Research On Spatial Anisotropic Random Simulation Under Constraint Of Sedimentary Facies

Geostatistics stochastic modeling is an important technology in the area of oil and gas exploration,it is also an important method to realize reliable stochastic modeling.Its implementation has an extremely significant influence on the reservoir prediction and gas exploration.It uses the known information(including logging attribute values and the corresponding three-dimensional coordinates)to build the variogram function,then simulating each single point through the sequential gaussian random way and kriging valuation method.But traditional sequential gaussian random simulation method is insufficient on building variogram function and estimating through kriging valuation.It mainly embodies in two aspects: 1)The variogram function is built on the basis of logging information in plane,which ignores the variable’s space characteristics;2)The traditional method hasn’t taken the influence of different sedimentary into consideration.Normally,the known points in the same sedimentary should be estimated heavily to the contribution.In a word,Traditional methods only consider the factor of the space distance,but the influence of sedimentary related to kriging valuation is considered less.In order to make up for the above,this thesis focuses on the research and improving of the construction of variogram function,and proposes multivariable controlled method of building a space anisotropic varoigram function,which considers the spatial variation characteristics of the geological variables.Furthermore,this thesis creativity introduces sedimentary facies into kriging estimation,realizing face-controlled three-dimensional spatial point searching and kriging interpolation,and also co-simulation,which makes the results can not only correspond the space variation characteristics,but also be in accordance with the geological distribution of the sedimentary facies.1.Proposing multivariable controlled method of building a space anisotropic varoigram function.This thesis constructs variogram functions from the horizontal and vertical direction.In horizontal direction,the variogram function is built through different parts in separation,then jointing together based on geometric anisotropy sets.In vertical direction,the variogram function is simple.After above,the horizontal part and the vertical part will be united through the way called ribbon heterogeneous.Because this thesis adopts the method of the co-simulation,so the construction of building variogram of primary variable and co-variogram of secondary variables is necessary.Actual cross-validation tests show that the variogram function’s stability performs significantly better than traditional methods.2.Putting forward the method of face controlled spatial anisotropic co-simulation.This method is fully combined with the distribution of spatial sedimentary facies of the work area,which makes the results of the simulation better fitted with the underground geological conditions.The innovation is mainly reflected in three aspects: On the method of known points’ determination,this thesis has realized the function based on the extended face controlled method to find known points,which can achieve higher precision of known points searching;On the method of building the variogram function,this thesis uses the multivariate spatial anisotropy variogram function,which makes the spatial changes can be reflected in the process of the simulation;In the process of simulation,this thesis uses the body of inversion or simulation data as the simulation’s secondary variable,and also the sedimentary face as the constraint conditions,which are made uses of getting the ratio of estimating the points of kriging function,and then the ratio will be used to participate the face controlled spatial anisotropic co-simulation.This method fully considered the situation that the neighborhood points of the estimated belong to different sedimentary facies.The actual verification of the simulation results show that the method is superior to traditional methods,and the valuation better accords with the actual characteristics of the underground circumstance.