Bi-dimensional empirical mode decomposition (BEMD) and the stopping criterion based on the number and change of extreme points

Ma, Xingmin; Zhou, Xianwei; F, An

doi:10.1007/s12652-018-0955-4

Cited by 4 publications

(1 citation statement)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, due to the harsh operating environment and the high incidence of rolling bearing failures, rolling bearing failures often cause huge casualties and economic losses. Using machine learning and deep learning methods to carry out abnormal detection and research on rolling bearings to achieve intelligent fault diagnosis is of important practical significance for timely detection of faults, early warning and predictive maintenance, safe operation of units, improvement of unit operation efficiency, and avoidance of accidents [2][3][4]. V. Purashotham [5] presents a new method for detecting localized bearing defects based on wavelet transform.…”

Section: Introductionmentioning

confidence: 99%

Rolling Bearing Fault Diagnosis and Prediction Based on VMD-CWT and MobileNet

Zhu¹,

Deng²,

Xue³

et al. 2021

Computer Science &Amp; Information Technology (CS &Amp; IT)

View full text Add to dashboard Cite

When deep learning is used for rolling bearing fault diagnosis, there are problems of high model complexity, time-consuming, and large memory. In order to solve this problem. This paper presents an intelligent diagnosis method of rolling bearings based on VMD-CWT feature extraction and MobileNet, VMD is used to extract the signal features, and then wavelet transform is used to extract the timefrequency features. After the image is enhanced, the MobileNet network is trained. In order to accelerate the convergence speed, this paper adds transfer learning in the network training process, and migrates the weights of the first several layers pretrained to the corresponding network. Experimental results based on bearing fault data sets show that after adopting VMD-CWT, the accuracy of mobilenet increased from 68.7% to 94%, and its network parameters were reduced by an order of magnitude compared with CNN.

show abstract

Section: Introductionmentioning

confidence: 99%