[1]周志成,杨志超,杨成顺,等.基于改进RADAG-SVM的电力变压器故障诊断[J].东南大学学报(自然科学版),2016,46(5):964-971.[doi:10.3969/j.issn.1001-0505.2016.05.011]
 Zhou Zhicheng,Yang Zhichao,Yang Chengshun,et al.Fault diagnosis of power transformer based on modified RADAG-SVM[J].Journal of Southeast University (Natural Science Edition),2016,46(5):964-971.[doi:10.3969/j.issn.1001-0505.2016.05.011]
点击复制

基于改进RADAG-SVM的电力变压器故障诊断()
分享到:

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
46
期数:
2016年第5期
页码:
964-971
栏目:
电气工程
出版日期:
2016-09-20

文章信息/Info

Title:
Fault diagnosis of power transformer based on modified RADAG-SVM
作者:
周志成1杨志超2杨成顺2陶风波1李建生1
1江苏省电力公司电力科学研究院, 南京 211103; 2南京工程学院电力工程学院, 南京 211167
Author(s):
Zhou Zhicheng1 Yang Zhichao2 Yang Chengshun2 Tao Fengbo1 Li Jiansheng1
1Jiangsu Electric Power Company Research Institute, Nanjing 211103, China
2School of Electric Power Engineering, Nanjing Institute of Technology, Nanjing 211167, China
关键词:
电力变压器 故障诊断 重排序自适应有向无环图 支持向量机
Keywords:
power transformer fault diagnosis reordering adaptive directed acyclic graph(RADAG) support vector machine(SVM)
分类号:
TM71
DOI:
10.3969/j.issn.1001-0505.2016.05.011
摘要:
为提高变压器故障诊断的正确率,提出一种基于改进重排序自适应有向无环图(reordering adaptive directed acyclic graph, RADAG)支持向量机(support vector machines, SVM)的电力变压器故障诊断方法.该方法首先利用基于K折交叉验证和人工蜂群算法优化SVM的核函数和惩罚因子参数,使二分类SVM获得最佳的分类性能;其次,为进一步提高多分类SVM的性能,提出利用交叉确认机制估计每个二分类SVM的泛化能力的方法,并将其用于改进RADAG-SVM的分类精度.最后,给出基于改进RADAG-SVM的变压器故障诊断流程并进行实例分析.结果表明,所提方法、原始RADAG-SVM和基于结点优化的DDAG-SVM方法对变压器故障诊断的平均正确率分别为94.16%,87.85%和90.77%.因而,与其他2种诊断方法相比,所提方法具有较好的故障诊断效果.
Abstract:
The method for fault diagnosis of power transformers based on the reordering adaptive directed acyclic graph support vector machines(RADAG-SVM)was proposed to enhance the accurate rate of the fault diagnosis of power transformers. Firstly, the scheme for parameter optimization of kernel and penalty factor based on K-fold cross-validation(K-CV)and artificial bee colony(ABC)was used to achieve the best classification performance of the binary SVM. Secondly, a method for estimating the generalization performance of each two classification SVM by using cross-validation mechanism was proposed to further improve the performance of multiclass SVM. And, it was used to improve the classification accuracy of RADAG-SVM. Finally, the process for fault diagnosis of power transformers based on the modified RADAG-SVM was addressed and experiments were carried out. The results show that the average correct rates for transformer fault diagnosis based on the proposed method, the original RADAG-SVM and the nodes refined decision directed acyclic graph(DDAG)SVM are 94.16%, 87.85% and 90.77%. Therefor, compared with the other two diagnosis methods, the proposed method has a better effect on the transformer fault diagnosis.

