SMART RAIL Project

The SMART RAIL project: “MACHINE LEARNING ANALYSIS OF MASS-CAPTURE GEOPHYSICAL DATA WITH MULTIARRAY TECHNOLOGY FOR INTELLIGENT MAINTENANCE OF RAILWAY INFRASTRUCTURE” led by Rover Rail, is part of the Cooperation R&D Projects program of the Valencian Institute for Competitiveness and Innovation (IVACE+i), and has the participation of Geozone Asesores S.L. This project, co-financed by the European Union within the framework of the Valencian Community program funds ERDF, develops an advanced system of geophysical exploration of the railway substructure using non-invasive techniques (georradar), based on machine learning data post-processing to measure the state of the ballast.

For this, a mass data collection system with multiarray technology has been implemented in railway machinery, for the subsequent processing of complex sets of geophysical data collected on the track. This allows for the detection of anomalies in ballast compactness, degree of contamination, subsidence and/or pathologies that may be found on the subplatform.

From the graphic and numerical correlation of patterns present in the data, machine learning models of the state of the track sections analyzed have been developed for implementation in predictive maintenance. In this way, it is possible to alert areas of the track susceptible to local intervention at incipient stages, to anticipate the appearance of pathologies or high levels of deterioration during the life cycle of the infrastructure.