Railway ballast condition assessment using ground-penetrating radar - an experimental, numerical simulation and modelling development

http://sfx.cilea.it:9003/sfxcab3/sf(opens in a new window)|View at Publisher| Export | Download | Add to List | More... Construction and Building Materials Volume 140, 1 June 2017, Pages 508-520 Railway ballast condition assessment using ground-penetrating radar – An experimental, numerical simulation and modelling development (Article) Benedetto, A.a , Tosti, F.b , Bianchini Ciampoli, L.a , Calvi, A.a , Brancadoro, M.G.a , Alani, A.M.b a Department of Engineering, Roma Tre University, Via Vito Volterra 62, Rome, Italy b School of Computing and Engineering, University of West London (UWL), St Mary's Road, Ealing, London, United Kingdom View references (52) Abstract This paper reports on the ground-penetrating radar (GPR)-based assessment of railway ballast which was progressively “polluted” with a fine-grained silty soil material. It is known how the proper operation of a ballast track bed may be undermined by the presence of fine-grained material which can fill progressively the voids between the ballast aggregates and affect the original strength mechanisms. This occurrence is typically defined as “fouling”. To this effect, a square-based methacrylate tank was filled with ballast aggregates in the laboratory environment and then silty soil (pollutant) was added in different quantities. In order to simulate a real-life scenario within the context of railway structures, a total of four different ballast/pollutant mixes were introduced from 100% ballast (clean) to highly-fouled (24%). GPR systems equipped with different air-coupled antennas and central frequencies of 1000 MHz and 2000 MHz were used for testing purposes. Several processing methods were applied in order to obtain the dielectric permittivity of the ballast system under investigation. The results were validated using the “volumetric mixing approach” (available within the literature) as well as by performing a numerical simulation on the physical models used in the laboratory. It is important to emphasize the significance of the random-sequential absorption (RSA) paradigm coupled with the finite-difference time-domain (FDTD) technique used during the data processing. This was proved to be crucial and effective for the simulation of the GPR signal as well as in generating synthetic GPR responses close to the experimental data.