The pantograph–catenary system is a critical component of railway vehicles, and its performance directly affects the quality of current collection. Accurately measuring the arcing rate is essential for monitoring the system’s condition and ensuring safe operation. However, traditional arc detection methods are prone to increased false detection rates and reduced measurement accuracy in complex railway environments due to the diversity of arc sizes and shapes, environmental interference, instability in current collection, and power fluctuations. While deep learning-based methods can effectively address environmental interference, obtaining sufficient labeled training data is challenging because arc events occur infrequently. Moreover, a large number of unlabeled images of pantograph–catenary contacts cannot be directly utilized due to the lack of annotations. To solve these issues, a novel arc detection method is proposed: a multimodal arc detection network based on denoising diffusion probabilistic models (DDPMs-MILNet). First, a DDPM is pretrained using a large set of unlabeled images to acquire advanced image features. This model serves as a feature extractor, and a hierarchical variation semantic decoder is fine-tuned, thereby improving performance under small-sample conditions and reducing dependence on extensive labeled datasets. Building on this, an audiovisual semantic decoder is designed to incorporate audio signals as semantic cues, providing additional modality information for visual features. This approach not only reduces the model’s reliance on visual information but also enables it to locate the visual target of the arc even when the object is not simultaneously seen and heard, further alleviating the challenges posed by limited sample sizes. Experimental results demonstrate that DDPM-MILNet achieves excellent detection performance with minimal data in complex railway environments, indicating significant application potential, particularly in the state monitoring and anomaly detection of railway systems.

Thanks to the development of deep learning, machine abnormal sound detection (MASD) based on unsupervised learning has exhibited excellent performance. However, in the task of unsupervised MASD, there are discrepancies between the acoustic characteristics of the test set and the training set under the physical parameter changes (domain shifts) of the same machine's operating conditions. Existing methods not only struggle to stably learn the sound signal features under various domain shifts but also inevitably increase computational overhead. To address these issues, we propose an unsupervised machine abnormal sound detection model based on Transformer and Dynamic Graph Convolution (Unsuper-TDGCN) in this paper. Firstly, we design a network that models time-frequency domain features to capture both global and local spatial and time-frequency interactions, thus improving the model's stability under domain shifts. Then, we introduce a Dynamic Graph Convolutional Network (DyGCN) to model the dependencies between features under domain shifts, enhancing the model's ability to perceive changes in domain features. Finally, a Domain Self-adaptive Network (DSN) is employed to compensate for the performance decline caused by domain shifts, thereby improving the model's adaptive ability for detecting anomalous sounds in MASD tasks under domain shifts. The effectiveness of our proposed model has been validated on multiple datasets.

The Pantograph-Catenary System (PCS) is a critical component of railway vehicles, and its performance directly affects current collection quality. The arc rate serves as an essential measurement indicator for monitoring PCS state. However, in complex railway environments—where arc sizes and shapes can vary significantly and are further influenced by factors such as reflected light, glare, and adverse weather—traditional arc detection methods are easily affected by unstable current collection and power fluctuations, resulting in increased false detection rates and reduced measurement accuracy. Deep learning methods, while promising, also face limitations when dealing with such diverse arc morphologies and strong external interference. To address these challenges, this paper proposes a Multi-Modal Imitation Learning-based Arc Detection Network (MILADNet). First, the measurement system fuses infrared and visible-light image features to enhance arc feature extraction in scenarios with strong glare or reflective interference, thereby mitigating false alarms caused by relying on a single sensor. Second, to overcome the lack of information on small arcs, an online imitation learning framework is introduced to improve the system’s detection sensitivity for small arcs. Finally, to address data bias arising from uneven arc distributions, an unsupervised transferable representation learning method is employed to reduce dependence on labeled data and enhance model generalization. Experimental results show that MILADNet exhibits outstanding detection performance for arcs of various sizes and in complex environments, demonstrating both high efficiency and accuracy during measurement and data processing. Beyond improving the precision and reliability of arc detection, this method offers a novel solution for the instrumentation and measurement field and shows significant potential for condition monitoring and anomaly detection in railway systems.