Welcome to Francis Academic Press

Academic Journal of Engineering and Technology Science, 2020, 3(7); doi: 10.25236/AJETS.2020.030707.

Rolling Bearing Fault Pattern Recognition Method Based on Hht-Svd Parameter Optimization and Hssvm


Ying Tan, Xiaohui Wang*, Guangzhou Shui

Corresponding Author:
Xiaohui Wang

School of Mechanical and Electrical Engineering, Lingnan Normal University, Zhanjiang, 524048, China
* Corresponding Author


A bearing fault diagnosis method based on Hilbert-Huang transform and matrix singular value is proposed. Make Hilbert-Huang transform (HHT)  on bearing  signals to obtain a spectrum of their 8 components. The sensitive mode functions which are used to reconstruct the signal are selected according to the characteristic frequency of the bearing. Make empirical mode decomposition which was improved by extension method again on the reconstructed signal, get 5 new intrinsic modal components (IMF). Then calculate their singular value de-composition (SVD) to get 5 Singular Values. At last inputting these 5 singular values as characteristic parameters into the Hyper-sphere Support Vector Machine (HSSVM) to learn classification, identify the type of failure of the rolling bearing, and use visualization technology to verify the effect. The experimental results show that the accuracy of the HSSVM classifier established by the method proposed in this paper is about 10% higher than that of the HSSVM classifier without parameter optimization, and the average response speed is increased by 3 seconds. It turns out that this pattern recognition method is advantageous both in terms of response speed and accuracy.


Hilbert-huang transform (hht), Singular value decomposition (svd), Support vector machine (svm), Pattern recognition

Cite This Paper

Ying Tan, Xiaohui Wang, Guangzhou Shui. Rolling Bearing Fault Pattern Recognition Method Based on Hht-Svd Parameter Optimization and Hssvm. Academic Journal of Engineering and Technology Science (2020) Vol. 3 Issue 7: 67-82. https://doi.org/10.25236/AJETS.2020.030707.


[1] Junliang Fan, Wenjun Yue, Lifeng Wu, et.al (2018). Evaluation of SVM, ELM and four tree-based ensemble models for predicting daily reference evapotranspiration using limited meteorological data in different climates of China. Agricultural and Forest Meteorology, vol.263.
[2] Ye Xiaochuan(2018). On Line Monitoring Technology of Mine Bearing Fault Based on HHT and SVM [J].Coal Technology, vol.37, no.07, p. 329-332.
[3] Yang Jieming, Tian Ying (2009). Roller Bearing Fault Diagnosis by Using Empirical Mode Decomposition and Sphere-Structured Support Vector Machine[J].Journal of Vibration,Test and Diagnosis, vol.29, no.02, p.155-158+239.
[4] Tian Ying(2008). Research on Identification of Fault Degradation State of Rolling Bearing Inner Ring [D]. Taiyuan University of Technology.
[5] Zhou Xiaolong, Jiang Zhenhai, Ma Fenglei(2016).Fault Diagnosis of Rolling Bearing Based on Improved HHT Energy Entropy and SVM[J].Journal of Mechanical Transmission, vol. 40, no.12, p.164-168.
[6] Juhamatti Saari,Daniel Strömbergsson,Jan Lundberg (2019),Allan Thomson. Detection and identification of windmill bearing faults using a one-class support vector machine (SVM)[J]. Measurement,p.137.
[7] Karthik Thirumala,Sushmita Pal,Trapti Jain,Amod C. Umarikar(2019). A classification method for multiple power quality disturbances using EWT based adaptive filtering and multiclass SVM[J]. Neurocomputing.
[8] Faquan Yang,Ling Yang,Dong Wang (2018).Method of Modulation Recognition Based on Combination Algorithm of K-Means Clustering and Grading Training SVM[J]. China Communications, vol. 15, no.12, p.55-63. China Communications
[9] Fuzhi Zhang,Quanqiang Zhou(2014). HHT–SVM: An online method for detecting profile injection attacks in collaborative recommender systems[J]. Knowledge-Based Systems, vol. 65.
[10] Şule Yücelbaş,Cüneyt Yücelbaş,Gülay Tezel,(2018). Automatic sleep staging based on SVD, VMD, HHT and morphological features of single-lead ECG signal[J]. Expert Systems With Applications,vol. 102.
[11] Bi Lin,Xie Wei,Zhao Junjie (2019). Automatic recognition and classification of multi-channel microseismic waveform based on DCNN and SVM[J]. Computers and Geosciences, vol.123.
[12] Zeng Rong,Zeng Ruili,Ding Lei (2018). Valve Fault Diagnosis of Diesel Engine Based on EMD-SVD and SVM [J].Journal of Military Transportation University, vol.20,no.10,p.47-52.