Train motion prognostics and classification from multi-source decentralised sensors using unsupervised data-driven technology

Key Points

• Effective classification of train motion reduces system failures and enhances safety measures, improving operational efficiency.
• The model utilizes multi-source data and unsupervised learning techniques, achieving a 95% accuracy rate on the test dataset.
• Unsupervised data-driven technology is employed to analyze train motion from various decentralised sensors, offering insights.
• This approach highlights the need for further validation in real-world scenarios to establish broader applicability.