Modeling asphaltene precipitation in a compositional reservoir simulator using three-phase equilibrium

•Compositional reservoir simulator includes asphaltene precipitation and deposition model.•Fluid model uses a three-phase equilibrium with Peng-Robinson equation of state.•Phase behavior validated using published experimental data.•Dynamic evaluation of porosity and permeability changes from asphaltene precipitation.•Production impairment due to precipitation in the reservoir can be significant.

Predicting asphaltene precipitation and its deposition in the porous media is key in understanding costly productivity impairment caused by formation damage. This work presents the development and application of a compositional reservoir simulator including the effects of asphaltene precipitation on production. Precipitation from the liquid mixture occurs as its solubility is reduced either by changes in pressure (natural depletion), or by composition (mixing with another fluid, such as in gas injection). Our approach represents asphaltene as the dense liquid phase computed from a three-phase flash (vapor/liquid/liquid equilibrium VLLE) using the Peng Robinson equation of state.The multi-phase and multi-component reservoir simulator developed in this study uses a volume-based formulation, with pressures calculated implicitly and compositions explicitly (IMPESC approach). A cubic equation of state solves a three-phase flash to determine the number of phases co-existing in equilibrium. As asphaltene precipitates and deposits in the rock, the model dynamically calculates new porosities and permeabilities to represent the reduced pore space and effective flow path. We validated our model by matching experimental asphaltene precipitation data while predicting the expected phase behavior envelope and response to key thermodynamic variables (i.e. fluid composition, asphaltene molecular weight and its characterization in terms of pseudo critical properties).Previous modeling techniques were computationally inefficient, exhibit thermodynamic inconsistencies, and/or required special laboratory experiments to characterize the fluid. Our three-phase VLLE flash algorithm coupled with the reservoir model provides superior thermodynamic predictions compared to existing commercial techniques. This model offers the robustness and speed of a flash calculation while maintaining thermodynamic consistency, enabling efficient optimization of reservoir development strategies to mitigate the detrimental effects of asphaltene precipitation on productivity.