Inversion Of Reservoir Parameters Based On Petrophysical Model

Porosity,saturation and other reservoir parameters characterize the oil and gas bearing property of the reservoir,which is an important basis for describing the distribution of underground reservoirs and calculating the energy storage and productivity of the reservoir.However,reservoir parameters cannot be obtained directly from seismic data.Usually,petrophysical models are used to establish the relationship between reservoir elastic parameters and reservoir parameters,and appropriate inversion algorithms are selected to obtain reservoir parameters.However,different reservoir types are affected by lithology,sedimentary environment and other factors,and the relationship between elastic parameters and physical parameters is different from the corresponding inversion algorithms.Therefore,the research on reservoir inversion method of petrophysical model for different types of reservoirs is of great significance to subsequent reservoir development.Based on the comprehensive research status and data conditions,this thesis puts forward the petrophysical reservoir inversion strategies corresponding to the reservoir characteristics in terms of model construction and algorithm selection for clastic rock reservoir and carbonate rock reservoir.The accuracy of the inversion results is determined by the coincidence degree between the petrophysical model and the reservoir rock.According to the laboratory measurement data of the core in the work area,this thesis obtains the modeling methods suitable for clastic rock and carbonate rock respectively by using a variety of petrophysical theories such as critical porosity model and equivalent medium theory,evaluates the degree of linearization,and based on the constructed petrophysical model,In view of the lack of shear wave logging data in the work area,considering the influence of matrix in low porosity reservoir,a shear wave prediction method of variable matrix is proposed,which provides shear wave data for seismic data inversion.Inversion algorithm has a direct impact on the accuracy of reservoir parameter inversion results,which can be divided into linear inversion method and nonlinear inversion method.According to the characteristics of clastic rock reservoir,the linear inversion method is applied in this thesis.By linearizing the petrophysical model and directly solving the linearized inverse problem by using the conjugate gradient method,the inversion efficiency of seismic reservoir parameters is improved;According to the characteristics of carbonate reservoir,the nonlinearity of the model is evaluated,and the simulated annealing trust region joint inversion algorithm is proposed.Aiming at the lack of matrix parameters in the model,the neural network is used to predict the dolomite content.Starting from the two aspects of model and algorithm,this thesis puts forward the corresponding inversion strategies of reservoir parameters for different lithologic reservoirs,which is of great significance to the subsequent development and interpretation.