加速油藏数值模拟的POD模型降阶方法  

The POD Model Order Reduction Method of Accelerating Reservoir Numerical Simulation

作　　者：曹静[1] 赵辉[1] 喻高明[1] 胡文明 

机构地区：[1]长江大学石油工程学院,湖北武汉430100 [2]中原油田普光气田净化厂,四川达州636100

出　　处：《新疆大学学报（自然科学版）》2016年第1期102-108,共7页Journal of Xinjiang University(Natural Science Edition)

基　　金：国家自然科学基金项目(51344003)

摘　　要：实际油田建立的油藏模型可能包括几十万或几百万个网格,采用传统油藏模拟器进行模拟非常耗时,这也导致闭环油藏管理中生产优化和历史拟合过程计算成本的增加,如何大幅度加速油藏模拟运算速度是亟待解决的问题.本文将本征正交分解(POD)模型降阶方法应用于油藏数值模拟,首先,在训练过程中运行全阶油藏模拟器获得快照,构造POD的基矩阵,然后再利用获得的基矩阵进行不同生产制度下的预测过程.数值实验显示,在保证有足够精度的条件下,该方法可降低算例油藏模型的维数,使模拟器计算该算例的运算速度提高近3倍,验证了POD方法的有效性.Because the reservoir simulation models arising from real field application may consist of hundreds of thousands or millions of grid blocks, traditional reservoir simulators can be quite time consuming, which also led to production optimization and history matching process of closed-loop reservoir management computational cost increase. How to greatly accelerate reservoir simulation speed is the urgent problem to be solved. In this paper, the proper orthogonal decomposition （POD） model order reduction method is applied in reservoir numerical simulation. First of all, in the training process, we run full order reservoir simulator to get snapshots and construct the POD base matrix. Then we reuse the base matrix to predict under different production schedule. Numerical experiments show that this method can reduce the dimensions of the reservoir model and improve the operation speed of simulator computing the numerical example by nearly three times under the conditions with sufficient precision to ensure, verify the validity of the proposed method.

关 键 词：油藏数值模拟 全隐式 模型降阶 本征正交分解(POD) 控制系统 

分 类 号：TE319[石油与天然气工程—油气田开发工程] O242[理学—计算数学]