Back to Search
Start Over
Comparison of Well Test Data with Reservoir Data Obtained by Well-Log, Using Well Test Software’s and Presenting an Accurate Model According to Both Analytical Solution and Artificial Neural Network for Horizontal Wells in Naturally Fractured Reservoirs.
- Source :
-
Petroleum & Coal . 2022, Vol. 64 Issue 2, p25-43. 19p. - Publication Year :
- 2022
-
Abstract
- Well test analysis is a method for evaluating the average properties of the reservoir by characterizing the ability of the fluid to flow through the reservoir and to the well. Well test output parameters that descript reservoir are permeability, reservoir heterogeneities, boundaries and pressure; also, parameters that descript well are skin factor, productivity index and well geometries. The suggested models for prediction of pressure drop in vertical well have obtained from the solution of diffusivity equation for radial and elliptical regime. The purpose of this project is to generate a model for the horizontal well in fracture reservoir and estimate a parameter for this type of well by ANN method and compare the results with well test software. In this study, pressure data versus time was obtained for horizontal wells in naturally fractured reservoirs by solving diffusivity Equation using Stehfest algorithm, and for each set of data, a polynomial was developed for pressure derivative data by applying Chebyshev polynomials method. The polynomial coefficients along with reservoir and horizontal well data were fed into Artificial Neural Network (ANN) as input data and as a result, a model was presented for the horizontal well in naturally-fractured reservoir. In addition, output of the model was compared to the well log and well test software’s data and it was found that the presented model has better and higher performance than well test software’s. [ABSTRACT FROM AUTHOR]
Details
- Language :
- English
- ISSN :
- 13353055
- Volume :
- 64
- Issue :
- 2
- Database :
- Academic Search Index
- Journal :
- Petroleum & Coal
- Publication Type :
- Academic Journal
- Accession number :
- 158553171