Automated Reservoir Model Selection in Well Test Interpretation

Abstract

This paper was selected for presentation by an SPE Program Committee following review of information contained in an abstract submitted by the authors(s). Contents of the paper, as presented, have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Papers presented at SPE meetings are subject to publication review by Editorial Committees of the Society of Petroleum Engineers. Permission to copy is restricted to an abstract of not more than 300 words. Illustrations may not be copied. The abstract should contain conspicuous acknowledgement of where and by whom the paper was presented. Write Librarian, SPE,

Using Particle Swarm Optimization (PSO) Algorithm in Nonlinear Regression Well Test Analysis and Its Comparison with Levenberg-Marquardt Algorithm

“…Although AI (Artificial Intelligence) techniques have been used in many branches of the petroleum engineering including well log interpretation (Mohaghegh et al, 1994;Al-Bulushi et al, 2012), seismic interpretation (Boerner et al, 2003), well test model recognition (Allian and Houze, 1995;Al-Kaabi and Lee, 1990;Ershaghi et al, 1993;Athichanagorn and Horne, 1995;Kumoluyi et al, 1995;Alajmi and Ertekin, 2007;Kharrat and Razavi, 2008;Adibifard et al, 2014), and fluid properties estimation (Gharbi et al, 1999;Elsharkwy and Gharbi, 2001;Osman et al, 2001;Huang et al, 2003), there are a few studies regarding the applications of AI in well test nonlinear regression analysis. Güyagüler et al (2001) used GA (Genetic Algorithm) to recognize reservoir model using nonlinear regression analysis technique. They used different individuals in the population of the GA as different well test models and allowed the evolution process to occur only within the same model.…”

“…They tested their proposed technique over pressure and pressure derivative data in a dual porosity reservoir with constant pressure boundary and found out that the novel technique is able to accurately predict the reservoir model compared with classical nonlinear regression methods. They also proposed that results achieved by the GA can be used as an initial point used by classical nonlinear regression well test analysis techniques (Güyagüler et al, 2001).…”

An Improved Model to Predict Reservoir Characteristics During Underbalanced Drilling

“…It has also been mentioned repeatedly in the literature that sometimes simple GA is plagued by symptoms of premature convergence without guarantee of optimality because of biased selection of an incomplete search space [20][21][22] . This could originate from inappropriate choices of variables resulting in solution convergence in bad regions.…”

“…Recently, both simple and hybrid (polytope and Tabu search, Fang's algorithm etc.) GA is gaining in popularity in the petroleum engineering literature [20][21][22] . Since GA starts with a population of decision variables instead of a single estimate, the entire search domain is covered from the outset.…”

An Efficient Approach for Automatic Complex Fractured Networks Parameter Inversion Based on Surrogate Model and Deep Reinforcement Learning

“…At present, a variety of deep learning algorithms are being integrated with energy development, which has led to a significant increase in computational efficiency and a prominent reduction in solution design cycles (Y. Li et al., 2019; Sun & Zhang, 2020). At the same time, many automatic interpretation methods have been proposed, including gradient‐based optimization algorithms (Dastan, 2010; Dastan & Horne, 2011; Nanba & Horne, 1992), gradient‐free optimization algorithms (Awotunde, 2015; Gomez et al., 2014; Guyaguler et al., 2001), and Ensemble Kalman filter method (Xue et al., 2022). Besides, C. Wang et al.…”