Forecasting the Fatigue Life of Rails R65 Using Digital Technologies and Artificial Intelligence

Abstract

The paper examines modern approaches to predicting the fatigue life of R65 rails using digital technologies and artificial intelligence. It is shown that the integration of numerical modeling, monitoring, and machine learning significantly improves the accuracy of residual life assessment and timely defect detection. Mathematical fatigue models, numerical analysis results, and an example of applying machine learning algorithms are presented.