Degree Name

Doctor of Philosophy


Department of Civil and Mining Engineering


This thesis was concerned with the design, analysis and geometrical optimisation of reinforced soil dams (RSDs, singular RSD)and aimed to develop a computer program for these tasks. In order to achieve this objective, the following tasks were carried out as parts of this thesis.

(a) Comprehensive literature survey. This part included an overview of the history of reinforced soil, its application, material components, design considerations, construction methods, and economic considerations of reinforced soil. The fundamentals of design and analysis of conventional earth dams were also considered. This included the history, classification, factors governing the choice of dam type, stability analysis, design criteria, and limitations of conventional earth dams. This project also gave a detailed evaluation and design criteria of a number of existing RSDs. A summary of recent investigations on the behaviour of RSDs was outlined. This continued by considering the classification of RSDs and the forces acting on them.
(b) Stability analysis of RSDs. This part was focused on some formulae in order to optimise the cross sectional area of RSDs. The external stability analysis of RSD was evaluated as a whole structure based on analytical approach. Some design formulae were given for RSD optimisation concerning the minimum base length required against sliding, overturning, overstressing, bond failure, and rupture failure. The internal stability of RSD was also taken into account based on a semi-empirical method. Some empirical relationships were proposed to eliminate the tangent discontinuity which exists in the Coherent Gravity Method formulae. These relationships reflect the non-linearity indicated by thefield data and eliminate unknown parameters existing in the formulae of Modified Coherent Gravity Method. They also offer a better fit with the available field observations. Relationships between the lateral earth pressure and the apparent friction factor with the fill depth were proposed. The apparent friction factor versus the reinforcement length were also undertaken.
(c) Analysis ofthe behaviour under seismic loads. Although many researchers have investigated the effects of earthquakes on soil dams, many problems are still unsolved, specially for RSDs. A comparison between the natural frequency of conventional earth dams and RSDs were considered in this project. The practice of inserting reinforcement into the earth dam material allows reduction in fill volume and reduction in displacement. However, this also leads to an increase in the natural frequency of such structures compared with conventional earth dams. This may increase the possibility of failure. The natural frequency of RSD is increased because of its geometry and its overall stiffness. In this project, the increases in natural frequency of RSDsdue to these two major factors were separately discussed. Formulae concerning the magnification of the natural frequency of the structure due to reinforcement insertion were derived, and in some cases tabulated and plotted.
(d) Development of a computer program. A computer program was developed for geometrical optimisation and stress-strain analysis of RSD. The outcomes of the program are (a) geometrical optimisation of RSD based on analytical and semi- empirical formulae, and (b) stress-strain analysis of the optimised RSD based on the finite element method.
(e) Analysis of models of RSDs. Six models of RSDs were analysed for various heights and safety factors to find the optimum geometry. A 30m high RSD was also analysed considering the following configurations: (a) without reinforcements, (b) with the assumed increased stiffness of the soil fill, (c) with horizontal reinforcements, and
(d) with inclined reinforcements to evaluate the variation of stresses versus the direction of reinforcements. It was concluded that putting reinforcements in soil dams decreases displacements and stresses values.



Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.