Data-Driven Wheel Slip Diagnostics for Improved Railway Operations

Wheel slip activity detection is crucial in railway maintenance, as it can contribute to avoiding wheel damage but also track deteriorations leading significant maintenance costs, trains delays, well the risk of accidents. characterised by lower adhesion between and wheel, especially braking conditions, locking wheels. It complex model or predict, being influenced a multitude factors including ambient global vehicle load, axle quality, leaves objects present on rail, steep incline, oxidation rails, forces applied This paper presents combined wavelet tuned Long-Short Term Memory (LSTM) approach for from time series data collected real-world trains. Results provide evidence superior performance over methods such decision trees random forests, naïve Bayes, k-nearest neighbours, logistic regression, support vector machines.