Soft Computing for Intelligent Reservoir Characterization

Abstract

This paper presents an overview of soft computing techniques for reservoir characterization. The key techniques include neurocomputing, fuzzy logic and evolutionary computing. A number of documented studies show that these intelligent techniques are good candidates for seismic data processing and characterization, well logging, reservoir mapping and engineering. Future research should focus on the integration of data and disciplinary knowledge for improving our understanding of reservoir data and reducing our prediction uncertainty.

Predicting injection profiles using ANFIS

“…The problems are: (1) there are no standard methods for transforming human knowledge or experience into the rule base; and (2) it is required to further tune the membership functions (MF) to minimize the output errors and to maximize the performance, as stated in [9]. There are many methods [27,20,14] that can be applied to identify the MF and FIS. In this paper, two commonly used methods are applied for FIS identification and refinement.…”

“…Soft computing methods have been widely applied in many areas in the petroleum industry, such as reservoir description [27], well logging interpretation [16], production prediction [29] and treatment optimization [17]. In this paper, two neuro-fuzzy systems, ANFIS-GRID and ANFIS-SUB, are employed to model the relationships of injection profiles and their influential parameters.…”

Modeling the permeability of carbonate reservoir using type-2 fuzzy logic systems

“…The use of functional network was also reported in [7] for predicting permeability from well logs, where functional equations were made use of to get better predictions. Recently, application of fuzzy logic in petroleum engineering field has received considerable attention and has been successfully applied to address problems on various oil and gas reservoirs such as identification of lithofacies and prediction of permeability using wire line logs [26][27][28]. For predicting these properties generally Gaussian membership and fuzzy clustering algorithm are applied and better estimates have been reported compared to that of conventional techniques.…”

“…In the case of regression, a margin of tolerance 2 is set in approximation to the SVM which would have already being inferred from the problem. Mathematically, since the main idea is to optimize the margin then the quadratic optimization problem becomes min T wð1=2W " T:ÞW (28) s:t:…”

History matching of petroleum reservoirs employing adaptive genetic algorithm

“…This optimization technique usually involves minimizing the objective function that describes the mismatch between the available field historic data and reservoir simulator response. GA as a stochastic optimization tool outperforms other gradient based methods (steepest descent, GaussNewton method, conjugate gradient etc.,) toward reaching a global optimal solution escaping the local optima (Gill 1981;Ouenes 1992;Tamhane et al 2000;Gomez et al 2001;Romero and Carter 2001;Schulze-Riegert et al 2001;Choudhary et al 2007). …”