Abstract:The symmetry of vibration signals collected from healthy machinery, which gradually degenerates with the development of faults, must be detected for timely diagnosis and prognosis. However, conventional methods may miss spatiotemporal relationships, struggle with varying sampling rates, and lack adaptability to changing loads and conditions, affecting diagnostic accuracy. A novel bearing fault diagnosis approach is proposed to address these issues, which integrates the Gramian angular field (GAF) transformation with a parallel deep convolutional neural network (DCNN). The crux of this method lies in the preprocessing of input signals, where sampling rate normalization is employed to minimize the effects of varying sampling rates on diagnostic outcomes. Subsequently, the processed signals undergo GAF transformation, converting them into an image format that effectively represents their spatiotemporal relationships in a two-dimensional space. These images serve as inputs to the parallel DCNN, facilitating feature extraction and fault classification through deep learning techniques and leading to improved generalization capabilities on test data. The proposed method achieves an overall accuracy of 96.96%, even in the absence of training data within the test set. Discussions are also conducted to quantify the effects of sampling rate normalization and model structures on diagnostic accuracy.

What problem does this paper attempt to address?

The paper attempts to address the following issues in rolling bearing fault diagnosis: 1. **Insufficient Spatiotemporal Relationship Capture**: Current time-frequency domain analysis methods combined with deep learning algorithms have limitations in capturing the spatiotemporal relationships between sampling points in the input time series. This is particularly challenging when dealing with complex and nonlinear bearing fault signals, as these methods struggle to fully reveal the intrinsic structure and dynamic characteristics of the signals, thereby affecting the accuracy and reliability of the diagnosis. 2. **Poor Adaptability to Different Sampling Rates**: When the sampling rate of the input time series is inconsistent with the training data, deep learning models produce biased diagnostic results. This indicates a need to enhance the model's ability to extract features and recognize patterns under different sampling rates. Current research on this issue is insufficient, and no optimized solutions have been proposed to improve model performance under varying sampling rates. 3. **Insufficient Generalization Ability**: In practical applications, although the model performs well on training data collected under specific load conditions, the diagnostic accuracy decreases when load conditions change and the current load situation is not included in the training set. This highlights the model's limitations in adapting to different operating conditions and load variations. To address the above challenges, this paper proposes a novel bearing fault diagnosis method based on Gramian Angular Field (GAF) transformation and parallel Deep Convolutional Neural Networks (DCNN), aiming to solve these three problems simultaneously. Specifically: - GAF is used to convert waveforms collected under different working conditions into image formats, generating Gramian Angular Summation Field (GASF) and Gramian Angular Difference Field (GADF) images, effectively mitigating common-mode and differential-mode interference in the signals. - For the situation where the sampling rate of the input time series differs from the training data, an upsampling method based on cubic spline interpolation is introduced to further improve the accuracy of diagnostic results, and the effectiveness of this method is verified through experiments. - A bearing fault diagnosis method based on parallel DCNN is proposed. Each CNN contains two convolutional layers to comprehensively extract vibration patterns under different operating conditions. These networks process image data generated by GASF and GADF respectively, and use an attention mechanism to fuse the extracted features, ultimately achieving comprehensive fault diagnosis. Experimental results show that this method has strong adaptability under different load conditions.

A Novel Method for Bearing Fault Diagnosis Based on a Parallel Deep Convolutional Neural Network

Bearing fault diagnosis with parallel CNN and LSTM

A Novel Cross-Domain Mechanical Fault Diagnosis Method Fusing Acoustic and Vibration Signals by Vision Transformer

Rolling bearing fault diagnosis based on feature fusion with parallel convolutional neural network

Bearing Feature Extraction and Fault Diagnosis Algorithm Based on Convolutional Neural Networks

A new bearing fault diagnosis method based on modified convolutional neural networks

Hierarchical adaptive deep convolution neural network and its application to bearing fault diagnosis

Optimal Time Frequency Fusion Symmetric Dot Pattern Bearing Fault Feature Enhancement and Diagnosis

A Novel Bearing Fault Diagnosis Method Based on 2D Image Representation and Transfer Learning-Convolutional Neural Network

A novel fault diagnosis method of rotating machinery via VMD, CWT and improved CNN

Bearing Fault Diagnosis of Single-Channel Data by a 3D DCN with Bilinear LBP and Modified KPCA

A Bearing Fault Diagnosis Method Based on Spectrum Map Information Fusion and Convolutional Neural Network

A novel diagnostic framework based on vibration image encoding and multi-scale neural network

Bearing fault diagnosis with varying conditions using angular domain resampling technology, SDP and DCNN

Bearing Fault Diagnosis using Normalized Diagnostic Feature-gram and Convolutional Neural Network

A novel two-stream multi-head self-attention convolutional neural network for bearing fault diagnosis

Rolling Bearing Fault Diagnosis Based on Convolutional Neural Network and Support Vector Machine

A multi-scale collaborative fusion residual neural network-based approach for bearing fault diagnosis

Intelligent Fault Diagnosis of Bearing Based on Convolutional Neural Network and Bidirectional Long Short-Term Memory

Convolutional Neural Network with Attention Mechanism and Visual Vibration Signal Analysis for Bearing Fault Diagnosis

Motor Bearing Fault Diagnosis Using Deep Convolutional Neural Networks with 2D Analysis of Vibration Signal