算待测样本点至各超球体中心的距离,确定样本所属的种类。试验结果表明,该方法可以有效处理高压隔离开关故障诊断中故障类别不完备的问题,在诊断出已知故障的同时可对未知故障给出判断。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	陈士刚
	关永刚
	张小青
	杨元威
	张一茗

关键词 ：隔离开关, 类别不完备, 多重支持向量域, 核函数, 故障诊断

Abstract：

Focusing on the problem of incomplete fault types existing in high voltage isolating switch fault diagnosis, a fault diagnosis method based on multi-support vector domain description is proposed. Firstly, by principal component analysisto order the normal and known fault samples as a new eigenvector according to their contributions, and the weighted Gaussian kernel function was constructed with the contribution of features to improve the ability to identify inter type feature differences. Then the PSO was used to optimize the kernel parameters so that the model had a higher promotion ability and smaller false positive rate. Secondly, the normal and known fault samples were trained, the hypersphere describing the different working states of the isolating switch was established as the predictive model. Finally, the sample space was divided by multi support vector data description (Multi-SVDD) and the distance from the sample point to the center of the hypersphere was calculated to determine the type of sample. The experimental results showed that this method could effectively deal with the problem of incomplete fault samples in the fault diagnosis of high voltage isolating switch, and it could judge the unknown fault while diagnosing the known fault.

Key words： Isolating switch incomplete class multi support vector data description kernel function fault diagnosis

收稿日期: 2018-01-08 出版日期: 2018-06-13

PACS:	TM564.1
	TM507

通讯作者: 关永刚男,1974年生,副教授,博士生导师,研究方向为高压电气设备及智能化技术。E-mail：guanyg@tsinghua.edu.cn

作者简介: 陈士刚男,1991年生,硕士研究生,研究方向为电力设备状态检测与故障诊断技术。E-mail：16121424@bjtu.edu.cn

引用本文:

陈士刚, 关永刚, 张小青, 杨元威, 张一茗. 不完备故障类别下基于Multi-SVDD的高压隔离开关故障诊断方法[J]. 电工技术学报, 2018, 33(11): 2439-2447. Chen Shigang, Guan Yonggang, ZhangXiaoqing, Yang Yuanwei, Zhang Yiming. Diagnosis Method of High VoltageIsolating Switch Fault Based on Multi-SVDD under Incomplete Fault Type. Transactions of China Electrotechnical Society, 2018, 33(11): 2439-2447.

链接本文:

https://dgjsxb.ces-transaction.com/CN/10.19595/j.cnki.1000-6753.tces.180026 https://dgjsxb.ces-transaction.com/CN/Y2018/V33/I11/2439

