科研之友
微信
新浪微博
Facebook
分享链接
摘要
Reservoir simulator constitutes the core of data analysis in petroleum reservoir development. With industry's advancements in unconventional resources, such as CBM, shale gas and shale oil, new recovery mechanisms have been identified. However, it may be impractical to implement some of the relevant mathematical models in simulators that are originally designed for conventional problems. In this work, we developed the UNConventional Oil and Gas simulator (UNCONG) under a more flexible architecture. UNCONG is based on the black oil model and the compositional model but does more than just to cover the aspects of traditional models. It possesses several noticeable features, which enable it to handle unconventional oil and gas problems more properly: 1) it provides multiple ways to model fractures and allows to mix structured and unstructured grid; 2) it uses the multi-point flux approximation (MPFA) to guarantee the accuracy of flow terms in non-K-orthogonal grid; 3) it realizes a flexible flow equation framework to incorporate various permeability models, in which the diffusion and the geomechanic effects can be considered; and 4) the linear solver is tailored for problems containing MPFA and complex wells. UNCONG is parallelized on shared memories, and has shown good acceleration ratio in limited tests. UNCONG has been validated with commercial simulators in conventional problems and successfully applied in studying the mechanisms of unconventional oil and gas recovery, producing reasonable and self-consistent results. Moreover, owing to its robustness, efficiency, and user-friendliness, UNCONG is in fact applicable to more realistic industrial problems.
- 出版日期2015-9
- 单位北京大学
