We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
基于GAF和DenseNet的滚动轴承故障诊断方法.
- Authors
姜家国; 郭曼利; 杨思国
- Abstract
Model-based and signal-based rolling bearing fault diagnosis methods have problems such as difficult modeling and cumbersome signal analysis; data-driven rolling bearing fault diagnosis methods mostly use convolutional neural networks, but gradients will appear as the number of network layers increases during network training. The problem of disappearance, and the direct use of the rolling bearing vibration signal as the network input will cause incomplete feature extraction. In response to the above problems, a fault diagnosis method for rolling bearings based on Gram angle field (GAF) and densely connected convolutional network (DenseNet) is proposed. The one-dimensional time series of the rolling bearing vibration signal is converted into a two-dimensional image through GAF, which retains the relevant information between the time series data; the two-dimensional image is used as the input of DenseNet, and the two-dimensional image is extracted through DenseNet to improve the feature information. Utilization rate, and then achieve fault classification. Experiments were carried out on the bearing data set of Case Western Reserve University. The results show that the method can effectively identify the type of rolling bearing faults, and the fault diagnosis accuracy rate is 99.75%. In order to further prove the superiority of the method, the gray map + DenseNet, GAF + residual network (ResNet), gray map + ResNet fault diagnosis methods are selected for comparison. The results show that the GAF + DenseNet method has the highest accuracy rate, and the gray map + The ResNet method has the lowest accuracy; compared with the grayscale image, the two-dimensional image converted by GAF retains the relevant information between the original time series data; compared with ResNet, DenseNet can more fully due to its denser connection method. Extract fault characteristics.
- Publication
Journal of Mine Automation, 2021, Vol 47, Issue 8, p84
- ISSN
1671-251X
- Publication type
Article
- DOI
10.13272/j.issn.1671-251x.2021040095