| description abstract | Abstract. Production forecasting in gas condensate reservoirs is challenging due to the complex behavior of its flow near the wellbore and the impact of condensate blockage. This article reports a numerical compositional simulation study that incorporates three crucial near-wellbore characteristics: three-region pseudo-pressure (TRPP), velocity-dependent relative permeability (VDRP), and non-Darcy flow (NDF) to address this challenge. Our approach involves sequentially analyzing each characteristic, forming three groups of analysis: individual characteristics, pairwise combinations, and the combination of all three characteristics. This decomposition quantitatively assesses the physical contribution of each characteristic to prediction outcomes. The results demonstrate that combining all three characteristics yields the most reliable predictions for well X in gas condensate reservoir Y, with the combined approach aligning closely with observed production data. Neglecting any of these characteristics results in noticeable prediction errors. The TRPP model provided valuable insights into reservoir region detection, enabling precise characterization of each region. For the VDRP model, we implemented a tree-structured parzen estimator (TPE) optimization algorithm, achieving a more accurate correlation between calculated and experimental relative permeabilities than previous studies. The contribution of the NDF characteristic is clarified by demonstrating its influence on flow parameters, specifically the pressure gradient and gas relative permeability, with increasing flow velocity in the near-wellbore region. Additionally, this study offers detailed guidelines for incorporating these three characteristics into simulations, including a comprehensive mathematical formulation of the TRPP model within a compositional model, addressing a gap in prior studies. | |
