Research Of Complex Surface Reconstruction Method Based On Uncertain Programming

Geological surface reconstruction is an important basic work in the exploration field of oil and gas resources,it is the basis of studies of geological structure mapping,geologic modeling and reservoir simulation.Its main task is to use structural interpretation data and well data to construct geological surfaces from the perspective of computer graphics.Different from common surface reconstruction,there is a great uncertainty in the reconstruction of geological surfaces.The uncertainty is mainly reflected on two aspects: one is that sparse structural interpretation data and well data lead to the lack of geological surface structural feature information;the other one is that the limitation of exploration technology leads to greater uncertainty in structural interpretation data itself.For the problem of reconstruction of the uncertain geological surface,the existing methods mainly start from the idea of random disturbance,and randomly generate multiple possible surfaces.However,due to the lack of quantitative analysis methods for each surface,existing researches cannot meet actual needs.Aiming at the deficiencies in existing researches,this thesis proposes a geological surface reconstruction method based on uncertain programming.The main idea is to transform the geological surface reconstruction problem into an uncertain programming model.Through the solution to the model,a more reasonable geological surface can be obtained.Research has theoretical value and practical application value.This thesis focuses on the reconstruction of the uncertain geological surface and starts from the uncertain programming.The main work and innovations are as follows:1.The structural feature description method of geological surface based on semantic and feature parameters is proposed.For the reconstruction of uncertain geological surfaces,geologists pay more attention to the uncertainty of structural features in reconstructed surfaces.Based on this,this thesis studies the characterization methods of geological surface feature morphologies,and proposes a characterization method of structural feature morphologies based on semantic description.The basic idea is to establish corresponding parametric models for the main geological structures.Thus,the geological structural features are transformed into geometric models under the control of different parameters.And the parameter extraction method of the fold structure of the geological surface was studied.Theoretical analysis and simulation verify the validity and feasibility of the description method.2.A geological surface reconstruction method based on feature and uncertain programming is proposed.Based on the structural morphological feature parameter model,the uncertainty of the geological surface is transformed into the uncertainty of the structural model.Aiming at this problem,this thesis proposes a surface reconstruction method based on feature and uncertain programming.The main idea is to establish an uncertain programming model based on accuracy and smoothness in the geological surface,and at the same time,the lower uncertainty structural feature description is incorporated into the model as constraint data.Then the uncertain programming model is transformed into a deterministic programming model.Finally,solving the problem to obtain an optimal reconstruction result under this model.Through theoretical analysis and simulation,the feasibility of the method is verified.This thesis studies the reconstruction methods of the uncertain surface from two aspects: the semantic description of the structural morphological features of the geological surface and the uncertain programming problem based on the structural morphological features.It has achieved some research results.And the follow-up research is prospected.