Use of Centrifuge-Based Physical Modelling for Understanding the Performance of Geostructures

Abstract

Centrifuge-based physical modelling is widely adopted for understanding the performance of geostructures, like levees subjected to flooding and drawdown and geogrid-reinforced soil walls subjected to seepage. In this paper, an attempt has been made to bring out the advantage of centrifuge-based physical modelling to understand (i) the performance of levees subjected to flooding using a custom-designed and developed in-flight flood simulator at 30 gravities with and without chimney drain and (ii) the performance of geogrid-reinforced soil walls with and without chimney drain subjected to seepage at 40 gravities. In both the cases, silty sand was used to model soil and fine sand was used in chimney drain. All centrifuge model tests were performed using the 4.5-m-radius large-beam centrifuge facility available at IIT Bombay. Models were instrumented with linearly variable differential transformers for measuring surface settlements and pore pressure transducers to measure raise in pore water pressure within the soil at the onset of flooding for levees and at the onset of seepage for geogrid-reinforced soil walls. Additionally, digital image analyses of photographs of front elevation of levee models and geogrid-reinforced soil wall models were carried out to obtain face movements, movements of markers embedded within the levee, markers stuck to geogrid layers of reinforced soil walls at the onset of flooding and seepage. The developed in-flight flood simulator was found to be capable of generating the flood rate ranging from 2.2 to 7 m/day. Further, results of centrifuge model tests conducted on levees without any chimney drain were noticed to undergo catastrophic failure within 4.25 days of flooding-induced seepage, whereas a levee with chimney drain was found to sustain flooding-induced seepage of 37.5 days. Geogrid-reinforced soil wall was constructed with silty sand as a structural as well as backfill, without any drainage system experienced catastrophic failure. Contrary to this, geogrid-reinforced soil walls with chimney drain as an external drainage system helped in averting catastrophic failure. However, probability of piping failure near the toe region of the wall cannot be ruled out. Further, the use of geocomposite layers as an internal drainage system within the reinforced zone also explored and the placement of geocomposite layers at one-third portion of height from bottom was found to be effective.