Research On Reservoir Three-dimensional Digital Representation Model And Its Visualization Implementation Technology

In recent years, most oil field in China has come to the middle, high water-cut stage. Inthe development process of injecting water, the developing work faces more and moredifficulty and the development contradictions have become increasingly prominent, whichcan be attributed to 'three high and two low'. The 'three high' refers to the comprehensivemoisture content, the recovery degree, and the production rate of remaining recoverablereserves. The 'two low' refers to reserve-production ratio and recovery ratio. There are a lotof surplus oil cannot be recovered. Due to the increase of water content, exploration workfaces more and more difficult, especially the discovery of new oil field. It are urgent toestablish a fine three-dimensional geological model from the point of the reservoirdevelopment to meet the needs of studying sophisticated numerical reservoir simulation andrevealing the space distribution law of the remaining oil, and to sum up a effective stochasticreservoir modeling techniques of suitable for oil reservoir development characteristics for theNo.1extraction factory. The above measures may provide powerful tools for oilfield EORresearch. In this paper, by comprehensively analyzing the existing reservoir structure models,methods, and EPDM data model, and applying three-dimensional geological modelingtechniques, the reservoir three-dimensional dynamic description method is studied indetailed.This paper conducts a preliminary study and exploration of the reservoirthree-dimensional digital representation model and visualization realization technology. Thespecific research topics include the following aspects. First, the foundation ofthree-dimensional geological modeling is addressed, which includes the basic framework ofthree-dimensional geological modeling and the reservoir three-dimensional digital modelingalgorithm. Secondly, the reservoir geological modeling method is elaborated, and thethree-dimensional digital dynamic generation algorithm is studied, which includes Kriginginterpolation algorithm, the variogram, and sequential directions modeling method. Thedeficiencies of the existing slicing algorithms are analyzed, the slicing algorithm is improved.Thirdly, a random particle swarm optimization algorithm is printed. The proposed algorithmhas only one control parameter, and the search step is dynamically determined by a randomvariable. The target location tracking is achieved through the rational design of the controlparameter. The experimental results of standard test functions show that compared with thequantum particle swarm optimization, both the optimization capabilities and efficiency have been improved. At the same time, the lack of variation function design method based on leastsquares method is analyzed, and the variogram based on the random particle swarmoptimization design methods is established, and the validity of the proposed method isverified through oil field data.Finally, this paper studied the key technologies involved insystem implementation, and achieve visual display of the system.