Modelling of Gas-condensate flow around horizontal and deviated wells and cleanup efficiency of hydraulically fractured wells.
MetadataShow full item record
Gas condensate reservoirs, when the pressure falls below dew point, are characterised by the appearance of condensate bank and exhibiting a complex phase and flow behaviour around the wellbore. The unique dependency of the gas and condensate relative permeability (kr) on the velocity and interfacial tension (IFT) complicates the well productivity calculations both in field simulation models and in simple engineering calculations, especially for complex well geometries such as horizontal or deviated or hydraulically fractured wells. The current research work has two parts. The first part is devoted to study the flow behaviour around horizontal wells (HWs) and deviated wells (DWs) in gas condensate reservoirs. Here, several in-house simulators have been developed for single-phase and two-phase gas condensate flows. The two phase in-house simulators accounts for the phase change and the dependency of relative permeability to interfacial tension and velocity, due to coupling (increase in kr by an increase in velocity or decrease in IFT) and inertia (a decrease in kr by an increase in velocity). The integrity of the in-house simulators has been verified by comparing some of its results with those obtained using the fine grid option of the ECLIPSE300 commercial reservoir simulator under the same flow conditions. Using the 3-D in-house simulator a large data bank has been generated covering a wide range of variations of pertinent geometrical and flow parameters. Then a general approach is proposed for estimation of an effective wellbore radius of an equivalent open-hole (EOH) radial 1-D system replicating flow around the 3-D HW system. The results of the proposed formulation, which benefits from suitable dimensionless numbers, has been tested against the simulator results not used in its development confirming the integrity of the approach. The proposed formulation, which is simple and easy to use, correctly converts to that suitable for single-phase non-Darcy (inertial) flow systems when total gas fractional flow (GTR) is unity. An extensive sensitivity study has also been conduct to highlight the limitations of current geometric skin formulations widely used in the petroleum industry for HW productivity calculations. The in-house improved geometric skin formulation is more efficient especially for anisotropy, partial penetration and location of HW in the vertical direction. The same exercises have been performed to study the flow behaviour around deviated wells. That is, the corresponding proposed mechanical and flow skin factors ii i are converted into an effective wellbore radius, before being applied in the pseudo-pressure calculation of the equivalent open hole system. Here due to the similarity of flow around HWs and DWs a simple relationship is proposed between the corresponding skin factors of these two well geometries. Therefore, in the proposed general method for modelling of the two-phase flow of gas and condensate around a DW, effective wellbore radius estimated for the HW with the same well length is converted to skin and then included in the proposed formulation before being converted to the effective wellbore radius of the equivalent open hole model replicating flow around 3-D flow geometry. Hydraulic fracturing is one of the most important stimulation techniques especially for tight gas reservoirs. The second part of this research work is devoted to conduct a parametric study to evaluate the impact of the pertinent parameters on the cleanup efficiency, as one of main reasons for poor performance of hydraulic fracturing operation, of gas and gas condensate reservoirs. This study has two parts. In the first part, a comprehensive sensitivity study conducted to evaluate the impact of pertinent parameters on the cleanup efficiency of a hydraulically (gas or gas-condensate) fractured well. Here the key parameters which have significant impact on the gas production loss (GPL) are identified. A new method is proposed to simulate a more realistic fracture fluid (FF) invasion into matrix and fracture, which proves to be one of the main reasons of the contradictory results found in the literature. However since none of such studies have embarked on a much needed extensive investigation of variation of all pertinent parameters, the second part of study is concentrated on a much more expanded study following statistical approaches. Here based on the results of the first part, key parameters have been identified. A 2-level full factorial statistical experimental design method has been used to sample a reasonably wide range of variation of pertinent parameters covering many practical cases for a total of 16 parameters. Since over 130000 simulation runs have been required, to cover the range of variation of all parameters the simulation process has been simplified and a computer code, which automatically links different stages of these simulations, has been developed. The analysis of the simulation runs using two response surface models (with and without interaction parameters) demonstrates the relative importance of the pertinent parameters after different periods.