Research And Implementation For Permeability Upscaling Algorithm Of Reservoir

Oil and gas Reservoirs occupies a very important position in China's petroleum industry. Three-dimensional geological model has large amounts of grid volume data, generally include several million to tens of millions of three-dimensional grid volume data, can not be directly put the geological model input to the simulation system with such a large amount of data. Upscale the geological model to reduce the amount of data, while keep its geological information, is a way to solve this problem. Based on the limited geological data, this thesis considered heterogeneity of three-dimensional geological model, and established the reservoir modeling with certain upscaling algorithm, to develop the geologists' thinking and guide the development of oil and gas field.The main contents of this dissertation reads as follows:First of all, this thesis studied the upscaling technology that relates to the 3d reservoir geology model, which is generated by the reservoir stochastic modeling. Considering the anisotropy of the permeability, we cannot upscale the modeling by the general methods. Based on this, this paper compared several existing upscaling technologies, discussed the reasonable way, and then proposed a new upscaling algorithm of heterogeneity which bases on the idea of refining part of the upscaling grids.Secondly, this paper utilized the object-oriented technology to analysis and to design the permeability upscaling system of reservoir. It used the idea of model design and the program principle which based on the interface-oriented technology; It also applied the object-oriented method and encapsulated upscaling algorithms into classes; And then employed the C++ language to carry out the system.Finally, this dissertation analyzed and validated the upscaling model. It upscaled the grids in uniform which on the same plane, filtered detailed information for the heterogeneity of the plane, and then refined them. Which made the processed data volume not only reduce the amount of data, but also improve the system speed, while maintain the reservoir's heterogeneity.