الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 182 |  |  |
| | | from | 182 | | |
Abstract
Groundwater considers the main source of water in the arid and semi-arid regions of Egypt; therefore, it is necessary to keep this source active and available by optimum usage and good management. Groundwater resources management is a highly complex issue covering a wide spectrum of activities in the field of assessment, planning, designing, operation and maintenance. Efficient management of groundwater system depends upon many activities such as budget, time, type of modeling, and availability data resources. Groundwater modeling primarily depends on adequate knowledge of system hydraulic parameters such as transmissivity, hydraulic conductivity, storativity, specific yield and aquifer recharge. These parameters can be determined by experimental and field studies but the entire process is time consuming, very complex and financially demanding. Parameters estimation of aquifer systems is a dynamic process, due to the fact that the state of any hydrological system keeps changing with time. Moreover, scientific techniques involved to evaluate such systems are evolving continuously. Research Scope: - To implement objectives of the study, the following strategy will be as follow: Collecting data and information which concern the study. Collect good recourses of books, theses and researches concern with the study area, MODFLOW and ANNs. Study and analysis the subject of modeling using both of MODFLOW program and ANNs technique then arrange data that will be used in modeling. MODFLOW program Version 3.0 is used to draw hydraulic contour map of the considered area using finite difference grid. Surfer version 11 is used to check contouring of some hydraulic parameters. Global Mapper version 12 is to plot geological sections in the study area. Construction of ANNs model then built in MATLAB program version 7. The construction means that choosing the best topology, No. of hidden layers and neuron, training, validation and testing ANNs. Using ANN model, to estimate transmissivity and storativity distribution in study area. MODFLOW program Version 3.0 is used to predict hydraulic head for different situations and periods using the ANN output (transmissivity & storativity). Research Objectives: - The major objectives of this study are applied a MODFLOW model to predict hydraulic head of an aquifer for different scenarios using the outputs of ANN model and developing a model using ANNs for identificate an aquifer parameters such as transmissivity and storativity distribution. The obtained transmissivity and storativity are used to obtain the hydraulic heads of the same aquifer for different scenarios. Hydraulic heads and other parameters are necessary for the management and optimum usage. Objectives may be summarized as follows: Applied a model using (MODFLOW) to estimate initial hydraulic head. Test the model (calibration and validation,). Construct ANN model to estimate transmissivity and storativity distribution. Train ANN model then validate and test it. Using ANN model to obtain the distribution of transmissivity and storativity. Applied a MODFLOW model to predict hydraulic head of an aquifer for different scenarios using the outputs of ANN model. Steps of Study: - In this study, MODFLOW program is used to develop a groundwater model to simulate the behavior of flow system under different conditions. MODFLOW is used twice, firstly to get the values of hydraulic head in each cell of the study area and lastly to predict the changes of an aquifer under stress for four expected scenarios for the study area. An Artificial Neural Networks (ANNs) technique used to develop a model to predict and identificate the hydraulic parameters of the study area. A model based on artificial neural networks, was examined to get optimum topology that can applied for prediction transmissivity and storativity. Summary of Study: - This study is based on coupling of Finite Difference Method (FDM) – (MODFLOW), which serves as forward modeling (FDM) and inverse modeling (ANNs) models. An inverse technique using ANN is considered for estimating parameters of groundwater system.