Mechanistic Development of Decline Curves for Gas-Condensate Reservoirs
Abstract
This study incorporates material balance, complex fluid behavior, reservoir characteristics, and well operating conditions into a mechanistic model to forecast production from gas-condensate reservoirs. The model can represent the behavior of the fluids and production from a gas-condensate reservoir, with considerations of condensate banking phenomena and adsorption, while running an order of magnitude faster than a commercial compositional reservoir simulator.
The constant volume depletion (CVD) process serves as a basis to estimate production from gas-condensate reservoirs by coupling a robust phase-behavior calculation using a thermodynamic cubic equation of state (EOS) and radial flow equations. The mechanistic method developed in this project establishes the relationship between production and time.
The CVD test is coded as a multicompositional molar-accounting program that simulates gas-condensate phase behavior. As near-wellbore and reservoir pressures drop below saturation pressure, in gas-condensate reservoir condensate banking occurs, hindering flow. The two-tank CVD approach developed in this work models the near-wellbore liquid saturation buildup caused by this phenomenon.
The modified CVD provides a material balance base for the method. The backbone of this model is the vapor/liquid equilibrium (VLE) calculation with the Peng-Robinson equation of state (EOS). Produced fluid volumes are calculated with each CVD pressure decrease. Using this information coupled with the flow equation, the model can forecast production behavior. Cumulative oil and gas production comparisons between the results from this model and those from a compositional reservoir simulator show an average absolute error of around 4%. Oil and gas production rate forecasts from this model have an average absolute error of 15% from the compositional reservoir simulator forecasts. Comparison of results from this model with those from the commercial simulator shows the importance iii of considering condensate accumulation effects on productivity and accurately predicting flow from the gas-condensate reservoir.
In the model, hydraulic fracture well completions are represented with an equivalent skin factor. In this case, the effect of condensate banking is less significance, and the new model produced results comparable to production forecasted by the commercial simulator.
This model is unique because it uses the CVD process as a material balance basis to predict production. This process provides the physical basis for reservoir engineers to predict production behavior of a gas-condensate reservoir, instead of the empirically based predictions from conventional decline analysis. This simple tool can also be incorporated to the evaluating prospects before further in-depth reservoir studies are conducted.
Citation
Ariwibowo, Riyan (2017). Mechanistic Development of Decline Curves for Gas-Condensate Reservoirs. Master's thesis, Texas A & M University. Available electronically from https : / /hdl .handle .net /1969 .1 /161397.