Parameter estimation of a static geotechnical model using a Bayes' approach

Parameters concerning the geometry (length and location of' an inclusion) and the material properties (elastic moduli) of a system, modelled as linearly elastic and in plane-strain, are estimated on the basis of the knowledge of displacements due to given loads. The problem is studied by means of a Bayesian approach, where the a priori information concerning the unknown parameters is optimally combined with the a posteriori knowledge supplied by the displacement measurements. This is a non -trivial problem since the geotechnical model is non -linear with respect to the parameters. An iterative estimation algorithm is used which is shown to be computationally efficient for the problem at hand. The results achieved are compared to those previously obtained by a simulation procedure