الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 118 |  |  |
| | | from | 118 | | |
Abstract
In this study the Artificial Neural Network (ANN) model was developed to propose a new approach to predict the hydraulic fracture geometry: fracture height, fracture half-length and dimensionless fracture conductivity. The workflow begins with an integrated ANN model, then sets of variable fracture parameters, formation rock properties and real pressure transient analysis (PTA) data were utilized for training and testing the ANN based on the most appropriate activation function, the number of hidden layers and the number of neurons. The available data for this study was acquired from different fields located in the Egyptian western desert and west of Nile concessions, which covers different reservoirs like Safa, Bahariya, Abu-Roash and Alam ElBueib formations and different hydraulic fracturing techniques like conventional fracturing, channel fracturing and artificial barrier hydraulic fracturing. The raw data for this study consists of 49 input parameter across 59 well divided as 70% for training, 15% for validation and 15% for testing. The proposed ANN model result was promising as compared with other commercial fracture simulators. The cross plot of the actual fracture geometry parameters versus the predicted ANN outputs showed a good match with the correlation coefficient (R) equal to 0.93. Then the relative importance of ANN input parameter on the fracture geometry optimization was employed by Garson method. The result of this study shows the potential of the approach developed based on the ANN model for predicting the fracture geometry.