The two most popular methods for extracting gas condensate from reservoirs are the gas cycle and natural depletion. The first method is the quickest and least expensive, but it also produces a significant amount of stagnant liquid drop-out that is lost during the recovery process. Gas cycling, which maintains the same pressure while re-evaporating the condensed oil, boosts liquid recovery. Yet, reservoir heterogeneities and money problems may make implementation more challenging. Pressure maintenance has been opposed in favor of water injection. In order to assess the latter method's viability, this thesis compares its efficacy to that of the gas cycle and natural depletion. An Equation of State (EOS)-based multi-phase, multi-component compositional technique was used to accomplish this. It was important to make changes to lower the cost of simulation and raise the physical realism of the projections before executing full scale recovery calculations. In order to reduce the amount of elements that describe hydrocarbon fluids, regression-based EOS software was created. The findings of the present study imply that even three components might be adequate to preserve the thermodynamic consistency of the predictions. By solving the pressure equation with a powerful algorithm, the cost of simulation might be decreased. In the prior investigation, several sparsity conserving techniques were used and conventional direct approaches were investigated. To deal with the drainage and imbibition cycles that happen during water injection, a relative permeability hysteresis methodology was also integrated into the procedure. According to the findings, recovery forecasts may be off in the absence of the hysteresis effect. To determine how sensitive the process of water injection was to different fluid and rock parameters, the approach was then applied to a variety of representative reservoir simulations. The effectiveness of the suggested approach was also compared to that of more widely used ones.
Keywords: Cycling, gas, condensate, recovery, modeling