INVESTIGATIONS OF PLANE FRAME - FOOTINGS - REINFORCED SOIL INTERACTION

Abstract

The present research work has been carried out to analytically and experimentally investigate the complete behaviour of plane frame -footingssoil/ reinforced soil system. The study would help in a better understanding of the behaviour of such interactive problem and may lead to more economical, realistic and safe design of foundations and superstructure especially in the areas where the soil is loose and weak. The technique of reinforcing weak soils to improve their bearing capacity has been in vogue for quite some time. It is relatively inexpensive and obviates the need of undertaking time consuming traditional methods of ground improvement, or alternatively going for deep foundations. However, the phenomenon of soil structure interaction for frame-footings-soil/reinforced soil system is not very well known and documented. The effect of material non-linearity on the complete response of such a system is not yet fully understood. Yet it is essential for the development of a safe and rational basis of design. The present study has, however, been taken to investigate the following aspects: (1) The method of analysis has to be formulated in such a way that, the framefootings- soil/reinforced soil system should represent a single, integral and compatible unit. (2) The weak and loose soil can be treated by providing a reinforcement of relatively high tensile strength such as geogrid or geotextile that can nicely complement the weakness of soil in tension. It is proposed to find the optimum amount of reinforcement by varying the size and the number of layers for the improved efficiency of the system. (3) To investigate the actual stress-strain relationship of the different materials in the laboratory and to suitably model it and incorporate in the proposed formulation using finite element method. (4) Laboratory tests on palne frame models and strip footing models resting on unreinforced/reinforced soil to study their behaviour and to use this data for the validation of the proposed method of analysis. (ii) (5) The development of finite element model to predict the complete response of frame-footings-unreinforced/reinforced soil system taking cognizance of material non-linearity as well as the reinforcement-soil interface shear slip characteristics. To facilitate the computer aided analysis a computer program NAFFRS has been written in fortran code and implemented on 486 machine. The theoretical study consists of the development of a plane stress-plane strain non-linear finite element model capable of predicting the complete response of the plane frame-footings-soil/reinforced soil and strip footings - soil/reinforced soil system. Further the proposed model was validated on the basis of the comparison of analytical and experimental results of models tested for both plane frame and strip footing resting on soil/reinforced soil, as well as using the published results of reported investigations. The experimental investigation includes the determination of physical and mechanical properties of sand, reinforcement and the their interface properties. For this purpose, seventy consolidated and drained triaxial tests have been performed on sand as well as on reinforced sand. The sand was placed at two relative densities of 50 % and 80%. To study the behaviour of reinforced sand, the samples were reinforced with two types of reinforcement. Geogrid Netlon-CE121 and Geotextile-PD380 placed in different number of layers were used. These tests yielded the stress-strain and Poisson's ratio-strain characteristics of plain soil as well as the composite material of reinforced soil. The stress-strain and Poisson's ratio-strain characteristics of both the materials, thus obtained, have been idealize in a nonlinear elastic mathematical model for use in the proposed formulation for theoretical prediction of the behaviour of model tests. Tensile strength tests have been conducted on a few samples of geogrid and geotextiles reinforcement to determine their stress-strain characteristics. The curves of stress-strain relationships have been idealized into tri-linear model. The yield stress as well as the ultimate stress have been defined in the model and the same incorporated in the proposed finite element formulation. Shear-slip phenomenon at the interface between reinforcement and adjacent soil have been defined and modelled in a mathematical nonlinear form and incorporated in the present finite element model. The modelling of such phenomenon is based on a series of sliding shear tests conducted on two relative densities of 50% and 80% of sand with geogrid Netlon CE-121. *- C1 C-*^ (iii) * Extensive laboratory investigations were carried out in two parts to study (i) plane frame-footings-soil/reinforced soil interaction and (ii) footing-soil/reinforced soil interaction. In the first part, eight model tests on a two-storyed single bay plane framefootings resting on soil/reinforced soil were tested under single point loads acting on both second and first floor beams at four different eccentricities given by e = 0, 0.25, 0.5 and 0.75 times half beam span from the center line of the frame. The deflections under the point of loads applications, sway of second and first storey and settlements of both footings were measured during the progress of the tests. The second part consists of 18 model tests on strip footing resting on unreinforced/reinforced soil. The number of layers as well as the width of the reinforcement are the main parameters influencing the improvement in bearing capacity and the settlement of foundation and thus have been studied. The number of layers as well as the width of the reinforcements were varied and the pressure-settlement characteristics of each model test were obtained and plotted in a graphical form. The ultimate bearing capacity as well as the corresponding settlement of each model test were evaluated. All the models tested experimentally, were analysed theoretically using .the specially developed non-linear finite element model for this purpose. Non-linear constitutive laws for soil, reinforcement and interface element between reinforcement and adjacent soils were recommended. Suitably assumed failure criteria based on the experimental results were incorporated in the finite element model. Alternatively, for improved computational efficiency, the non-linear constitutive laws of the composite materials were also evaluated to obviate the need of treating soil and reinforcement separately. This approach assumed perfect bond between soil and reinforcement. The results of two approaches of analysis (namely the discrete and the composite material approach of reinforced soil system) were compared for a strip footing-reinforced soil system. The theoretical values of settlement were predicted at different load levels and compared with the experimentally obtained results. In almost all cases it was found that the predicted values by discrete approach were in good agreement with the experimental data, whereas only in few cases the theoretical predicted values were in good agreement with the experimental data using the composite approach. In case of plane frame-r footings •Tsoil/reinforced soil interactive analysis, deflections, sway, bending moment and settlement of both footings have been predicted and these results (iv) were compared with the available experimental data from the model tests. It was found that the predicted values agreed with the experimental data. Based upon the theoretical and experimental studies undertaken, the following conclusions have been drawn: (1) An appropriate mathematical model using finite element technique was developed to predict the complete response of: (a) Plane frame-footings-soil/reinforced soil interactive problem. (b) Strip footing-soil/reinforced soil interactive problem. The proposed formulation incorporates effects of non-linearities caused by material's nonlinear stress-strain behaviour. The employment of eight-noded planar elements makes such an analysis economical viable propositions for predicting the behaviour of such type of interactive problems. (2) The experimental investigations clearly established the feasibility and the advantages of providing reinforcement layers in a loose sand, in terms of increasing the ultimate bearing capacity and reducing the settlement as well as the differential settlements between the footings of the plane frame and hence the sway and deflections of the frame. (3) The evaluation of the realistic properties of soil, reinforcement, shear-slip phenomenon and the reinforced sand composite and modelling them in a suitable non-linear mathematical form, made the finite element model capable of predicting the complete behaviour of highly interactive problems such as framefootings- reinforced soil system which is in good agreement with observed experimental test results. (v)