Abstract
Rolling mill multi-row bearings are subjected to axial loads, which cause damage of rolling elements and cages, so the axial vibration signal contains rich fault character information. The vertical shock caused by the failure is weakened because multiple rows of bearings are subjected to radial forces together. Considering the special characters of rolling mill bearing vibration signals, a fault diagnosis method combining Adaptive Multivariate Variational Mode Decomposition (AMVMD) and Multi-channel One-dimensional Convolution Neural Network (MC1DCNN) is proposed to improve the diagnosis accuracy. Additionally, Deep Convolutional Generative Adversarial Network (DCGAN) is embedded in models to solve the problem of fault data scarcity. DCGAN is used to generate AMVMD reconstruction data to supplement the unbalanced dataset, and the MC1DCNN model is trained by the dataset to diagnose the real data. The proposed method is compared with a variety of diagnostic models, and the experimental results show that the method can effectively improve the diagnosis accuracy of rolling mill multi-row bearing under unbalanced dataset conditions. It is an important guide to the current problem of insufficient data and low diagnosis accuracy faced in the fault diagnosis of multi-row bearings such as rolling mills.
Highlights
Rolling mill multi-row bearings are the core of the main drive system of the rolling mill, which support the rolling mill roll system and withstand huge radial forces
After each channel signal is reconstructed by Adaptive Multivariate Variational Mode Decomposition (AMVMD), it is input into M1DCNN for individual convolution calculation, and the multi-channel can more comprehensively explore the information of fault vibration signal characters than the single channel
Deep Convolutional Generative Adversarial Network (DCGAN), which further verifies the superiority of the fault diagnosis model proposed in this paper
Summary
Rolling mill multi-row bearings are the core of the main drive system of the rolling mill, which support the rolling mill roll system and withstand huge radial forces. One-dimensional convolution solves the problem of time series feature loss, and makes CNN lose the ability to handle high-dimensional data; the analysis of a single-channel signal cannot fully explore the fault character information of the large equipment. In [28], Liu applied Generating Adversarial Network (GAN) to deep feature enhancement of bearing data and demonstrated that GAN can overcome the problems of insufficient fault data and unbalanced dataset, and GAN can improve the model training effect to improve the diagnosis accuracy. In order to reduce the effect of noise on the feature extraction ability of MC1DCNN, AMVMD was combined with MC1DCNN and applied to multi-channel signal fault diagnosis of rolling mill multi-row bearings.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
