A simulation-based approach for railway applications using GPR

In this work a numerical model capable to predict the electromagnetic response of railway ballast aggregates under different physical conditions has been calibrated and validated by a simulation-based approach. The ballast model is based on the main physical and geometrical properties of its constituent material and it is generated by means of a random-sequential absorption (RSA) approach. A finite-difference time-domain (FDTD) simulator is then employed to calculate the ground-penetrating radar (GPR) signal response to the scenario. The calibration of the model has been performed by taking into account the main physical properties and the grain size characteristics of both the reference ballast material and a fine-grained pollutant material, namely, an A4 soil type material, according to the AASHTO soil classification. The synthetic GPR response has been generated by using the gprMax freeware simulator. Several scenarios have been considered, which in turn were reproduced in laboratory environment and used for the validation of the model. Promising results have demonstrated the high potential of such approach in characterizing the simulated response of complex coarse-grained heterogeneous materials.