Bearing Fault Diagnosis with a Feature Fusion Method Based on an Ensemble Convolutional Neural Network and Deep Neural Network

Abstract

Rolling bearings are the core components of rotating machinery. Their health directly affects the performance, stability and life of rotating machinery. To prevent possible damage, it is necessary to detect the condition of rolling bearings for fault diagnosis. With the rapid development of intelligent fault diagnosis technology, various deep learning methods have been applied in fault diagnosis in recent years. Convolution neural networks (CNN) have shown high performance in feature extraction. However, the pooling operation of CNN can lead to the loss of much valuable information and the relationship between the whole and the part may be ignored. In this study, we proposed CNNEPDNN, a novel bearing fault diagnosis model based on ensemble deep neural network (DNN) and CNN. We firstly trained CNNEPDNN model. Each of its local networks was trained with different training datasets. The CNN used vibration sensor signals as the input, whereas the DNN used nine time-domain statistical features from bearing vibration sensor signals as the input. Each local network of CNNEPDNN extracted different features from its own trained dataset, thus we fused features with different discrimination for fault recognition. CNNEPDNN was tested under 10 fault conditions based on the bearing data from Bearing Data Center of Case Western Reserve University (CWRU). To evaluate the proposed model, four aspects were analyzed: convergence speed of training loss function, test accuracy, F-Score and the feature clustering result by t-distributed stochastic neighbor embedding (t-SNE) visualization. The training loss function of the proposed model converged more quickly than the local models under different loads. The test accuracy of the proposed model is better than that of CNN, DNN and BPNN. The F-Score value of the model is higher than that of CNN model, and the feature clustering effect of the proposed model was better than that of CNN.

A Review on Deep Learning Applications in Prognostics and Health Management

“…The stride parameter should also be selected carefully, because an overly large stride parameter will inevitably produce undesired shift-variant features. To improve accuracy, several studies attempted to fuse features produced by multiple 1D CNN [127] and GRU [128]. In the end-to-end learning regime, although rarely reported, AE-based models for vibration data have been investigated; see [129]- [131].…”

Fault Diagnosis of Rotating Machinery Based on Combination of Deep Belief Network and One-dimensional Convolutional Neural Network

“…In the experiments, the proposed method is compare with the other five representative methods, including BPNN and SVM with 10 time-domain features and 9 frequency-domain features manually extracted from the raw signals [8], and 1D-CNN [9], CNNEPDNN [31], M2D-CNN [19] with the raw signals.…”

“…Chen et al [23] developed a fault diagnosis method based on CNN and discrete wavelet transform. Li et al [31] developed a network based on an ensemble CNN and deep neural network (CNNEPDNN). Gong et al [19] designed a modified 2D-CNN for fault diagnosis.…”

A Rolling Bearing Fault Diagnosis Method Based on EEMD-WSST Signal Reconstruction and Multi-Scale Entropy

“…Feature Extraction Classification Accuracy (%) [30] HOSA + PCA "one-against all" SVM 96.98 [31] Time-frequency domain ANN 93.00 [32] Time-and frequency-domains SVM 98.70 [33] IMFs decomposed by EEMD SVM with parameter optimized by ICD 97.91 [34] EEMD-MPE SSDAE 99.60 [35] CNNEPDNN CNNEPDNN 98.10 [36] FF_FC_MIC SVM 99.17 [37] HHT-WMSC SVM…”