Application of Avant-Garde Simulation Technology to Explore the Complexity of the SAGD Process
The inherent complexity in the thermal physics and operational aspects of SAGD, offers many challenges to the SAGD numerical modeling process. This work explores how a group of advanced reservoir simulation techniques such as thermal multi-segment wells, parallel processing and non-linear gradient based optimization algorithms can be fully integrated with a Compositional Thermal Reservoir Simulator to gain maximum efficiency in modeling the complex interaction of the physical, thermodynamical and operational aspects of SAGD and achieve an unprecedented level of accuracy and performance in SAGD simulations. We investigate the applicability of thermal Multi- segment wells to model complex wellbore flow and thermal aspects of the process, such as the so important start-up phase and the implicit modeling of the pressure and heat transfer dynamics in the wellbore during SAGD operations. Further, we present a detailed analysis of alternative SAGD configurations, such as Cross-SAGD (X-SAGD) and Opposed SAGD, based on a multi-segment well description. Advanced parallel processing capabilities are combined with high performance computing (HPC) technologies in an effort to address one of the most important factors of SAGD thermal simulations: the associated cost in processing time. By proving the increase in speed and scalability that can be achieved by running SAGD thermal models using parallel computing techniques, we aim to provide a definite solution to simulate realistic high- resolution models in practical time frames. Finally, we evaluate the application of non-linear gradient based optimization algorithms, fully embedded in the thermal simulator, to automatically estimate the optimal SAGD operation parameters given a user-selected objective function, i.e. the Steam- to-Oil-Ratio (SOR). Based on a detailed numerical analysis, established from extensive simulation runs, we present a set of examples that illustrates the implications of integrating these technologies in the different engineering aspects of the SAGD modeling process.
C. PEREZ-DAMAS
Schlumberger
国际会议
首届世界重油大会(The Technical Sessions of the First World Heavy Oil Conference)
北京
英文
443-454
2006-11-12（万方平台首次上网日期，不代表论文的发表时间）