Centrifuge modeling and analysis of active trapdoor in sand

The trapdoor problem is a useful model for providing a clearer understanding of stress distribution around basic geotechnical engineering structures such as anchor plates and tunnels. The stability of the trapdoor is considered to be a classical problem of soil arching action where soil pressure is transferred from a yielding support to an adjacent non-yielding support. The interest is mainly directed towards the determination of ground pressure on the trapdoor, which can be substantially different from the initial geostatic pressure when the door is moved even slightly.; Experimental, analytical and numerical investigations were conducted and studied to understand the behavior of active circular trapdoor, with dry cohesionless overburden soil. The study was undertaken mainly because of four reasons: (i) there has been no successful centrifuge modeling of the problem, especially in obtaining a good initial static load under high gravity condition, (ii) lack of understanding of the problem under axisymmetric condition, (iii) lack of understanding of the behavior of dead-load surface footing on trapdoor, and (iv) lack of the centrifuge experimental data to validate the finite element and analytical models for this specific problem. A new design trapdoor system and a special inflight precompression technique were developed to perform a series of tests involving the different overburden depths and weights of surface footing. In addition, a set of half-cut container tests were conducted to observe localization of shear bands under single and high gravity conditions.; Classical and recent analytical methods were adapted and compared with the experimental results in both plane strain and axisymmetric conditions. The experiments were simulated using a finite element code, PLAXIS. The comparisons were interpreted in order to gain an understanding of the basic behavior of the trapdoor problem.; The reliable failure load curve with a correct initial geostatic load was obtained and found to be more accurate than the past researches. Both modeling of models and repeatability experiments were conducted successfully in terms of load and surface settlement. Localisation of shear bands under a high gravity field differed considerably from the normal gravity condition. Evaluations of analytical solutions in terms of input parameters and basic assumptions were made. The numerical results showed that good predictions of load at the maximum arching point were obtained. There was no success in matching the state of force recovery and minimum arching. However, rate of surface settlement was predictable for the entire range of H/D ratio.