基于多特征融合多核学习支持向量机的液压泵故障识别方法

中国机械工程

基于多特征融合多核学习支持向量机的液压泵故障识别方法

刘志强1,2,3;姜万录1,2;谭文振3;朱勇1,2

1.燕山大学河北省重型机械流体动力传输与控制重点实验室,秦皇岛,066004
2.先进锻压成形技术与科学教育部重点实验室,秦皇岛,066004
3.唐钢高强汽车板有限公司,唐山,063002

出版日期:2016-12-25 发布日期:2016-12-28
基金资助:
国家自然科学基金资助项目(51475405)；国家重点基础研究发展计划(973计划)资助项目(2014CB046405)；河北省自然科学基金资助项目(E2013203161)

Fault Identification Method for Hydraulic Pumps Based on Multi-feature Fusion and Multiple Kernel Learning SVM

Liu Zhiqiang1,2，3;Jiang Wanlu1,2;Tan Wenzhen3;Zhu Yong1,2

1.Hebei Provincial Key Laboratory of Heavy Machinery Fluid Power Transmission and Control,Yanshan University,Qinhuangdao，Hebei,066004
2.Key Laboratory of Advanced Forging & Stamping Technology and Science,Ministry of Education of China,Qinhuangdao，Hebei,066004
3.Tangsteel High Strength Automotive Strip Co.,Ltd.,Tangshan，Hebei,063002

Online:2016-12-25 Published:2016-12-28
Supported by:

摘要/Abstract

摘要： 提出基于多特征融合多核学习支持向量机的液压泵故障识别方法。该方法首先对原始信号进行集总经验模态分解，然后分别用AR模型和奇异值分解两种特征提取方法提取故障特征，最后将不同类型的特征分别用相应的核函数进行映射，用多核学习支持向量机来识别液压泵的工作状态和故障类型。实验结果表明该方法显著地提高了故障诊断的准确性。

关键词: 多核学习, 多特征融合, 支持向量机, 故障识别, 液压泵

Abstract: A hydraulic pump fault identification method was put forward based on multiple feature fusion and multiple kernel learning SVM. Firstly, the original signals were processed by the ensemble empirical mode decomposition. Then, the feature vectors of hydraulic pump faults were obtained by using the autoregressive model and the singular value decomposition. Through different types of features mapped by corresponding different kernel functions, the hydraulic pump working conditions and fault types might be finally identified by multiple kernel learning SVM. The experimental results show that the approach improves the accuracy of fault diagnosis significantly.

Key words: multiple kernel learning, multi-feature fusion, support vector machine(SVM), fault identification, hydraulic pump

中图分类号:

TP277

刘志强, 姜万录, 谭文振, 朱勇, . 基于多特征融合多核学习支持向量机的液压泵故障识别方法[J]. 中国机械工程.

Liu Zhiqiang, Jiang Wanlu, Tan Wenzhen, Zhu Yong, . Fault Identification Method for Hydraulic Pumps Based on Multi-feature Fusion and Multiple Kernel Learning SVM[J]. China Mechanical Engineering.

参考文献

[1]汪洪桥，孙富春，蔡艳宁，等.多核学习方法[J].自动化学报，2010，36(8)：1037-1050.

Wang Hongqiao, Sun Fuchun, Cai Yanning, et al. On Multiple Kernel Learning Methods[J]. Acta Automatica Sinica, 2010, 36(8): 1037-1050.

[2]郭创新，朱承治，张琳，等.应用多分类多核学习支持向量机的变压器故障诊断方法[J].中国电机工程学报，2010，30(13)：128-134.

Guo Chuangxin, Zhu Chengzhi, Zhang Lin, et al. A Fault Diagnosis Method for Power Transformer Based on Multiclass Multiple-kernel Learning Support Vector Machine[J]. Proceedings of the CSEE, 2010, 30(13): 128-134.

[3]Lanckriet G, Cristianini N, Bartlett P, et al. Learning the Kernel Matrix with Semi-definite Programming[J]. Journal of Machine Learning Research, 2004, 5: 27-72.

[4]Lewis D P, Jebra T, Noble W S. Nonstationary Kernel Combination[C]//Proceeding of the 23rd International Conference on Machine Learning. Pittsburgh: The International Machine Learning Society, 2006: 553-560.

[5]Ong C S, Smola A J, Williamson R C. Learning the Kernel with Hyperkernels[J]. The Journal of Machine Learning Research, 2005, 6: 1043-1071.

[6]Sonnenburg S, Ratsch G, Schafer C. A General and Efficient Multiple Kernel Learning Algorithm[C]//Proceedings of the Advances in Neural Information Processing Systems. Vancouver: The MIT Press, 2005: 1273-1280.

[7]Rakotomamonjy A, Bach F, Canu S, et al. More Efficiency in Multiple Kernel Learning[C]//Proceeding of the 24th International Conference on Machine Learning. Corvallis: The International Machine Learning Society, 2007: 775-782.

[8]Bach F R, Lanckriet G, Jordan M. Multiple Kernel Learning, Conic Duality, and the SMO Algorithm[C]//Proceedings of the 21st International Conference on Machine Learning. Alberta: The International Machine Learning Society, 2004: 41-48.

[9]Sonnenburg S, Ratsch G, Schafer C, et al. Large Scale Multiple Kernel Learning[J]. The Journal of Machine Learning Research, 2006, 7: 1531-1565.

[10]何正嘉，訾艳阳，张西宁.现代信号处理及工程应用[M].西安：西安交通大学出版社，2007.

[11]Huang N E, Shen Z, Long S R, et a1. The Empirical Mode Decomposition and the Hilbert Spectrum for Nonlinear and Nonstationary Time Series Analysis[J]. Proceedings of the Royal Society, 1998, 454(1971): 903-995.

[12]Wu Z H, Huang N E. Ensemble Empirical Mode Decomposition: a Noise Assisted Data Analysis Method[J]. Advances in Adaptive Data Analysis, 2009, 1(1): 1-41.

[13]于德介，杨宇，程军圣.一种基于SVM和EMD的齿轮故障诊断方法[J].机械工程学报，2005，41(1)：140-144.

Yu Dejie, Yang Yu, Cheng Junsheng. Fault Diagnosis Approach for Gears Based on EMD and SVM[J]. Chinese Journal of Mechanical Engineering, 2005, 41(1): 140-144.

[14]程军圣，于德介，杨宇.基于内禀模态奇异值分解和支持向量机的故障诊断方法[J].自动化学报，2006，32(3)：475-480.

Cheng Junsheng, Yu Dejie, Yang Yu. Fault Diagnosis Approach Based on Intrinsic Mode Singular Value Decomposition and Support Vector Machines[J]. Acta Automatica Sinica, 2006, 32(3): 475-480.

[15]Rakotomamonjy A, Bach F R, Canu S, et al. Simple MKL[J]. The Journal of Machine Learning Research, 2008, 9: 2491-2521.

[16]Vedaldi A, Gulshan V, Varma M, et al. Multiple Kernels for Object Detection[C]//Proceedings of IEEE 12th International Conference on Computer Vision. Kyoto: Institute of Electrical and Electronics Engineers， 2009： 606-613.

[17]刘志强.基于多核学习支持向量机的旋转机械故障识别方法研究[D].秦皇岛：燕山大学，2014.

[18]Lee Y, Lin Y, Wahba G. Multicategory Support Vector Machines: Theory and Application to the Classification of Microarray Data and Satellite Radiance Data[J]. Journal of the American Statistical Association, 2004, 99(465): 67-81.