Machine learning framework detects railway wheel defects using ultrasonic signals

Researchers have developed a machine-learning framework to detect defects in railway wheels using passive ultrasonic signals. The system analyzes acoustic emission data from wheelsets, identifying key features in both time and frequency domains. A Random Forest classifier, trained with these features, achieved a balanced accuracy of approximately 0.66 and a Macro-F1 score of 0.65 across nine different defect classes, demonstrating the potential for non-contact inspection systems. AI

IMPACT This research demonstrates a novel application of machine learning for safety-critical infrastructure, potentially improving inspection efficiency and reliability.

RANK_REASON Academic paper detailing a new machine learning framework for defect detection. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• arXiv
• Kruskal–Wallis one-way analysis of variance
• random forest

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →