Abstract:This study presents a novel fault diagnosis model for urban rail transit systems based on Wavelet Transform Residual Neural Network (WT-ResNet). The model integrates the advantages of wavelet transform for feature extraction and ResNet for pattern recognition, offering enhanced diagnostic accuracy and robustness. Experimental results demonstrate the effectiveness of the proposed model in identifying faults in urban rail trains, paving the way for improved maintenance strategies and reduced downtime.
What problem does this paper attempt to address?
This paper attempts to address the issue of fault diagnosis in urban rail transit systems. Specifically, the authors propose a novel fault diagnosis model based on Wavelet Transform Residual Neural Network (WT-ResNet). This model combines the advantages of wavelet transform in feature extraction and residual neural network (ResNet) in pattern recognition, aiming to improve the accuracy and robustness of fault diagnosis.
### Main Issues:
1. **Deficiencies of Existing Methods**:
- Traditional maintenance methods usually perform maintenance only after detecting mechanical faults, known as "firefighting" maintenance, leading to a passive state of maintenance work.
- Periodic diagnostic methods, although they can regularly assess the operational status of equipment, have timeliness issues and can easily miss accidents.
2. **Need for Intelligence**:
- With the development of modern information technology and artificial intelligence, especially the application of big data analysis, machine learning, and deep learning algorithms, intelligent operation and maintenance have become an effective way to improve equipment maintenance efficiency.
- Intelligent operation and maintenance systems enhance maintenance efficiency, accuracy, and reliability through intelligent functions such as remote monitoring, fault detection, expert diagnosis, and health management, while reducing labor costs and operational errors.
3. **Application of Deep Learning in Fault Diagnosis**:
- Traditional machine learning methods (such as Support Vector Machine SVM and Artificial Neural Network ANN) have limited capabilities in handling natural data and rely on domain-specific expertise.
- Deep Learning (DL) methods have been widely applied in fault diagnosis technology, particularly excelling in discovering complex structures in high-dimensional data, making them suitable for classification or detection tasks.
### Solution:
- **WT-ResNet Model**:
- **Wavelet Transform (WT)**: Used for feature extraction, capable of capturing local features of signals, particularly suitable for time-frequency analysis of non-stationary signals.
- **Residual Neural Network (ResNet)**: Used for pattern recognition, enhancing information flow through residual blocks, addressing the degradation problem in deep networks, and improving the training efficiency and generalization ability of the model.
### Experimental Results:
- Experimental validation shows that the WT-ResNet model performs excellently in identifying faults in urban rail transit systems, improving the accuracy and robustness of diagnosis, and providing an effective means to improve maintenance strategies and reduce downtime.
### Conclusion:
- The study indicates that combining wavelet transform with residual neural networks in the fault diagnosis model for urban rail transit systems yields satisfactory results. Future work should focus on extending the model to handle more types of faults and enhancing its real-time application capabilities. These findings are significant for the development of intelligent diagnostic technology in the transportation sector, with a wide range of potential applications.