Green Rail Equipment Safety Prediction Integrating Few-Shot Learning and Deep Models

Abstract

Against the backdrop of growing demand for intelligent transportation and green low-carbon development, urban rail transit systems have put forward higher requirements for efficient monitoring and safety prediction of equipment health status. In particular, straddle-type monorail pantographs, as key power supply components, play a vital role in ensuring the safe operation of the system. With the rapid development of urban rail transit, the pantograph of straddle-type monorails, as a key component for power supply, plays a crucial role in ensuring the safe operation of the system. However, due to the scarcity of fault data for the pantograph, traditional fault prediction methods perform poorly under conditions of small sample sizes. This study proposes a deep learning approach based on few-shot learning for fault prediction of straddle-type monorail pantographs. By combining Convolutional Neural Networks (CNN), Long Short-Term Memory Networks (LSTM), Generative Adversarial Networks (GAN), and transfer learning techniques, the study successfully constructs an efficient and accurate fault prediction model under multimodal signals. Experimental verification shows that the model is superior to traditional machine learning methods in terms of accuracy, precision, recall, F1 score, and AUC value, especially in the case of data scarcity, showing strong advantages. In addition, the robustness and adaptability of the model also indicate that it has strong practical application potential and can effectively help build a green, safe, and intelligent urban rail transit system. This study provides new ideas for the intelligent operation and maintenance of sustainable infrastructure and the safety of green rail equipment in the future.