Implementation of In flow Control Devices in Gas Condensate Reservoirs using Reservoir Simulation

Abstract

Inflow Control Device (ICD) is a relatively new completion item. The objective of this device is to balance the inflow coming from the reservoir toward the wellbore by introducing an extra pressure drop.ICDs have already been implemented in many oil reservoirs all over the world. The main objective of this thesis is to investigate the possibility of using ICDs in gas condensate reservoirs. This workis without prior studies in literature. Synthetic reservoir data associated with economic calculations were used to carry out this investigation.

A single-well reservoir model combined with a r-z radial grid is developed. This model is based on two non-communicating layers. Different properties are assigned to these layers to generate a desired uneven inflow. This model is run many times with different properties each time (initial reservoirpressure in both layers, dew point pressure in both layer, layer permeability, layer thickness, etc.).This "reservoir mapping" is intended to cover a large spectrum of potential gas condensate reservoirs.ICD is modeled by a skin factor, generating an extra pressure drop, in these reservoirs simulated.

For each reservoir simulated, an optimization process is performed using Nelder Mead Simplex Reflection solver. The objective function for the optimization is Net Present Revenue of the project(also referred to as Total Discounted Value in this thesis). Net Present Revenue is derived from economic calculations using as input reservoir simulation results. The only optimization variable is skin value, representing ICD. For each reservoir simulated, a Net Present Revenue comparison is made between cases with no skin and cases with a given skin value (result of optimization process).

This thesis shows that among all cases simulated, only few of them show an increase in Net Present Revenue when using ICDs, and this increase is not significant. Some sensitivity analysis have been performed focusing mainly on two parameters: permeability and layer thickness. These analysis show a high dependency of optimization results on both intrinsic values of these parameters (permeabilityintrinsic value and layer thickness intrinsic value) and ratios of this parameters (permeability ratio and layer thickness ratio).

The results of this thesis, even if they may lack generality, raise interrogation regarding the relevance of using ICDs in gas condensate reservoirs. No definitive answer can be given yet. Better modeling and further investigations are required. Some possible continuation steps to this study are suggested at the end of this thesis.