Train Tilt:Machine Learning Model for Computation of Real Time Angles for Trains

This study introduces a machine learning-based model to predict permissible tilting angles for tilting trains, with the goal of improving travel efficiency and passenger comfort. The model leverages an interpolation algorithm to estimate tilting angles using two critical inputs: train speed and track curvature radius. Historical data on train operations, including speeds, curvature radii, and associated tilting angles, were collected and processed to construct a comprehensive dataset. Nonlinear relationships between inputs were captured through polynomial feature engineering, and a linear regression model was trained to deliver precise tilting angle predictions. These predictions enable real-time adjustments to the train's tilting mechanism, enhancing performance on curved tracks. The approach provides an economical solution to enhance train speed and stability on existing railways without requiring extensive infrastructure upgrades. Real-world data validation confirmed the model's effectiveness in ensuring smooth and secure travel experiences. Integrating this predictive model into train control systems can revolutionize rail transport by boosting efficiency, passenger comfort, and operational dependability. This work underscores the transformative potential of machine learning in advancing rail system optimization.