Determination of elastic properties of a film-substrate system by using the neural networks

An inverse method based on artificial neural network (ANN) is presented to determine the elastic properties of films from laser-genrated surface waves. The surface displacement responses are used as the inputs for the ANN model; the outputs of the ANN are the Young's modulus, density, Poisson's ratio, and thickness of the film. The finite element method is used to calculate the surface displacement responses in a film-substrate system. Levenberg Marquardt algorithm is used as numerical optimization to speed up the training process for the ANN model. In this method, the materials parameters are not recovered from the dispersion curves but rather directly from the transient surface displacement. We have also found that this procedure is very efficient for determining the materials parameters of layered systems. (C) 2004 American Institute of Physics.