A study on valve clearance anomaly feature extraction of diesel engines based on VMD and SVD

被引：0
|
作者：
Jiang Z. [1 ]
Wei D. [1 ]
Zhang J. [2 ]
Mao Z. [2 ]
机构：
[1] Beijing Key Laboratory of High-end Mechanical Equipment Health Monitoring and Self-Recovery, Beijing University of Chemical Technology, Beijing
[2] Key Lab of Engine Health Monitoring-Control and Networking of Ministry of Eduation, Beijing University of Chemical Technology, Beijing
来源：
Zhendong yu Chongji/Journal of Vibration and Shock | 2020年 / 39卷 / 16期
关键词：
Feature extraction; Singular value decomposition (SVD); Valve clearance failure; Variational mode decomposition (VMD);
D O I：
10.13465/j.cnki.jvs.2020.16.004
中图分类号：
学科分类号：
摘要：
A novel feature extraction method based on variational mode decomposition (VMD) and singular value decomposition (SVD) was proposed to effectively extract valve clearance fault characteristics from cylinder head surface vibration signals of diesel engines. The vibration signal of cylinder head was decomposed by the VMD algorithm, and the modal component was used to construct the characteristic matrix. Then the SVD theory was applied to transform the eigenmatrix into a singular value sequence representing the frequency characteristics. Since the influence of changes in working conditions such as rotating speed and load on the signal characteristic layer might be very similar to the signal characteristic changes caused by faults, the sequence of singular values was taken as characteristic parameters and input into the random forest classifier to build a classification model for the diagnosis of valve clearance faults under variable working conditions of diesel engine. Experimental results show that this method can effectively identify valve clearance faults and highlight fault sensitive features. Compared with the traditional SVD feature extraction method based on the Hankel matrix and the wavelet packet coefficient matrix, the feature parameters proposed by this method have higher recognition rate under variable working conditions of diesel engine. © 2020, Editorial Office of Journal of Vibration and Shock. All right reserved.
引用
收藏
页码：23 / 30
页数：7
相关论文
共 23 条
  • [1] JIANG Z N, MAO Z W, WANG Z J, Et al., Fault diagnosis of internal combustion engine valve clearance using the impact commencement detection method, Sensors, 17, 12, pp. 1-19, (2017)
  • [2] GRZEGORZ M, SZYMANSKI, TOMASZEWSKI F., Diagnostics of automatic compensators of valve clearance in combustion engine with the use of vibration signal, Mechanical Systems & Signal Processing, 68, 7, pp. 479-490, (2015)
  • [3] LIU Yu, ZHANG Junhong, BI Fengrong, Et al., Diesel engine valve fault diagnosis based on LMD marginal spectrum, Chinese Internal Combustion Engine Engineering, 35, 6, pp. 96-100, (2014)
  • [4] LU Yanchun, XU Yuxiu, Fault diagnosis of valve clearance of diesel engine based on Laplace wavelet correlation filtering, Vehicle Engine, 3, pp. 48-54, (2008)
  • [5] FLETT J, BONE G M., Fault detection and diagnosis of diesel engine valve trains, Mechanical Systems and Signal Processing, 72, 73, pp. 316-327, (2015)
  • [6] SI Jingping, REN Qingshuang, LIANG Hongbo, Et al., Valve clearance fault detection based on wavelet packet analysis, Journal of Vibration and Shock, 30, 12, pp. 64-68, (2011)
  • [7] SHEN Zhixi, HUANG Xiyue, MA Xiaoxiao, Fault diagnosis of diesel engine based on empirical mode decomposition and support vector machine, Journal of Vibration Measurement & Diagnosis, 30, 1, pp. 19-22, (2010)
  • [8] ZHANG Junhong, LIU Yu, BI Fengrong, Et al., Diesel engine valve fault diagnosis based on LMD and SVM, Transactions of CSICE, 30, 5, pp. 469-473, (2012)
  • [9] DRAGOMIRETSKIY K, ZOSSO D., Variational mode decomposition, IEEE Transactions on Signal Processing, 62, 3, pp. 531-544, (2014)
  • [10] WANG Y, MARKERT R, XIANG J, Et al., Research on variational mode decomposition and its application in detecting rub-impact fault of the rotor system, Mechanical Systems & Signal Processing, 60, 61, pp. 243-251, (2015)
123