A Novel Enhanced-Oil-Recovery Screening Approach Based on Bayesian Clustering and Principal-Component Analysis

Abstract

We present and test a new screening methodology to discriminate amongst alternative and competing Enhanced Oil Recovery (EOR) techniques to be considered for a given reservoir. Our work is motivated by the observation that, even if a considerable variety of EOR techniques have been successfully applied to extend oilfield production and lifetime, an EOR project requires extensive laboratory and pilot tests prior to field-wide implementation and preliminary assessment of EOR potential in a reservoir is critical in the decision-making process. Since similar EOR techniques may be successful in fields sharing some global features, as basic discrimination criteria we consider fluid (density and viscosity) and reservoir formation (porosity, permeability, depth and temperature) properties. Our approach is observation-driven and grounded on an exhaustive data-base which we compile upon considering worldwide EOR field experiences. A preliminary reduction of the dimensionality of the parameter space over which EOR projects are classified is accomplished through Principal Component Analysis (PCA). A screening of target analogs is then obtained by classification of documented EOR projects through a Bayesian clustering algorithm. Considering the cluster which comprises the EOR field under evaluation, an intercluster refinement is then accomplished by ordering cluster components on the basis of a weighted Euclidean distance from the target field in the (multidimensional) parameter space. Distinctive features of our methodology are that (a) all screening analyses are performed on the database projected onto the space of principal components, and (b) the fraction of variance associated with each principal component is taken as weight of the Euclidean distance we determine. As a test bed, we apply our approach on three fields operated by eni. These include light, medium and heavy-oil reservoirs, where Gas, Chemical and Thermal EOR projects have been respectively proposed. Our results are (a) conducive to the compilation of a broad and extensively usable data-base of EOR settings and (b) consistent with the field observations related to the three tested and already planned/implemented EOR methodologies, thus demonstrating the effectiveness of our approach.

Efficient Assessment of Reservoir Uncertainty Using Distance-Based Clustering: A Review

“…Other distance calculation formulas have been employed in previous studies [14,36,39,42,52]. For example, Patel et al [16] employed the Kantorovich distance to compare the probability distributions of the reservoir models.…”

“…Compared with MDS, which is only used to reduce the data dimension based on the distance information, PCA can analyze data characteristics and reduce their dimension. Therefore, PCA is extensively employed in various research areas, such as face cognition, seismic interpretation, and reservoir engineering, which require very complex calculations in high dimensions [5,24,25,42,49,50,52,54,55].…”

“…The criterion of EOR screening is usually determined by previous EOR projects. Some researchers have classified the conditions of EOR projects to develop a decision-making system for a proper EOR method [47,52]. In [47], after 151 EOR projects were reviewed, the following nine variables were applied for a feature vector: four rock properties (permeability, porosity, reservoir depth, and thickness) and five fluid properties (oil saturation, gravity, temperature, viscosity, and pressure).…”

“…Siena et al [52] also considered various reservoir properties for discrimination criteria: density, viscosity, porosity, permeability, depth, and temperature. They reduced the dimensionality of the parameter space by PCA and measured the similarity between EOR projects.…”

Characterization of three-dimensional channel reservoirs using ensemble Kalman filter assisted by principal component analysis

“…PCA is a useful mathematical tool to manage high-dimensional data by extracting primary parameters of the data (Lim et al 2015;Siena et al 2016;Jung et al 2018). Some researchers utilized PCA with other schemes to apply to multipoint geostatistics (Vo and Durlofsky 2014, 2016Chen et al 2016). We use PCA to extract the primary characteristics of channel distribution and select good models efficiently.…”