Advanced Modeling and Simulation Techniques for Heavy Oil and Oil Sands Reservoirs

点击翻译按钮获取中文摘要

Alberta's 174 billion barrels of established oil sands reserves make Canada one of the top three countries globally - along with Saudi Arabia and Venezuela - in terms of proven or recoverable crude oil reserves. The key questions in recovering these resources are: What is the best technology available to recover them? Where and how can sustainable recovery be improved? How much is recoverable? The current major production methods for oil sands reserves include mining and enhanced oil recovery (EOR) methods. More than 90% of the vast energy resources in Alberta is found in deposits too deep underground to be mined at the surface. Thus the EOR methods must be used to coax the molasses-like bitumen, the "raw" heavy crude oil, to the surface. The commonly used EOR methods for heavy oil and oil sands reserves are thermal recovery methods such as steam assisted gravity drainage (SAGD) and cyclic steam stimulation (CSS). These methods take advantage of the dependence of the oil phase viscosity on temperature. Environmental impacts of these processes and the use of a high volume of water and natural gas in them suggest that extensive research is required for the economical and environmentally friendly development of these reservoirs. The development of novel EOR recovery methods for reservoirs with detailed geological description requires advanced mathematical models, solution algorithms, parallel techniques, and techniques capable of treating multi-scale phenomena. Also required is a reservoir simulator with unstructured grid capabilities to handle adaptive grid refinement to focus the grid to regions of the domain where small-scale phenomena occur. The proposed research program will develop mathematical models and simulation technologies capable of predicting heavy oil and oil sands reservoir performance under various operating conditions. It will include the design of advanced parallel, thermal reservoir simulators and process optimization toolkits suitable for fast, multiple runs that enable risk and uncertainty analysis and decision making in practical timeframes.