Volume 19, Issue 1 (Spring 2025)
2025, 19(1): 136-158 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Azari T. A novel approach to determine hydraulic parameters of double porosity aquifers based on MLP neural network. Journal of Engineering Geology 2025; 19 (1) :136-158
URL: http://jeg.khu.ac.ir/article-1-3163-en.html
URL: http://jeg.khu.ac.ir/article-1-3163-en.html
Kharazmi university , t.azari@khu.ac.ir
Abstract: (447 Views)
Accurately determining hydraulic parameter values is the first step in sustainably developing an aquifer. Since Theis (1935) introduced the type curve matching technique (TCMT), it has been used to estimate aquifer parameters from pumping test data. However, the TCMT is subject to graphical error. To eliminate this error, a multi-layer perceptron (MLP) artificial neural network (ANN) was developed as an alternative to the conventional TCMT. This MLP ANN models the Bourdet-Gringaten well function to determine fractured double porosity aquifer parameters. The MLP model was developed using a four-step protocol and trained using the backpropagation method and the Levenberg-Marquardt optimization algorithm for the well function of double-porosity aquifers. Through a trial-and-error procedure and by applying principal component analysis (PCA) to the training input data, the optimal network structure with the topology [3×6×3] is determined. We evaluated the validity of the developed network with synthetic and real field data. The network receives pumping test data and provides the user with aquifer parameter values. This network provides an automatic, fast procedure for determining double-porosity aquifer parameters, eliminating the graphical errors inherent in the conventional TCMT.
Keywords: Aquifer parameters, Double porosity, Well function, Multi Layaer Perceptron (MLP), Principal Component Analysis (PCA).
Type of Study: Original Research |
Subject:
Hydrogeology
Received: 2025/04/9 | Accepted: 2025/06/3 | Published: 2025/06/20
Received: 2025/04/9 | Accepted: 2025/06/3 | Published: 2025/06/20
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
