Use of rescaled exponential models for boundary-dominated liquid-rich gas flow analysis under variable bottomhole pressure conditions

摘要

The analysis of production data from natural gas reservoirs can serve as one of the most powerful tools to estimate remaining reserves and provide a forecast of its future performance. While fundamentals of decline for dry gas reservoirs are well described in literature, those for liquid-rich gas reservoirs are yet to be well-understood. Any predictive model used to analyze these reservoirs must account for the inherent changes in reservoir fluid composition during their producing life due to condensate dropout within the reservoir, once reservoir pressure falls below dew point pressure. This study presents a mathematical model capable of predicting non-linear flow behavior in multiphase gas-condensate reservoirs using rescaled exponential models applicable to boundary-dominated flow regimes under variable bottomhole pressure conditions. We develop a set of analytical solutions for surface oil and total hydrocarbon flowrates, for wells producing under variable bottomhole pressure. We model hydrocarbon production from gas-condensate reservoirs by employing a material balance over produced condensate and total hydrocarbons. In this material balance approach, we use equivalent fluid molar densities in multiphase systems, resulting in analytical equations for the different flowing phases. The developed set of analytical solutions aims at providing an accurate estimate of reservoir behavior and available reserves, which can be used to inform critical economic decisions for further development of the reservoir. The proposed rescaled models minimize assumptions and stay true to the physics of multiphase flow. We demonstrate that the developed analytical model closely predicts the numerical simulation results for the hydrocarbon flowrates as well as the estimated reserves where a wide range of gas-condensate reservoirs, from lean to liquid-rich, is considered. (C) 2017 Elsevier B.V. All rights reserved.