PARAMETER AND VARIANCE-ESTIMATION IN GEOTECHNICAL BACKANALYSIS USING PRIOR INFORMATION

A probabilistic framework to perform inverse analysis of geotechnical problems is presented. The formulation allows the incorporation of exi sting prior information on the parameters in a consistent way. The met hod is based on the maximum likelihood approach that allows a straight forward introduction of the error structure of field measurements and prior information. The difficulty of ascribing definite values to the uncertainties associated with the various types of observations is ove rcome by including the corresponding variances in the set of parameter s to be identified. The inverse analysis results in a minimization pro blem that is solved by coupling the optimization technique to the fini te element method. Two examples are presented to illustrate the perfor mance of the method. The first one corresponds to a synthetic case sim ulating the excavation of a tunnel. Young's modulus, K-0 value and mea surements variances ape identified. The second case concerns the excav ation of a large underground cavern in which again Young's modulus and K-0 are identified. It is shown that introduction of prior informatio n permits the estimation of parameters more consistent with all availa ble informations that include not only monitored displacements but als o results from in situ tests carried out during the site investigation stage.