CO2 Foam Flooding with Integrated EOR and Storage
Research output: Thesis › Master's Thesis › Research
This thesis is part of a collaboration between two prestigious academic institutions, the University of Leoben (Austria) and the University of Bergen (Norway) to investigate the applicability of the ECLIPSE 100 black-oil foam model to a Texan carbonate reservoir sector model. The project aims to find an injection scheme with reduced CO2 mobility in a 5-spot pattern, in order to get a uniform sweep with an evenly propagating flood front in a reservoir with multiple high permeability streaks. Firstly, the foam model has been investigated in a generic way with homogeneous block model reflecting reservoir properties such as porosity, permeability and fluids. The intention with this approach was to study the physics of the model, to answer whether it is suitable for further analysis predicting the performance of enhanced oil recovery and CO2 storage. Secondly, a systematic sensitivity analysis has been performed to test the robustness of simulation outputs to extreme input values. Finally, the sector model, which was used for the numerical simulation, has been populated with data that are based on well logs and core measurements, honoring the actual the geology. A detailed qualitative analysis is presented investigating the effect of each model parameter with respect to the producing gas-oil ratio. Furthermore, the mobility reduction factor, the oil recovery efficiency, the productivity index as well as oil production rates are discussed. Considering all aspects, this module is applicable because its sensitive reflects a behavior, which is well known from core flood experiments. However, because of the limitations of black oil models, it should not be used for performance prediction of the enhanced oil recovery process, but for validation of compositional simulation models. For example, gas injection processes with significant and changing component partitioning during displacement are not accurately be modelled, because the PVT behavior cannot be described by simple tables.
|Award date||30 Jun 2017|
|State||Published - 2017|