Doctor of Philosophy
School of Civil, Mining and Environmental Engineering
Biabani, M Mahdi, Behaviour of geocell-reinforced subballast under cyclic loading in plane strain condition, Doctor of Philosophy thesis, School of Civil, Mining and Environmental Engineering, University of Wollongong, 2015. https://ro.uow.edu.au/theses/4376
Ballasted rail tracks have gained a competitive edge over other modes of transportation systems in terms of long term performance, better ride quality, higher safety, lower cost of construction and relatively acceptable speed and efficiency of services. In order to keep rail infrastructure costs minimal, the railway industry needs to use locally available materials during track construction, some of which do not have sufficient shear strength. At times owing to use of poor quality material, ballast and subballast cause excessive lateral spreading that leads to differential track settlement, cases of derailment and regular costly maintenance. In addition, presence of soft estuarine clay deposits along the coastal belt of Australia pose serious concerns on track stability. On other hand, considering significant demand for urban transportation, substantial urban growth, the construction of new tracks as well as the maintenance and modernisation of existing tracks have been more challenging. As result, other engineering solutions should be pursued to improve ballasted rail track substructure, which can help to maintain railways as the most economical and safest mode of transportation in Australia.
In the view of above, reinforcing the subballast is an economical alternative for stabilizing the track substructure. Unlike conventional rigid reinforcement such as steel and timber, flexible geosynthetics have shown a promising approach for improving the performance of granular media (ballast and subballast) placed over weak and soft subgrade. In the recent past, different varieties of geosynthetics, including planar (two dimensional) and cellular (three dimensional) geosynthetics, have been successfully employed. Geosynthetics have been proven to be effective in terms of reducing the settlements and enhancing track stability. Nevertheless, among different types of reinforcement, a geocell mattress due to its unique honeycomb shape, provides an effective cellular confinement, to reduce lateral displacement. Additional confinement induced by the geocell, mobilized by the tensile stresses of the membrane (i.e. hoop stress), arrests almost all lateral spreading of the infill material and increases the overall material stiffness.
It is important to note that the potential use of geocells to stabilise the ballast layer has often been regarded with some scepticism from a track maintenance point of view. In other words, cleaning and replacement of spent material is not convenient if a geocell mattress interferes with the tamper ‘tines’. In this context, Australian rail organisations have now made attempts to use geocells and other methods of stabilisation to improve the subballast that rarely requires maintenance, rather than the overlying ballast. This study was the result of applied research undertaken within the Cooperative Research Centre for Rail Innovation in collaboration with the rail organisations, namely ARTC and Sydney Trains.
In this study, triaxial tests were conducted to characterize the behaviour of reinforced and unreinforced subballast under cyclic loading using large-scale process simulation prismoidal triaxial apparatus (PSPTA) designed and built at the University of Wollongong, Australia. The laboratory tests were conducted in plane strain condition and stress controlled mode. Cyclic loading with different frequencies under very low confining pressure was applied to study the performance of subballast. Granular material with an average particle size of 3.3 mm and a geocell system with a depth of 150 mm and a nominal area of 46 x 103 mm2, made from high density polyethylene (HDPE) material, were used in this study. The laboratory results revealed that subballast stabilisation was influenced by the number of cycles, the confining pressure and the frequency.
The results proved that the geocell reinforcement is an ideal technique to improve subballast performance under very low confining pressure. The outcome of this investigation confirmed that the geocell could effectively arrest lateral spreading and reduce excessive settlement of the subballast under cyclic loading, hence increase track longevity. The results also showed that the geocell performs effectively, especially under low confining pressure (5 ≤ σˊ3 ≤ 30 kPa) and higher frequencies (10 ≤ f ≤ 30 Hz). Moreover, the geocell increased the resilient modulus of the composite layer, providing enhanced track stability of increased train speed. An optimum confining pressure required to reduce excessive volumetric dilation of the subballast was also identified in this study.
The interface shear resistance developed between the subballast and geocell has important consequences on the shear behaviour of the geocell reinforced soil. In this regard, the interface shear resistance of unreinforced and reinforced subballast with different types of geosynthetics was also investigated using a large-scale direct shear box apparatus (DSBA). The results showed that the loading mechanism had a significant impact on the interface shear strength of the subballast.
A new analytical model was developed to calculate the additional confining pressure induced by the geocell mattress. The proposed model investigate the influence of several factors i. e. (a) frequency (b) confining pressure (c) number of cycles (d) the tensile strength on the behaviour of geocell reinforced subballast. Practical design guidelines in terms of allowable train speeds for different levels of confining pressure are provided for unreinforced and geocell reinforced subballast.
Finally, a three-dimensional numerical analysis was developed for unreinforced and geocell-reinforced subballast to simulate practical or real-life railroad conditions to support experimental observations. The numerical predictions indicated that the loaddistribution mechanism of subballast could be improved by the geocells. The finite element predictions were found to be in good agreement with the laboratory data. This numerical analysis can be used as a primary tool in the design of geocellreinforced granular material with known shear strength, subjected to cyclic loading in typical railway environments.