Bayesian two-sample tests

In this paper, we present two classes of Bayesian approaches to the two-sample problem. Our first class of methods extends the Bayesian t-test to include all parametric models in the exponential family and their conjugate priors. Our second class of methods uses Dirichlet process mixtures (DPM) of such conjugate-exponential distributions as flexible nonparametric priors over the unknown distributions.

Experimental and theoretical comparison between static and dynamic torsional soil tests

The dynamic properties of dry Leighton Buzzard sand have been investigated using a resonant column test apparatus. These data are compared with very low frequency cyclic tests on identical specimens of sand. The comparison indicates that the properties of dry sand are independent of frequency. A simple one-dimensional model of kinematic hardening plasticity is used to predict the dynamic behaviour of the sand. The input parameters for this model are based on the results of static tests. These may be conducted on standard laboratory equipment with only minor modifications. The predictions are in good agreement with the measured data.

Centrifuge tests investigating inclined grout micro-piles as a method of liquefaction remediation for existing buildings

This paper presents a series of centrifuge tests carried out to investigate the performance of non-structural inclined micro-piles as a potential liquefaction remediation method for existing buildings. Both a single-degree-of-freedom frame structure and a two-storey, two-degree-of-freedom frame structure were used as model buildings in these tests. Centrifuge tests were carried out with and without micro-piles in the foundation soil for each structure. Results primarily from the tests with the SDOF structure are presented in this paper. It is found that the micro-piles have some beneficial effect by increasing shear strains in the soil in their vicinity and hence causing dilation in these zones. However, they also increase structural accelerations by transmitting accelerations from deep in the soil and the beneficial effects from increased dilation are outweighed by the detrimental migration of pore pressures.

The influence of peel angle on the mechanics of peeling flexible adherends with arbitrary load-extension characteristics

The peel test is commonly used to determine the strength of adhesive joints. In its simplest form, a thin flexible strip which has been bonded to a rigid surface is peeled from the substrate at a constant rate and the peeling force which is applied to the debonding surfaces by the tension in the tape is measured. Peeling can be carried out with the peel angle, i.e. the angle made by the peel force with the substrate surface, from any value above about 10° although peeling tests at 90 and 180° are most common. If the tape is sufficiently thin for its bending resistance to be negligibly small then as well as the debonding or decohesion energy associated with the adhesive in and around the point of separation, the relation between the peeling force and the peeling angle is influenced both by the mechanical properties of the tape and any pre-strain locked into the tape during its application to the substrate. The analytic solution for a tape material which can be idealised as elastic perfectly-plastic is well established. Here, we present a more general form of analysis, applicable in principle to any constitutive relation between tape load and tape extension. Non-linearity between load and extension is of increasing significance as the peel angle is decreased: the model presented is consistent with existing equations describing the failure of a lap joint between non-linear materials. The analysis also allows for energy losses within the adhesive layer which themselves may be influenced by both peel rate and peel angle. We have experimentally examined the application of this new analysis to several specific peeling cases including tapes of cellophane, poly-vinyl chloride and PTFE. © 2005 Elsevier Ltd. All rights reserved.

Alternative uses of partitioning interwell tracer tests for DNAPL source zone characterisation

The heterogeneous nature of the subsurface and associated DNAPL morphologies often poses the greatest limitation to source zone clean-up strategies. Hence, detailed site characterisation techniques are required. The data presented in this paper has been collected from a series of laboratory 2-D tank experiments and numerical simulations of Partitioning Interwell Tracer Tests (PITT) in a wide range of aquifer conditions and DNAPL morphologies. Alternative uses of tracer breakthrough data have been developed In order to characterise the mass flux generated from the DNAPL source. By combining the laboratory and numerical data, a relationship between normalised mass flux and tracer-based average source zone DNAPL saturation has been established. Knowledge of such a relationship allows remediation targets to be identified, clean-up efficiencies to be evaluated, and increases the accuracy of any risk assessment.