Methods And Applications Study For Improving The Accuracy Of Reservoir Prediction And Modeling

During the levels of reservoir evaluation and development the object of reservoir investigation is reservoir modeling, but now the difficulty of modeling is how to improve the prediction accuracy across wells. When the well data is sparse the seismic data takes an important role in modeling. The low vertical resolution and strong ambiguity of the seismic data affects the accuracy of reservoir prediction across wells. In order to improve the accuracy of reservoir prediction across wells and modeling, based on the post-stack and pre-stack seismic data, an study in three levels which are theory exploring, novel methodology investigation and techniques application is conducted respectively.In order to improve the prediction accuracy of applying the seismic data, the advanced poroelastic AVO theory is chosen in this paper. The exploring study of the AVO method for reservoir prediction in poroelastic media is conducted, which includes principle and methodology, programming, applying mathematical model data to test, especially putting the complexity theory into real application. In the SN31 area, the two-dimension inversion for P-wave velocity and porosity has been done across well profile by using the real pre-stack seismic data.For post-stack seismic data, the in-depth study of reservoir prediction method from seismic waveform structure discrimination is conducted by applying the simplified poroelastic theory which is the equation of reflection coefficient under the condition of the equivalent media. The principle and methodology of waveform structure discrimination and methodology of building the model bank of seismic response are studied. The novel idea of reservoir interval spectrum is introduced and the spectrum is built. Then the method of automatic interpreting the reservoir interval spectrum is studied and extended from well site to three-dimension seismic data. So the three-dimension seismic data is converted directly into three-dimension data of sand distribution. Thus there has been a novel method of predicting the three-dimension distribution of sand.In order to excavate the technologic potentiality of existing impendence inversion, the influence of optimizing seismic data and sampling rate on inversion accuracy are analyzed quantitatively. How to improve the accuracy of seismic-well tie and the initial impendence building is investigated. The exploring study about the essence of seismic-well joint inversion to improve the seismic resolution is conducted. So the inversion accuracy of existing techniques is improved effectively.The reservoir prediction and modeling are conducted in the SN31 well area which is selected as the test project. The methodology and techniques in the paper are applied to real data. The reservoir modeling is applied to check the prediction accuracy quantitatively. The structural modeling, reservoir configuration modeling and porosity modeling are done. The result of the real data indicates the accuracy of reservoir prediction and model is good.