论文部分内容阅读
The multi-layers feedforward neural network is used for inversion of material constants of flu-id-saturated porous media.The direct analysis of fluid-saturated porous media is carried out with the bound-ary element method.The dynamic displacement responses obtained from direct analysis for prescribed materi-al parameters constitute the sample sets training neural network.By virtue of the effective L-M training algo-rithm and the Tikhunov regularization method as well as the GCV method for an appropriate selection of regu-larization parameter,the inverse mapping from dynamic displacement responses to material constants is per-formed.Numerical examples demonstrate the validity of the neural network method.
The multi-layers feedforward neural network is used for inversion of material constants of flu-id-saturated porous media. The direct analysis of fluid-saturated porous media is carried out with the bound-ary element method. analysis for prescribed materi-al parameters constitute the sample sets training neural network. By virtue of the effective LM training algo-rithm and the Tikhunov regularization method as well as the GCV method for an appropriate selection of regu-larization parameter, the inverse mapping from dynamic displacement responses to material constants is per-formed. Numerical examples demonstrate the validity of the neural network method.