Full-scale model study on variations of soil stress in geosynthetic-reinforced pile-supported track bed with water level change and cyclic loading

This study presents a full-scale model investigation on variations of soil stress in a geosynthetic-reinforced pile-supported track bed at various water levels and loading cycles, with four testing procedures: water level rising, cyclic loading at high water level, water level lowering, and cyclic loading at low water level. The soil arching effect was revealed, characterized by higher stress above the pile cap. With the water level rising and loading cycles increasing at high water level, this effect becomes more pronounced, until a peak value of dynamic stress concentration ratio is reached. The stable state of soil arching is obtained earlier near the crown of soil arching, but this arching effect develops more significantly at the foot of soil arching. With the water level lowering and loading at low water level, the soil arching effect remains steady, with slightly changed dynamic stresses in the track bed. The geogrid shows a significant impact on the load transfer mechanism for the quasi-static stress: the quasi-static pile-cap stress presents higher values below the geogrid, whereas the opposite trend is observed for the water-bag (subsoil) area. Nevertheless, this mechanism is not obvious with respect to the dynamic stress, with the values showing no distinct difference above and below the geogrid.