ESTIMATION OF PARAMETERS IN GEOTECHNICAL BACKANALYSIS .1. MAXIMUM-LIKELIHOOD APPROACH

The estimation of soil and rock parameters based on field instrumentat ion data is a common procedure in geomechanics. The use of system iden tification and optimization techniques allows the performance of this type of analyses in a more rational-and objective manner. In this pape r a probabilistic formulation for the backanalysis problem is presente d. The procedure described involves the evaluation of the measurement covariance matrices, which are derived for some geotechnical instrumen ts used in field instrumentation. The algorithm used to solve the math ematical problem of optimization is also presented, as well as its cou pling to a finite element code. The algorithm requires the computation of the sensitivity matrix, which can be evaluated ''exactly'' in term s of the finite element method. Finally, a synthetic example, based on the excavation of a tunnel, is presented in which the elastic modulus E and the K-0 parameter of the material are identified from measured displacements. The effect of the number of measurements and their erro r structure is also discussed.