Please use this identifier to cite or link to this item:
|Title:||AMELIORATION DE L’ESTIMATION DE LA TRANSMISSIVITE DANS UNE NAPPE ALLUVIALE A L’AIDE DES RESEAUX DE NEURONES ARTIFICIELS|
Artificial Neural Networks
|Abstract:||Transmissivity is often estimated using specific capacity data when the drawdown is stabilized early or standard pumping test data are not available, as in this study. Previous researchers studied the relationship between transmissivity and specific capacity in the aquifer system of alluvial basin on Lorraine, France, using the Cooper–Jacob, Boulton and Hantush equations. The linear relationship between transmissivity and specific capacity on a log–log scale for alluvial aquifer on Moselle basin is remarkably strong, with a correlation coefficient of 0,85. In this study, the ANN multilayer perceptron was employed to model transmissivity using data derived from pumping test conducted in Moselle Valley, France. Training parameters such as the stopping criteria and the type of transfer function affected the efficiency and the generalization capability of the ANN. The ANN resulted in a satisfactory network with an average R of 0.97.|
|Appears in Collections:||CS N 13|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.