为充分利用时域、频域以及时频域中的有效特征,提高滚动轴承故障诊断准确率,提出一种混合域特征集构建方法,利用原始信号分别生成时域和频域特征集,通过经验模式分解提取固有模态函数的排列熵和Hilbert谱的奇异值作为时频域特征集,使得混合域特征集比单域特征更能全面准确反映轴承运行状态。针对混合域特征集存在维数过高、特征之间冗余性严重的问题,采用加权最大相关最小冗余的特征选择方法,以支持向量机分类正确率为依据,选取7个有效特征向量。实验结果表明:基于WMRMR的混合域特征选择方法的分类准确率可达98%,能够有效的识别轴承故障信息。
Abstract
In order to improve the accuracy of rolling bearings fault diagnosis by making full use of effective features from time domain,frequency domain and time-frequency domain,a mixed domain feature built approach is proposed,which generate time domain and frequency domain features using the original signal, and extract permutation entropy of intrinsic mode function and singular values of the Hilbert spectrum as the time-frequency domain feature sets by empirical mode decomposition, making mixed domain feature set more fully and accurately reflect the bearing running than the single domain features. For the problem of mixed domain feature sets which have the shortcomings of too high dimension and serious redundancy, a feature selection method based on weighted minimal redundancy maximal relevance is proposed, which can select seven major feature vectors based on the classification accuracy of support vector machine. It can be shown from experiment results that: the classification accuracy of mixed domain feature selection can reach to 98% based on weighted minimal redundancy maximal relevance, and effectively identify the bearing fault information.
关键词
混合域 /
经验模式分解 /
Hilbert谱奇异值 /
排列熵 /
加权最大相关最小冗余
{{custom_keyword}} /
Key words
mixed domain /
empirical mode decomposition /
singular values of Hilbert spectrum /
permutation entropy /
weighted minimal redundancy maximal relevance
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1]Randall R B, Antoni J. Rolling element bearing diagnostics A tutorial[J]. Mechanical Systems and Signal Processing, 2011, 25(2): 485-520.
[2]罗荣,田福庆,李克玉等.卷积型小波变换实现及机械早期故障诊断应用[J]. 振动与冲击, 2013, 32(7): 64-69.
LUO Rong, TIAN Fu-qing, LI Ke-yu, et al. Realization of convolution wavelet transform and its application in mechanical incipient fault diagnosis[J]. Journal of Vibration and Shock, 2013, 32(7): 64-69.
[3]张超,陈建军,郭迅.基于EMD能量熵和支持向量机的齿轮故障诊断方法[J].振动与冲击, 2010, 29(10): 216-220.
ZHANG Chao, CHEN Jian-jun, GUO Xun. A Gear Fault Diagnosis Method based on EMD Energy Entropy and SVM [J]. Journal of Vibration and Shock, 2010, 29(10): 216-220.
[4]The Case Western Reserve University Bearing Data Center Website [DB/OL]. http://csegroups.case.edu/bearingdatacenter/pages/download-data-file
[5]Sassi S, Badri B, Thomas M. TALAF and THIKAT as innovative time domain indicators for tracking ball bearings[C]. Proceedings of the 14th Seminar on Machinery Vibration, Vancouver, IEEE, 2006: 24-27.
[6]杨国安.滚动轴承故障诊断实用技术[M].北京:中国石化出版社,2012:65-70.
YANG Guo-an. Practical technology of rolling bearings fault diagnosis[M].Beijing: China Petrochemical Press,2012:65-70.
[7]Huang N E, Shen Z, Long S R, et al. The Empirical Mode Decomposition and the Hilbert Spectrum for Nonlinear and Non-stationary Time Series Analysis [J]. Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences, 1998, 454(1971): 903-995.
[8]赵志宏,杨绍普,李韶华.基于Hilbert谱奇异值的轴承故障诊断[J].中国机械工程, 2013, 24(3): 346-350.
ZHAO Zhi-hong, YANG Shao-pu, LI Shao-hua. Bearings fault diagnosis based on singular values of Hilbert spectrum [J]. China Mechanical Engineering, 2013, 24(3): 346-350.
[9]饶国强,冯辅周,司爱威等. 排列熵算法参数的优化确定方法研究[J]. 振动与冲击, 2014 , 33(11): 188-193.
RAO Guo-qiang, FENG Fu-zhou, SI Ai-wei, et al. Method for optimal determination of parameters in permutation entropy algorithm [J]. Journal of Vibration and Shock, 2014, 33(1): 188-193.
[10]Bandt C, Pompe B. Permutation Entropy: a Natural Complexity Measure for Time Series [J]. Physical Review Letters, 2002, 88(17): 1-4.
[11]Ding C, Peng H. Minimum redundancy feature selection from microarray gene expression data [J]. Journal of bioinformatics and computational biology, 2005, 3(2): 185-205.
[12]李扬,顾雪平.基于改进最大相关最小冗余判据的暂态稳定评估特征选择[J].中国电机工程学报, 2013, 33(34): 179-186.
LI Yang, GU Xue-ping. Feature selection for transient stability assessment based on improved maximal relevance and minimal redundancy criterion [J]. Proceedings of the CSEE, 2013, 33(34): 179-186.
[13]彭涛,杨慧斌,李健宝,等.基于核主元分析的滚动轴承故障混合域特征提取方法[J]. 中南大学学报(自然科学版), 2011, 42(11): 3384-3391.
PENG Tao, YANG Hui-bin, LI Jian-bao, et al. Mixed-domain feature extraction approach to rolling bearings faults based on kernel principle component analysis[J]. Journal of Central South University (Science and Technology), 2011, 42(11): 3384-3391.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}