参考文献/References:

[1] 师瑞峰,史永锋,牟军,等. 油中溶解气体电力变压器故障诊断专家系统[J]. 电力系统及其自动化学报,2014,26(12):49-54.
  Shi Ruifeng, Shi Yongfeng, Mou Jun, et al. Power transformer fault diagnosis expert system with dissolved gas analysis in oil[J]. Proceedings of the CSU-EPSA, 2014, 26(12): 49-54. DOI:10.3969/j.issn.1003-8930.2014.12.008.(in Chinese)
[2] Souahlia S, Bacha K, Chaari A. MLP neural network-based decision for power transformers fault diagnosis using an improved combination of Rogers and Doernenburg ratios DGA[J]. International Journal of Electrical Power & Energy Systems, 2012, 43(1): 1346-1353. DOI:10.1016/j.ijepes.2012.05.067.
[3] Duraisamy V, Devarajan N, Somasundareswari D, et al. Neuro fuzzy schemes for fault detection in power transformer[J]. Applied Soft Computing, 2007, 7(2): 534-539. DOI:10.1016/j.asoc.2006.10.001.
[4] 周爱华,张彼德,方春恩,等.基于模糊免疫识别方法的电力变压器故障诊断[J].电网技术,2009,33(3):99-102.
  Zhou Aihua, Zhang Bide, Fang Chunen, et al. Power transformer fault diagnosis based on fuzzy immune recognition approach[J]. Power System Technology, 2009, 33(3): 99-102.(in Chinese)
[5] 厉劼翀,周宁,吕彬.基于神经网络、模糊理论的变压器油中溶解气体诊断专家系统[J].电网技术,2006,30(增刊):125-128.
  Li Jiechong, Zhou Ning, Lü Bin. Transformer DGA diagnosis expert system based on neural network and fuzzy theory[J]. Power System Technology, 2006, 30(supp): 125-128.(in Chinese)
[6] 杨廷方,周力行,李景禄,等.基于最优权值的组合模型诊断变压器故障[J].电网技术,2013,37(1):190-194.
  Yang Tingfang, Zhou Lixing, Li Jinglu.Power transformer fault diagnosis based on optimal weights combined model[J].Power System Technology, 2013, 37(1):190-194.(in Chinese)
[7] 李俭,孙才新,陈伟根,等.灰色聚类与模糊聚类集成诊断变压器内部故障的方法研究[J].中国电机工程学报, 2003, 23(2):112-115.
  Li Jian, Sun Caixin, Chen Weigen, et al.A method of synthesis based on the grey cluster and fuzzy cluster about internal fault diagnosis of transformer[J]. Proceedings of the CSEE, 2003, 23(2): 112-115.(in Chinese)
[8] 宋志杰,王健.AI优化模糊核聚类算法的变压器DGA分析[J].电力系统及其自动化学报,2014,26(4):39-43.DOI:10.3969/j.issn.1003-8930.2014.04.008.
  Song Zhijie, Wang Jian.Transformer DGA analysis based on fuzzy kernel clustering algorithm optimized by artificial immune[J]. Proceedings of the CSU-EPSA, 2014, 26(4): 39-43.DOI:10.3969/j.issn.1003-8930.2014.04.008.(in Chinese)
[9] 易辉,宋晓峰,姜斌,等.基于结点优化的决策导向无环图支持向量机及其在故障诊断中的应用[J].自动化学报,2010,36(3):427-432.
