Determination of probabilistic parameters of concrete: solving the inverseproblem by using artificial neural networks

The probabilistic approach, based on the Monte Carlo method, has been recen tly introduced to simulate cracking of concrete in the framework of a finit e element analysis [Rossi P, Wu X, le Maou F, Belloc A. Mater Struct 1994;2 7(172):437-44; Rossi P, Ulm F-J, Hachi F. J Engng Mech ASCE 1996;122(11): 1 038-43; Rossi P,Richer S. Mater Struct 1987;20(119):334-7; Rossi P, Ulm F-J . Mater Struct 1997;30(198):210-6; Fairbairn EMR, Pat CNM, Alves JLD, Silva RCC. Proceedings of XVIII CILAMCE-Iberian Latin American Congress on Compu tational Methods in Engineering, Brasilia, vol. 2, 1997;709-15]. If the unc ertainties of the material parameters are assumed to vary spatially followi ng a normal distribution, the samples corresponding to a simulation are fun ction of the mean and the standard deviation that define the Gauss density function. The problem is that these statistical moments are not known, a pr iori, for the characteristic volume of the finite elements. In this paper, neural networks are used to evaluate the parameters characterizing the stat istical distribution for a given response of the structure following an inv erse analysis procedure. It is shown that this procedure improves a recentl y proposed algorithm [Fairbairn EMR, Guedes QM, Ulm F-J. Mater Struct 1999; 32(215):9-13], which is able to solve the problem, but is very hard to oper ate. Finally, the procedure presented in this paper is used to identify the probabilistic parameters of a beam tested at TU-Delft. (C) 2000 Civil-Comp Ltd. and Elsevier Science Ltd. All rights reserved.