An attempt to recycle the massive quantum of mine overburden (OB) waste, generated as a by-product during coal extraction has been made in this study. Experimental and numerical investigations have been carried out to better understand its behavior as a railway subballast material under cyclic loading. On model railway tracks with varying OB-subballast layer thicknesses reinforced with geocells of various pocket sizes, a series of cyclic model tests were conducted. Three-dimensional cyclic FEM models were used to validate the model tests. The validated FEM models served as training data sets for the artificial neural network (ANN) technique, which was used to predict deformations. As the loading cycles increased, the geocells became more effective at reducing track deformation rates. Correspondingly, vertical strains and stresses in the track beds were also reduced in a noticeable way. The findings also supported the use of a shallow OB-subballast thickness in the presence of a geocell, which can withstand track degradations for longer load cycle durations and thus prolong costly track maintenance cycles. The ANN technique, which demonstrated good agreement with the numerical output, can predict the displacements of geocell reinforced tracks in a reliable manner and serve as an efficient deformation prediction tool for railways.
All Science Journal Classification (ASJC) codes
- Civil and Structural Engineering
- Geotechnical Engineering and Engineering Geology