  Yi Hui, Song Xiaofeng, Jiang Bin, et al.Support vector machine based on nodes refined decision directed acyclic graph and its application to fault diagnosis[J]. Acta Automatica Sinica, 2010, 36(3): 427-432.(in Chinese)
[10] Fei Shengwei, Zhang Xiaobin. Fault diagnosis of power transformer based on support vector machine with genetic algorithm[J]. Expert Systems with Applications, 2009, 36: 11352-11357. DOI:10.1016/j.eswa.2009.03.022.
[11] Platt J C, Cristianini N, Shawe-Taylor J.Large margin DAG’s for multiclass classification[C]//Advances in Neural Information Processing Systems. Denver, Colorado, USA, 2000: 547-553.
[12] 赵春晖,陈万海,郭春燕.多类支持向量机方法的研究现状与分析[J].智能系统学报,2007,2(2):11-17. DOI:10.3969/j.issn.1673-4785.2007.02.002.
  Zhao Chunhui, Chen Wanhai, Guo Chunyan.Research and analysis of methods for multiclass support vector machines[J]. CAAI Transactions on Intelligent Systems, 2007, 2(2): 11-17. DOI:10.3969/j.issn.1673-4785.2007.02.002.(in Chinese)
[13] Phetkaew T, Kijsirikul B, Rivepiboon W.Reordering adaptive directed acyclic graphs: An improved algorithm for multiclass support vector machines[C]//Proceedings of the 2003 International Joint Conference on Neural Networks. Portland, OR, USA, 2003:1605-1610.
[14] Bartlett P, Shawe-Taylor J.Generalization performance of support vector machines and other pattern classifiers[C]//Advances in Kernel Methods-Support Vector Learning.Cambridge, USA, 1999: 43-54.
[15] Takahashi F, Abe S. Optimizing directed acyclic graph support vector machines[C]//International Workshop on Artificial Neural Networks in Pattern Recognition. Florence, Italy, 2003:166-170.
[16] Burges C.A tutorial on support vector machines for pattern recognition[J]. Data Min Knowl Discov, 1998, 2: 121-167.
[17] Markatou M, Tian H, Biswas S, et al. Analysis of variance of cross-validation estimators of the generalization error[J]. Journal of Machine Learning Research, 2005, 6(1): 1127-1168.
[18] Cook W, Rohe A. Computing minimum-weight perfect matchings[J]. Informs Journal on Computing, 1999, 11(2): 138-148. DOI:10.1287/ijoc.11.2.138.
[19] Karaboga D, Akay B. A comparative study of artificial bee colony algorithm[J]. Applied Mathematics and Computation, 2009, 214(1): 108-132. DOI:10.1016/j.amc.2009.03.090.
[20] Leisch F, Jain L C, Hornik K.Cross-validation with active pattern selection for neural network classifiers[J]. IEEE Transactions on Neural Network, 1998, 9(1): 35-41. DOI:10.1109/72.655027.

