No abstract

No abstract

Understanding the microscopic flow behaviour of hydrocarbons and water in porous media gains importance as more and more reservoirs are being exploited. Network modelling techniques could be extended to tighter media as long as Darcy's law is applicable. 3D random networks are constructed in order to represent the Mesaverde formation which is located in north Wyoming, USA. The network modelling software solves the fundamental equations of single-phase and two-phase immiscible flow incorporating wettability and contact angle assuming a quasi-static displacement mechanism. Macroscopic properties of the porous media network representation such as porosity, absolute permeability, and formation factor are calculated and whenever possible compared to experimental data. Subsequently, immiscible two-phase flow properties such as capillary pressure, relative permeability, and resistivity curves are predicted and compared to available experimental data. The effect of interfacial tension alteration is also investigated as an attempt to demonstrate the capability of the network modelling technique to show physical fluid behaviour. It is observed that the capillary pressure curve obtained using MICP data can be used to calibrate and validate the network model generated to represent the sample. The study shows that the modified random network modelling technique is capable of modelling low permeable porous medium and predicting single-phase and immiscible two-phase flow properties assuming quasi-static displacement mechanism.

ConocoPhillips operates Surmont, which is the first Steam-Assisted Gravity Drainage (SAGD) project to implement Flow Control Devices (FCDs) in producer wells. This study was conducted to evaluate the production performance of different liner completion strategies. The analysis compared well pairs completed with slotted liners (SL) to producers completed with FCDs, both liner deployed (LD-FCD) and tubing deployed (TD-FCD), and investigated the impact of FCDs in injectors. An extensive analysis was conducted using available production and temperature data along the wells. The wells were completed using various fixed-resistance FCD settings, while some wells were completed using variable setting designs. As time went on, several of the slotted liner producer wells were retrofitted with tubing-deployed FCD completions. One of the key objectives of the study was to determine the success rate of tubing-deployed FCDs and their performance relative to liner-deployed FCD wells. Another objective was to evaluate the impact of retrofitting slotted liner SAGD injectors with tubing-deployed FCD completions. In this study, a grading system was established based on the reservoir quality along the well for both injector and producer. For similar graded well pairs, LD-FCDs had better production performance than TD-FCDs. Considering similar graded reservoir quality, FCDs consistently performed better than slotted liners, in both conformance and production acceleration. The production analysis showed that the FCD flow restriction was a major controller of the conformance, but considering the self-choking phenomenon of the reservoir, most FCDs can perform positively in different circumstances. In this study, the self-choking effect of the liquid pool is discussed and explained for different reservoirs and variable subcool. Generally, if erosion is not a factor, FCDs can create a more controlling system than liquid-pool dominant systems. In these cases, both conformance and production acceleration is enhanced if operators yield lower subcools and greater draw-down pressures.

No abstract