Multimodal Learning for Arcing Detection in Pantograph-Catenary Systems

2026-02-09

AI Summary

This work addresses the challenge of detecting pantograph-catenary arcs in electrified railways, which are highly transient, obscured by significant noise, scarce in real-world samples, and easily confused with other transient phenomena. To tackle this, the authors propose MultiDeepSAD, a multimodal anomaly detection framework that fuses high-resolution visual data with force sensor measurements. The model is trained on both real and synthetically generated data acquired through cross-modal synchronized collection. A key innovation lies in modality-specific pseudo-anomaly generation strategies—such as synthetic arc artifacts in images and simulated perturbations in force signals—which effectively mitigate data scarcity and enhance generalization. Experimental results demonstrate that the proposed method significantly outperforms baseline approaches on both real and synthetic datasets, maintaining high sensitivity and robustness even under domain shift and with extremely limited real arc samples.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data

Abstract

The pantograph-catenary interface is essential for ensuring uninterrupted and reliable power delivery in electrified rail systems. However, electrical arcing at this interface poses serious risks, including accelerated wear of contact components, degraded system performance, and potential service disruptions. Detecting arcing events at the pantograph-catenary interface is challenging due to their transient nature, noisy operating environment, data scarcity, and the difficulty of distinguishing arcs from other similar transient phenomena. To address these challenges, we propose a novel multimodal framework that combines high-resolution image data with force measurements to more accurately and robustly detect arcing events. First, we construct two arcing detection datasets comprising synchronized visual and force measurements. One dataset is built from data provided by the Swiss Federal Railways (SBB), and the other is derived from publicly available videos of arcing events in different railway systems and synthetic force data that mimic the characteristics observed in the real dataset. Leveraging these datasets, we propose MultiDeepSAD, an extension of the DeepSAD algorithm for multiple modalities with a new loss formulation. Additionally, we introduce tailored pseudo-anomaly generation techniques specific to each data type, such as synthetic arc-like artifacts in images and simulated force irregularities, to augment training data and improve the discriminative ability of the model. Through extensive experiments and ablation studies, we demonstrate that our framework significantly outperforms baseline approaches, exhibiting enhanced sensitivity to real arcing events even under domain shifts and limited availability of real arcing observations.

Problem

• arcing detection
• pantograph-catenary system
• multimodal learning
• transient phenomena
• data scarcity

Innovation

• multimodal learning
• arcing detection
• MultiDeepSAD
• pseudo-anomaly generation
• pantograph-catenary system