Personal Website of Asaad Abdollahzadeh

Aso

www.aso.im

My Blog

Papers Abstracts

Papers Abstracts

The Quantification of Uncertainties in Production Prediction Using Integrated Statistical and Neural Network Approaches: An Iranian Gas Field Case Study

Asaad Abdollahzadeh ; M. Hosseini; and Gh. Zargar

ABSTRACT

Uncertainty in production prediction has been due to numerous investigations. Geological and reservoir engineering data comprise a huge number of data entries to the simulation models. Thus, uncertainty of these data can largely contaminate the reliability of the simulation model. For this reason, it is worthy to present the desired quantity with a probability distribution instead of a single sharp value. For the case-study, numbers of parameters which are believed to contribute largely in the uncertainty of Field Gas Production Total are recognized. A sensitivity analysis was done to find the most significant ones. We have designed experiments of screening objective to recognize the main factors and the significant interactions of factors that we need to certainly include in the response function. Later, we designed experiments of response surface objective to model the response surface function of Field Gas Production Total. This has been done utilizing two methods, Response Surface Methodology and Artificial Neural Networks. The probability distribution of Field Gas Production Total was then plotted utilizing Monte Carlo simulation.

KEYWORDS

Reservoir, Simulation, Uncertainty, Gas, Sensitivity, Statistics, Experimental Design, Neural Networks

Accelerating the GMRES Iterative Linear Solver of an Oil Reservoir Simulator using the Multi-Processing Power of Compute Unified Device Architecture of Graphics Cards

Nima Ghaemian1, Asaad Abdollahzadeh, Zoltan Heinemann, Hossein Beyrami, Andreas Harrer, Mohsen Sharifi1, and Gabor Heinemann

ABSTRACT

The Generalized Minimal Residual Method (GMRES) is an iterative method for numerical solution of a system of linear equations. It approximates the solution by a vector in a Krylov subspace with minimal residual, using the Arnoldi iteration to find the vector. To gain higher performance, we have applied GMRES to the Solver of an in-house Reservoir Simulator Software that has been basically designed and operated as an ordinary CPU-based program. We have used the multi-processing capabilities of the Graphics Processor Unit (GPU) of a specific brand of a Graphics Card to our advantage, using Compute Unified Device Architecture (CUDA). As well as presenting our acceleration technique, we show the acceleration rate experimentally.

KEYWORDS

Reservoir Simulator, Iterative Numerical Method, GPU, CUDA, GMRES

The Quantification and Evaluation of Uncertainty in Reserves; Case Study in an Oil Field

Asaad Abdollahzadeh

ABSTRACT

In this paper after an introduction to the Uncertainty Studies, a methodology for estimating quality and quantity of the reservoir uncertainties is introduced in five stage: 1) choose or prepare best reservoir dynamic and static model 2) define uncertainty parameters 3) design the experiments 4) run designed experiments 5) analyze of the results and estimate the distribution of response parameter. In the case study, this integrated methodology, based on some statistical approaches such as Experimental Design, Response Surface Methodology and Monte Carlo Sampling, is applied to the reservoir model and the defined uncertainty parameters. Therefore, impact of each uncertainty on the cumulative oil production as response parameter is obtained and finally, uncertainty in the reserves is estimated based on the stochastic modeling of the uncertain parameters.

KEYWORDS

Reservoir Study, Uncertainty, Reserve