一种基于属性分布的非均匀油藏数值模拟网格优化算法

油藏数值模拟是油气田勘探开发的基础,三维地质网格模型是数值模拟的基础数据之一,网格的数量和大小决定了数值模型的精细程度,同时也影响了数值模拟的效率。为了有效保证地质模型的准确性,又可以提高数值模拟的效率,借鉴自适应网格法的思想,提出了一种改进的不均匀网格优化方法。该方法在储层物性非均质性强烈或重点研究区域采用细网格,而在储层物性均质性相对较好的区域采用粗网格,相邻粗、细网格间均匀渐变过渡。首先,设计了网格优化的基本原则;随后,在网格优化基本原则的基础之上,设计并实现了一种改进的不均匀网格优化算法;最后,以陕西省X区油田实例数据对算法的有效性和准确进行了验证。结果表明:该算法可以实现网格的均匀渐变过度,既有效减少了网格数量,又很好的保证了模型的精确度。

A Grid Optimization Algorithm Used in Reservoir Numerical Simulation Based on Grid Attribute Distribution

Reservoir numerical simulation is the basis of oil and gas exploration and development. Three-dimensional geological grid model is one of the basic data of numerical simulation. The number and size of grid determines the precision of numerical model, and also affects the efficiency of numerical simulation. In order to effectively guarantee the accuracy of geological model and improve the efficiency of numerical simulation, an improved non-uniform mesh optimization method is proposed based on the idea of adaptive mesh method. In this method, fine grids are used in areas where reservoir physical heterogeneity is strong or key research areas, while coarse grids are used in areas where reservoir physical homogeneity is relatively good, and the transition between adjacent coarse and fine grids is uniform and gradual. Firstly, the basic principles of grid optimization are designed. Then, based on the basic principles of grid optimization, an improved non-uniform grid optimization algorithm is designed and implemented. Finally, the validity and accuracy of the algorithm are validated by an example data of X oilfield in Shaanxi Province. The results show that the algorithm can realize the uniform gradient transition of the grid, which not only effectively reduces the number of grids, but also guarantees the accuracy of the model.