相似文献/References:

[1]杨建文,贾民平,许飞云.经验模式分解在循环平稳故障信号分析中的应用[J].东南大学学报(自然科学版),2006,36(1):77.[doi:10.3969/j.issn.1001-0505.2006.01.016]
 Yang Jianwen,Jia Minping,Xu Feiyun.Application of EMD to cyclostationarity fault signal[J].Journal of Southeast University (Natural Science Edition),2006,36(5):77.[doi:10.3969/j.issn.1001-0505.2006.01.016]
[2]许飞云,胡建中,贾民平,等.嵌入式旋转机械状态监控与故障诊断系统研究[J].东南大学学报(自然科学版),2004,34(6):761.[doi:10.3969/j.issn.1001-0505.2004.06.010]
 Xu Feiyun,Hu Jianzhong,Jia Minping,et al.Research on the embedded condition monitoring and fault diagnosis system for the rotation machine[J].Journal of Southeast University (Natural Science Edition),2004,34(5):761.[doi:10.3969/j.issn.1001-0505.2004.06.010]
[3]聂长海,徐宝文,史亮.一种基于组合测试的软件故障诊断方法[J].东南大学学报(自然科学版),2003,33(6):681.[doi:10.3969/j.issn.1001-0505.2003.06.001]
 Nie Changhai,Xu Baowen,Shi Liang.Software fault diagnosis method based on combinatorial testing[J].Journal of Southeast University (Natural Science Edition),2003,33(5):681.[doi:10.3969/j.issn.1001-0505.2003.06.001]
[4]杨国安,许飞云,贾民平,等.小波消噪用于钻井泵阀泄露故障诊断[J].东南大学学报(自然科学版),2001,31(2):50.[doi:10.3969/j.issn.1001-0505.2001.02.013]
 Yang Guoan,Xu Feiyun,Jia Minping,et al.Application of Wavelet De-Noise to Leaking Fault Diagnosis for Drilling Pump Valve[J].Journal of Southeast University (Natural Science Edition),2001,31(5):50.[doi:10.3969/j.issn.1001-0505.2001.02.013]
[5]朱大奇,于盛林.电子电路故障诊断的神经网络数据融合算法[J].东南大学学报(自然科学版),2001,31(6):87.[doi:10.3969/j.issn.1001-0505.2001.06.021]
 Zhu Daqi,Yu Shenglin.Neural Network Data Fusion Algorithm of Circuit Fault Diagnosis[J].Journal of Southeast University (Natural Science Edition),2001,31(5):87.[doi:10.3969/j.issn.1001-0505.2001.06.021]
[6]史治宇,张令弥,吕令毅.基于模态应变能诊断结构破损的修正方法[J].东南大学学报(自然科学版),2000,30(3):84.[doi:10.3969/j.issn.1001-0505.2000.03.018]
 Shi Zhiyu,Zhang Lingmi,Lu Lingyi.Improved Structural Damage Detection Method Based on Modal Strain Energy[J].Journal of Southeast University (Natural Science Edition),2000,30(5):84.[doi:10.3969/j.issn.1001-0505.2000.03.018]
[7]赵林度,盛昭瀚.形象思维故障诊断方法[J].东南大学学报(自然科学版),2000,30(6):66.[doi:10.3969/j.issn.1001-0505.2000.06.015]
 Zhao Lindu,Sheng Zhaohan.Methods of Thinking with Imagery for Fault Diagnosis[J].Journal of Southeast University (Natural Science Edition),2000,30(5):66.[doi:10.3969/j.issn.1001-0505.2000.06.015]
[8]杨国安,许飞云,吴贞焕,等.基于小波包和解调分析的多类故障综合诊断方法研究[J].东南大学学报(自然科学版),2004,34(1):42.[doi:10.3969/j.issn.1001-0505.2004.01.010]
 Yang Guoan,Xu Feiyun,Wu Zhenhuan,et al.Research on the multi-fault comprehensive diagnosis method based on wavelet packet and demodulation[J].Journal of Southeast University (Natural Science Edition),2004,34(5):42.[doi:10.3969/j.issn.1001-0505.2004.01.010]
[9]蒋全胜,贾民平.基于Web的连铸旋转塔远程状态监测与诊断系统[J].东南大学学报(自然科学版),2006,36(3):407.[doi:10.3969/j.issn.1001-0505.2006.03.015]
 Jiang Quansheng,Jia Minping.Remote monitoring and diagnosis system for turret of continuous casting machine based on Web[J].Journal of Southeast University (Natural Science Edition),2006,36(5):407.[doi:10.3969/j.issn.1001-0505.2006.03.015]
[10]赵林度,盛昭瀚.基于小波变换的分形数据压缩不变性[J].东南大学学报(自然科学版),2000,30(3):41.[doi:10.3969/j.issn.1001-0505.2000.03.009]
 Zhao Lindu,Sheng Zhaohan.Invariance Properties of Fractal Data Compression Based on Wavelet Transform[J].Journal of Southeast University (Natural Science Edition),2000,30(5):41.[doi:10.3969/j.issn.1001-0505.2000.03.009]

备注/Memo

备注/Memo:
收稿日期: 2015-12-02.
作者简介: 周志成(1977—),男,博士,高级工程师, zhouzhicheng1977@126.com.
基金项目: 国家电网公司总部科技资助项目(5299001352U0)、江苏省产学研联合前瞻性资助项目(BY2015008-05,BY2016008-05).
引用本文: 周志成,杨志超,杨成顺,等.基于改进RADAG-SVM的电力变压器故障诊断[J].东南大学学报(自然科学版),2016,46(5):964-971. DOI:10.3969/j.issn.1001-0505.2016.05.011.
更新日期/Last Update: 2016-09-20