Train Classification Using a Weigh-in-Motion System and Associated Algorithms to Determine Fatigue Loads.

Area of Science:

• Railway engineering
• Mechanical engineering
• Data science

Background:

• Accurate calibration of railway fatigue load models is crucial for infrastructure safety and maintenance.
• Existing methods may lack the precision to identify individual vehicles and continuously calibrate weigh-in-motion (WIM) systems.
• Understanding train types and vehicle loads is essential for predicting structural fatigue.

Purpose of the Study:

• To develop and present a methodology for classifying train passages using a WIM system.
• To enable continuous calibration of railway WIM stations by identifying individual vehicles from in-service trains.
• To ensure the quality assurance of measured responses for fatigue load model calibration.

Main Methods:

• Utilizing a weigh-in-motion (WIM) system for data acquisition.
• Implementing a data processing method for raw measurements.
• Developing an algorithm for automatic identification of train types and individual vehicles.
• Applying statistical methods for quality assurance of measured responses.

Main Results:

• A methodology for classifying train passages and identifying individual vehicles was successfully developed.
• The quality assurance of measured responses was demonstrated to be satisfactory for fatigue load model calibration.
• Measurement errors were found to be within acceptable limits for the intended application.
• The study identified the limits of the obtained load spectra.

Conclusions:

• The proposed methodology effectively classifies train passages and identifies individual vehicles using WIM data.
• The system allows for continuous calibration of railway WIM stations, improving data accuracy.
• The findings support the use of WIM systems for accurate railway fatigue load model calibration and understanding traffic conditions.