New Technique to Estimate Temperature Distribution During Thermal EOR Operations

Amjed Hassan, Salaheldin Mahmoud Elkatatny, Mohamed Mahmoud, Abdulazeez Abdulraheem, Enamul Hossain

2018

Closed access Metadata (CSL-JSON)
DOI 10.2118/192350-ms
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

Abstract Thermal operations are the most effective enhanced oil recovery (EOR) techniques used to increase hydrocarbon production, especially in heavy oil reservoirs. Thermal EOR involves injecting of steam or hot fluids into underground reservoirs to alter the physical properties, specifically, the fluid viscosity and effective mobility. Numbers of mathematical (numerical or analytical) models were developed to estimate the temperature variations during such operations, however, those models assume constant fluid velocity throughout the reservoir, then, severe estimations errors could be generated. The objective of this paper is to develop a new approach for determining the temperature distribution during thermal-EOR processes, and the heat propagations with time and distance from the wellbore with acceptable tolerance. The main aim of this work is to develop a reliable model to predict the temperature distributions in porous rocks during thermal EOR operations. Artificial intelligence (AI) methods were utilized to compute the temperature profiles, those models would minimize the complexity and uncertainties of numerical approaches. The impact of formation permeability, injection time and distance from wellbore were considered to develop effective models. To ensure a high level of model reliability, more than 220 data set was used for training and testing the proposed models. Temperature distribution was determined using the neural network, fuzzy logic system, and generalized intelligent networks. Different model's parameters were used to optimize the intelligent networks, average absolute error and correlation coefficient were utilized to measure the model performance. ANN model showed the best prediction performance, an average absolute error of 6.2 % and a correlation coefficient of 0.98 was obtained using unseen data set.