Development of New PVT Correlations for Reservoir Fluids From Unconventional Reservoirs
Abstract
The estimation of Pressure-Volume-Temperature (or PVT) properties is an extremely important aspect of oil and gas exploration and production as the parameters derived from these functions are used for reserve estimations, reservoir modeling, production, and pressure analysis, and for predicting well production performance. Reservoir fluids from unconventional reservoirs tend to be significantly "lighter" (i.e., contain less heavy hydrocarbon molecules) compared to the so-called "black oil" reservoir fluids, and as such, generally require specialized correlations.
The primary focus of this work is to estimate the "saturation pressure" and "oil formation volume factor" of a relatively large sampling of reservoir fluids from unconventional (shale) reservoirs where these fluids tend to be near-critical volatile oils or retrograde gas condensate fluids. In this work, we utilize data from 138 PVT lab reports donated by industry collaborators (on the condition that the reports and data remain anonymous/confidential). These data were obtained from various unconventional reservoirs in the United States and Latin America (where again, the origin of the data is to remain confidential).
These data are used as a test case against the most common saturation pressure correlations available in the industry since the 1940s. It is important to recognize that this is a common practice, i.e., to test a "standard' model against a new data set. However, virtually all of the historical saturation pressure correlations were developed for "black oil" reservoir fluids. As such, we test each historical model with and without regression (MS Excel and Python algorithms are used to perform the regressions).
To assess a "best fit" in a statistical sense, we also utilize so-called "non-parametric" correlation methods that provide a multivariate optimization without using a specific model. These methods provide guidance on the best correlation of a given multivariate data set, but these methods cannot be used as a predictor (at least not directly). Lastly, in this work, we have developed a family of exponential-polynomial and rational polynomial functions to relate the saturation pressure and oil formation volume factor with reservoir temperature, stock tank oil gravity, separator gas gravity, and solution gas-oil-ratio.
Our proposed correlations have yielded excellent functional and statistical performances for our specific database, and we believe these correlations provide a new methodology for representing saturation pressure and oil formation volume factor based on these inputs. We also note that we have used certain variations of our proposed model to predict oil viscosity, but this was more of a test-of-concept as we are not confident that the volume of oil viscosity data in our database is sufficient to provide unique and robust correlations.
Subject
PVTCorrelation
Saturation Pressure
Formation Volume Factor
Unconventional Reservoir
light fluids
Hydrocarbons
Citation
Mejia Martinez, Nelson David (2022). Development of New PVT Correlations for Reservoir Fluids From Unconventional Reservoirs. Master's thesis, Texas A&M University. Available electronically from https : / /hdl .handle .net /1969 .1 /198154.