Experimental and numerical analyses of geosynthetic-reinforced soil slopes
Date of Award
Doctor of Philosophy
School of Engineering
Associate Professor Sanjay Kumar Shukla
The application of geosynthetic reinforcements in civil/geotechnical engineering projects (retaining walls, foundations, pavements, dams, slopes, etc.) has gained much popularity during the past few decades due to several benefits, including cost-effectiveness, environmentally friendly and sustainability. A detailed literature review as presented in this thesis has indicated that when a slope is reinforced with the geosynthetic layer(s), it improves the overall stability of the slope with or without loaded footing on the slope crest. However, studies on the performance of strip footings embedded in the slopes are very limited, and, especially for the geosynthetic-reinforced slopes, there is no work when the slope is reinforced with geosynthetic layers with or without wraparound ends. Also, there is no available literature on the design charts for low-height slopes, without footing/surcharge loads on the crest, which are usually constructed for the landscape developments in many countries. Furthermore, the literature has no information on the stability charts for reinforced sand slopes carrying embedded strip footing subjected to loads.
This thesis work is based on the laboratory experiments and numerical simulations. The laboratory model tests were conducted on a sand slope supporting an embedded strip footing (width B = 75 mm ) in a rigid test tank (internal dimensions of 1250 mm × 445 mm in plan and 800 mm in height). The slope was reinforced with a single and multilayer geotextile with and without wraparound ends as different test trials. The model tests were conducted to evaluate the effect of the footing embedment depth D, footing edge distance e , number of geotextile layers N , and wraparound end of geotextile on the behaviour of the embedded footing. The footing was subjected to incremental loads to observe the corresponding stabilised settlements until it failed. The slope angle and relative density of the sand were maintained at constant values, β = 35 and Dr = 70%, respectively, throughout the laboratory experiments. For the case of the single geotextile layer with no wraparound ends, the geotextile was installed at the depth ratio u / B = 0.5 below the base of the footing which was first fixed at the edge distance ratio e / B = 1, while the depth ratio (D/ B)was varied from 0 to 1.5. After that, the footing was maintained at a constant depth ratio D/ B = 1 while the edge distance ratio (e / B) was varied from 0 to 3. In the case of the multilayer geotextile (N = 2, 3) , with no wraparound ends and single layer geotextile with wraparound ends, the top geotextile layer was placed at the depth ratio u / H = 0.5 below the base of the footing and the subsequent layers were positioned at a constant vertical spacing(h) to footing width ratio h / B = 0.5 from the top layer. The footing edge distance ratio was kept constant as e / B = 1 while depth ratio (D/ B) was varied from 0 to 1. The numerical models for the laboratory experiments were developed using the Plaxis 2D, a finite element package. The numerical analysis utilised the Mohr-Coulomb criterion to model the slope soil, the geogrid option to model the geotextile layer(s), the gravity force to simulate the initial stress condition within the slope and prescribed footing load option to simulate the applied footing loads accompanied by iterative analysis until failure occurred. The developed numerical after validation has been used for a detailed parametric study in order develop design charts for the stability of slopes with embedded footing. Additionally, the stability (factor of safety) analysis of a geotextile-reinforced low-height sandy slope, without footing or surcharge loads, was carried out using the limit equilibrium method available in Slope/W package.
The experimental results indicate that the bearing capacity of the footing increases with increasing D/ B , e / B and N . The benefits derived from reinforcing the slope with geotextile layers have been evaluated using a non-dimensional parameter, called the ultimate bearing capacity ratio BCRu , defined as the ratio of ultimate bearing capacity of the reinforced case to that of unreinforced case. In the case of the single layer geotextile without wraparound ends, the maximum value of BCRu ≈ 2.5 − 3 is observed for D/ B = 0 and e / B = 0 , while the minimum value of BCRu ≈1.5 has been obtained for D/ B =1and e / B = 3 . The BCRu for the multilayer geotextile with no wraparound ends improves with an increase in N but reduces with an increase in D/ B . The minimum BCRu , BCRu (min) ≈ 2 , is observed for N =1 and D/ B =1, while the maximum BCRu , BCRu (max) ≈ 6 is attained when the footing is placed at D/ B = 0 and N = 3 . The installation of the single layer geotextile with wraparound ends brings an additional improvement in the bearing capacity of the footing compared to the case of no wraparound ends. The results obtained from the numerical simulations, on the load-settlement analysis of the embedded footing, closely agree with the experimental data, particularly for low settlements.
The results from the numerical slope analysis show that the factor of safety (F) of the unreinforced sandy slope with an embedded footing increases with an increase in the footing edge distance ratio (e / B) , footing depth ratio (D/ B) and soil relative density(Dr ) , but it decreases with an increase in the slope angle (β ) and applied pressure on the footing(q) . For the surface footing (D/ B = 0) , F increases to a critical value at e / B = 3 then remains constant for e / B > 3. Though in the experimental study, only Dr = 70% was used, in the numerical simulations, = 50% r D and = 90% r D have also been considered. The study shows that with respect to increase in Dr , F significantly improves until Dr = 70%; after that, further increase in reduces the rate of increase in F .
For the low-height sandy slopes, placing a single geosynthetic reinforcement layer at the depth ratio u / H = 0.5 in the 40° slope results in a stable slope with a maximum factor of safety Fr (max) = 1.61 , but this depth is not appropriate to stabilize the 50° and 60° slopes. The study shows that three geosynthetic layers are generally not be required as the two-reinforcement layers are adequate to attain the minimum factor of safety as usually recommended in most standards on stability of slopes. This thesis has many graphical presentations, which can be used as the design charts by the practising engineers.
Access to this thesis is embargoed until 14 August 2021. At the expiration of the embargo period, access to the thesis will be restricted to current ECU staff and students. Email queries to firstname.lastname@example.org
Baah-Frempong, E. (2019). Experimental and numerical analyses of geosynthetic-reinforced soil slopes. https://ro.ecu.edu.au/theses/2231