Research Of Geological Space-Time Methods Based On Reservoir

Reservoir is a gathering place for important natural resources such as oil and gas, as well as of the target area of oil and gas exploration and exploitation. Accompanying with development of oil and gas exploration, put forward higher requirements for the acquaintance of geological structure of the reservoir. Some existing methods have not been able to help geologists to keep abreast of changes in the reservoir oil and gas resources effectively, making it difficult to make a mining plan. Therefore, in the thesis geological 3D modeling algorithm is combined with multi-parameter calibration algorithm and fluid modeling algorithm based on 4D seismic to form Geological Space-Time Methods based on Reservoir, in order to improve the recovery of oil and gas resources and make a contribution to the development of national economic.The thesis firstly introduces the DEM data which is used to describe the geological structure of the reservoir, and discuss the generating algorithms of the triangulation network. In order to put forward 3D modeling method which suitable for the reservoir geology, it is necessary to make some improvements on the traditional geological modeling methods. The generated model can meet the requirements of real-time display while make sure of obtaining lower distortion.Since the amount of reservoir data are often very large, rendering speed is extremely slow only rely on the CPU to generate reservoir geological model. In order to increase the speed of modeling, GPU with is based on the architecture of CUDA is introduced. Compared with the CPU, architecture of GPU determines it suit for handling problem in large quantities, GPU has stronger capability of rendering, relatively lower of power consumption and cost. The utilization of GPU parallel computing to greatly enhance the speed of reservoir modeling.During the operation of monitoring and measurement of oil and gas resources in the work area, time-lapse seismic data multiple observation have some differences. In order to remove such differences and obtain more precise information of oil and gas resources, multi-parameter calibration algorithm based on four-dimensional seismic data is applied. Calibration and processing on multi-parameter, such as waveform, time, phase and amplitude, it can help people acquaint the changes in oil and gas resources, and speculate the storage of the remaining oil and gas resources in the reservoir.Aiming at seismic data of time-varying which includes multi-step phenomenon, modeling algorithm of fluid property based on 4D seismic is adopted which is by extracting and tracking of fluid properties that make a good solution to the problem. It can isolate the fluid property of interest from the data so that the size of data reduce to one-tenth of the original or less, thus, enhance the speed of real-time rendering on time-varying seismic data.