Research of Mechanical Vibration Signal Classification Based on LMD

Abstract:

The traditional signal processing methods are difficult to accurately extract fault information, because mechanical fault vibration signals have non-stationary, which will cause system instability. Local mean decomposition is adaptive signal processing method. However, in the local mean decomposition of the signal, the trend of the endpoint can not be predicted which cause contaminating the entire signal sequence, the original moving average of the signal used over-smoothing treatment, resulting in fault characteristics can not accurately extract. The article introduces waveform matching to solve the original features of signals at the endpoints, using linear interpolation to get local mean and envelope function, then obtain production function PF vector through making use of the local mean decomposition. The energy entropy of PF vector take as identification input vectors. These vectors are respectively inputted BP neural networks, support vector machines, least squares support vector machines to identify faults. Experimental result show that the accuracy of least squares support vector machine with higher classification accuracy has been improved.

• Periodical: Applied Mechanics and Materials (Volumes 764-765)
• Pages: 350-358
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.764-765.350
• Online since: May 2015
• Authors: Zeng Shou Dong*, Zhao Jing Ren, You Dong
• Keywords: End Effect, Least Squares Support Vector Machines, Local Mean Decomposition, State Recognition, Vibration Signals

References

[1] Chongfeng Cao, Shixi Yang, Jiangxin Yang: Vibration Mode Extraction Method Based on The Characterristics of White Noise [J]. Journal of Mechanical Engineering, 2010, 46(3): 65-70.

[2] Baojia Chen, Zhengjia He, Xuefeng Chen, Yanyang Zi: Locomotive Fault Diagnosis Based on Local Mean Decomposition Demoddulating Approach[J]. Journal of Xi'an Jiao Tong University, 2010, 44(5): 40-44.

[3] Zhinong Li, Weibing Liu, Yaoxian Xiao, Guanhua Wu: Fault Diagnosis Method of Rolling Bearing Based on Local Mean Decomposition Envelope Spectrum and Support Vector Machine [J]. Journal of Mechanical Design and Manufacturing, 2011, 11(5): 170-173.

[4] Shuqing Zhang, Guoxiu Sun, Liang Li, Xinxin Li, Xiong Jian: Study on mechanical Fault Diagnosis Method Based on LMD Approximate Entropy and Fuzzy C-means Clustering [J]. Chinese Journal of Scientific Instrument, 2013, 34(3): 714-720.

[5] Yanxue Wang, Zhengjia He, Yanyang Zi, Jing Yuan: Several Key Issues of Local Mean Decomposition Method Used in Mechanical Fault Diagnosis[J]. Journal of Vibration and Shock, 2012, 31(9): 10-12.

[6] Liang Xu: Nonlinear Feature Extraction and LSSVM Application in The Chemical Process Faults Diagnosis [J]. ournal of Computer Applications, 2010, 30(1): 236-239.

[7] Chunxia Tang, Chunhua Yang, Weihua Gui, Hongqiu Zhu: Based on KPCA-LSSVM silicon-manganese alloy smelting slag basicity prediction research process [J]. Chinese Journal of Scientific Instrument, 2010, 31(3): 689-693.

[8] Xiaobin Wang: Application of Neural Network Path Optimization in Fault Diagnosis [D]. Chengdu: University of Electronic Science and Technology，2012, 10-12.

[9] Tingxin Wen, Junjun Wang: Fault Diagnosis of Rolling Element Bearing Based on Wavelet Package and LSSVM[J]. Journal of Computer Simulation, 2012, 29(6): 202－205.

[10] L Gopalakrishnan K, Kim S: Support vector machines approach to HMA stiffness prediction [J]. Joumal of Engineering Mechanics, 2011, 137(2): 138-146.

DOI: 10.1061/(asce)em.1943-7889.0000214

[11] Liu L M, Wang A N, Sha M: Multi-class classification methods of cost-conscious LS-SVM for fault diagnosis of blast furnace [J]. Journal of Iron and Steel Research，International, 2011, 18(10): 17-23, 33.

DOI: 10.1016/s1006-706x(12)60016-8

[12] Smith JS: The local mean decomposition and its application to EEG perception data [J]. Journal of the Royal Society Interface, 2005, 2(5): 443-454.

DOI: 10.1098/rsif.2005.0058

[13] Daqian Ren: Rotating machinery fault feature extraction method and system research based on local mean decomposition [D]. Zhejiang: University of Zhejiang, 2008, 45-47.