[1] Chapelle O, Vapnik V,Bengio Y.Model selection for small sample regression[J]. Journal of Machine Learing, 2002, 48(1-3): 9-23.
[2] 陈佳成, 唐凯, 徐建. GW□-252隔离开关的机械故障及对策[J]. 电气技术, 2007(11): 69-70, 73.
Chen Jiacheng, Tang Kai, Xu Jian.Mechanical failure of GW□-252 switch-disconnector and the method of solving[J]. Electrical Engineering, 2007(11): 69-70, 73.
[3] 张一茗, 李少华, 陈士刚, 等. 基于ReliefF特征量优化及BP神经网络识别的高压隔离开关故障类型与位置诊断方法[J]. 高压电器, 2018, 54(2): 307-314.
Zhang Yiming, Li Shaohua, Chen Shigang, et al.Fault type and position diagnosis method of high-voltage isolating switch based on ReliefF characteristic quantity optimization and BP neural network recognition[J]. High Voltage Apparatus, 2018, 54(2): 307-314.
[4] 王雷,张瑞青,盛伟,等. 基于支持向量机的回归预测和异常数据检测[J]. 中国电机工程学报,2009, 29(8):92-96.
Wang Lei, Zhang Ruiqing, Sheng Wei, et al.Regression forecast and abnormal data detection based on support vector regression[J]. Proceedings of the CSEE, 2009, 29(8):92-96.
[5] 唐炬,林俊亦,卓然,等. 基于支持向量数据描述的局部放电类型识别[J]. 高电压技术, 2013, 39(5): 1046-1053.
Tang Ju, Lin Junyi, Zhuo Ran, et al.Partial discharge type recognition based on support vector data description[J]. High Voltage Engineering, 2013, 39(5): 1046-1053.
[6] 顾笑伟, 王智伟, 王占伟, 等. 基于密度权重支持向量数据描述的冷水机组故障检测[J]. 化工学报, 2017, 68(3): 1099-1108.
GuXiaowei, Wang Zhiwei, Wang Zhanwei, et al. Chiller fault detection by density weighted support vector data description[J]. CIESC Journal, 2017, 68(3): 1099-1108.
[7] Shan Jianfeng, Wang Liangwei. Circuit fault diagnosis based on OEMD and SVDD classifier of KFPCM optimal algorithm[J]. Applied Mechanics and Materials, 2015, 738-739: 366-372.
[8] 许敏, 王士同, 顾鑫, 等. 基于最小包含球的大数据集域自适应快速算法[J]. 模式识别与人工智能, 2013, 26(2): 159-168.
Xu Min, Wang Shitong, Gu Xin, et al.A fast learning algorithm based on minimum enclosing ball for large domain adaptation[J]. PR & AI, 2013, 26(2): 159-168.
[9] 黄南天,方立华,王玉强,等. 基于局域均值分解和支持向量数据描述的高压断路器机械状态监测[J]. 电工电能新技术, 2017, 36(1): 73-80.
Huang Nantian, Fang Lihua, Wang Yuqiang, et al.Machinery condition monitoring of high voltage circuit breakers based on local mean decomposition and support vector data description[J]. Advanced Technology of Electrical Engineering and Energy, 2017, 36(1): 73-80.
[10] Huang Nantian, Fang Lihua, Wang Yuqiang, et al.Machinery condition monitoring of high voltage circuit breakers basrd on local mean decomposition and support vector data description[J]. Advanced Technology of Electrical Engineering and Energy, 2017, 36(1): 73-80.
[11] 李赢, 舒乃秋. 基于模糊聚类和完全二叉树支持向量机的变压器故障诊断[J]. 电工技术学报, 2016, 31(4): 64-70.
Li Ying, ShuNaiqiu. Transformer fault diagnosis based on fuzzy clustering and complete binary tree support vector machine[J]. Transactions of China Electrotechical Society, 2016, 31(4): 64-70.
[12] 赵犁丰, 王栋. 基于加权多宽度高斯核函数的聚类算法[J]. 现代电子技术, 2011, 34(10): 78-81.
Zhao Lifeng, Wang Dong.Support vector clustering algorithm based on WGKMW[J]. Modern Electronics Technique, 2011, 34(10): 78-81.
[13] Qiao Feng, ZhaoHaoming, Zhang Feng, et al. A novel method for fault diagnosis based on PCA enhanced by wavelet denoising[J]. Advanced Materials Research, 2013, 756-759: 3450-3454.
[14] Hannan S A, Bhagile V D, Manza R R, et al.Development of an expert system for diagnosis and appropriate medical prescription of heart disease using SVM and RBF[J]. International Journal of Computer Science and Information Security, 2010, 8(5):795-802.
[15] Luo Hui, Wang Youren, Cui Jiang.A SVDD approach of fuzzy classification for analog circuit fault diagnosis with FWT as preprocessor[J]. Expert Systems with Applications, 2011, 38(8): 10554-10561.
[16] 黄飞腾, 南余荣, 翁国庆, 等. 计及监测可信度的扰动源定位粒子群算法[J]. 电工技术学报, 2016, 31(16): 215-222.
Huang Feiteng, Nan Yurong, Weng Guoqing, et al.Particle swarm optimization of disturbance source location with monitoring reliability[J]. Transactions of China Electrotechical Society, 2016, 31(16): 215-222.
[17] 何怡刚,祝文姬,周炎涛,等.基于粒子群算法的模拟电路故障诊断方法[J].电工技术学报,2010, 25(6):163-171.
He Yigang, Zhu Wenji, Zhou Yantao,et al.An analog circuit diagnosis method based on particle swarm optimization algorithm[J]. Transactions of China Electrotechical Society, 2010, 25(6):163-171.
[18] 尉军军,全力,彭桂雪,等. 基于最小二乘支持向量机的励磁特性曲线拟合[J]. 电力系统保护与控制,2010,38(11): 15-17, 24.
Wei Junjun, Quan Li, Peng Guixue, et al.Curve fitting of excitation characteristics based on the least squares support vector machine[J]. Power System Protection and Control, 2010,38(11):15-17, 24.
[19] 付文龙,周建中,李超顺,等.基于模糊K近邻支持向量数据描述的水电机组振动故障诊断研究[J].中国电机工程学报, 2014, 34(32): 5788-5795.
Fu Wenlong, Zhou Jianzhong, Li Chaoshun, et al.Vibrant fault diagnosis for hydro-electric generating unit based on support vector data description improved with fuzzy K nearest neighbor[J]. Proceedings of the CSEE, 2014, 34(32): 5788-5795.
[20] 韩富春,高文军,廉建鑫,等. 基于免疫优化多分类SVM的变压器故障诊断新方法[J].电力系统保护与控制, 2012, 40(2): 106-110.
Han Fuchun, GaoWenjun, Lian Jianxin, et al. A novel approach based on multi-class support vector machine of immune optimization for transformer fault diagnosis[J]. Power System Protection and Control, 2012, 40(2): 106-110.
[21] 程序, 关永刚, 张文鹏, 等. 基于因子分析和支持向量机算法的高压断路器机械故障诊断方法[J].电工技术学报, 2014, 29(7):209-215.
Cheng Xu, Guan Yonggang, Zhang Wenpeng, et al.Diagnosis method on the mechanical failure of high voltage circuit breakers based on factor analysis and SVM[J]. Transactions of China Electrotechical Society, 2014, 29(7):209-215.