Selection and Evaluation EOR Method Using Artificial Intelligence
E.M. Shokir, H.M. Goda, Mohamed Helmy Sayyouh, Kh. Fattah · 2002
Selection and Evaluation EOR Method Using Artificial Intelligence E.M. El-M. Shokir; E.M. El-M. Shokir King Saud University Search for other works by this author on: This Site Google Scholar H.M. Goda; H.M. Goda Cairo University Search for other works by this author on: This Site Google Scholar M.H. Sayyouh; M.H. Sayyouh Cairo University Search for other works by this author on: This Site Google Scholar Kh. A. Fattah Kh. A. Fattah King Saud University Search for other works by this author on: This Site Google Scholar Paper presented at the Annual International Conference and Exhibition, Abuja, Nigeria, August 2002. Paper Number: SPE-79163-MS https://doi.org/10.2118/79163-MS Published: August 05 2002 Cite View This Citation Add to Citation Manager Share Icon Share Twitter LinkedIn Get Permissions Search Site Citation Shokir, E.M. El-M., Goda, H.M., Sayyouh, M.H., and Kh. A. Fattah. "Selection and Evaluation EOR Method Using Artificial Intelligence." Paper presented at the Annual International Conference and Exhibition, Abuja, Nigeria, August 2002. doi: https://doi.org/10.2118/79163-MS Download citation file: Ris (Zotero) Reference Manager EasyBib Bookends Mendeley Papers EndNote RefWorks BibTex Search Dropdown Menu toolbar search search input Search input auto suggest filter your search All ContentAll ProceedingsSociety of Petroleum Engineers (SPE)SPE Nigeria Annual International Conference and Exhibition Search Advanced Search AbstractRecently, the theory of Artificial Neural Network has found a worldwideapproval from those who are concerned with the research in the field ofpetroleum and natural gas engineering. This may be due to the neural networkpotentialities in solving problems to which the network is designed, dependingon the gained experiences of some similar problems joined with their solutions.This theory has been found to give high possibilities with amazing results asapplied and tested in petroleum engineering.A model composed of two neural networks was designed to work in series inorder to perform the task of EOR project selection through the first network(Technical network), and then evaluate the selected project economically usingthe second network (Economics network). As the model, with the two networks, performs its task successfully, it is then tested with data seeded with certainlevel of error affecting the parameters given to the model in the test featuresin order to determine the power of the model and its ability to overcome whatwe call noisy data.IntroductionTypically, the use of the natural stored reservoir energy may not producenot more than 50% of the original oil in place during the usage of the primarymechanisms. As these energies are exhausted, programs of water flooding or gasinjection may then be applied to the reservoir during the secondary recoveryphase in order to maintain the energy of the reservoir.The term enhanced oil recovery (EOR) covers all reservoir treatmentprocesses designed to recover the stranded fraction of oil in place. Namely thefraction that has not been able to move under the natural effect of thepressure gradient and expansion of the dissolved gas, pressure maintenance bywater or gas injection (1).These processes, which may be applied to improve the petroleum recovery fromexisting reservoir, have gained more attention as a result of increasing costsof exploring new oil fields and reducing opportunities for discovery of new oilreservoir with high reserve and good quality (1).Since the production of oil by any means of EOR process is a ratherdifficult, risky, and need huge capital cost, the proper selection of the EORmethod for a certain reservoir is important to attaining a successful andprofitable project. The most difficult problem that faces reservoir engineersand experts is how to select the most appropriate method to enhance the oilproduction based on technical and economical factors (2). To achieve thismission, many reservoir factors must be studied carefully such as reservoirdepth, reservoir area, reservoir temperature, porosity, permeability, oilgravity, and oil viscosity, and also how these factors may affect each other(3, 4).Shindy et. al. (5), developed a knowledge-based expert system for theselection of the EOR method in oil reservoirs. An analytical method based onstatistical evaluation was used to produce the rules used in the expert systemformulation. They solved the problem of technical selection of EOR techniquesfor different application. No solution for the economical selection of EORproject was presented in their paper.The choice of any improved oil recovery method is based on technical andeconomical criteria. The main problem faced by the petroleum engineer is toidentify technically, and evaluate economically the enhanced recovery processesapplicable for the oil reservoir.In our study, both the technical, and economical EOR selection problem hasbeen solved using the neural network approach. This approach is as efficientand flexible as the previous developed expert system (5). However, theeconomical analysis for the EOR selection, which was not solved by the previousexpert system, has been solved in the present work. Keywords: efficiency, shokir, spe 79163, selection, waterflooding, neuron, society of petroleum engineers, cairo university, evaluation eor method, input layer Subjects: Improved and Enhanced Recovery, Information Management and Systems, Waterflooding, Neural networks This content is only available via PDF. 2002. Society of Petroleum Engineers You can access this article if you purchase or spend a download.