Method for predicting reservoir stratum by seismic inversion data

Abstract

The embodiment of the invention provides a method for predicting a reservoir stratum by seismic inversion data. The method comprises the following steps: well inversion is carried out on the seismic data of a certain study area based on a model to obtain wave impedance inversion data of the study area; when geology and logging information used in wave impedance inversion of the study area is analyzed and that each reservoir stratum and adjacent non reservoir stratums can be distinguished on a wave impedance curve is determined, a judgment curve of a well reservoir stratum on wave impedance is established; the established judgment curve of the well reservoir stratum on the wave impedance is interpolated and extrapolated within the same spacial scale in the same interpolation algorithm used for establishing an initial wave impedance model in inversion to form reservoir stratum judgment data; and the reservoir stratum judgment data is used for judging inversion data so as to predict the reservoir stratum of the inversion data. The embodiment of the invention provides a new means for lithologic character or reservoir stratum prediction under a complicated condition directly by wave impedance or speed inversion data obtained on the basis of model inversion.

Description

Technical field [0001] The invention relates to seismic exploration data processing technology, in particular to a method for predicting reservoirs with seismic inversion data. Background technique [0002] Seismic wave impedance or velocity inversion can convert the underground reflection information received on the ground into information reflecting underground rocks. It is an important technical method for people to use post-stack seismic data for reservoir prediction and reservoir physical parameter estimation. In sand and mudstone formations, when the wave impedance of sand and mudstone has a clear and uniform difference in the vertical and horizontal directions, a single numerical threshold can be used to identify and predict sandstone reservoirs in the wave impedance inversion data. Simple obvious differences are rare today. Due to the undercompaction of the formation or changes in the physical properties of the reservoir, the response difference between the reservoir and...

AI Technical Summary

Problems solved by technology

This method must be carried out in three dimensions and there must be a large number of known wells with relatively uniform distribution. In the vertical and horizontal directions, the lithology data of the borehole and the seismic data extracted from the seismic data can roughly reflect the distribution of lithology. The attributes determine the vertical and horizontal variograms, and the variograms will be applied to the whole area after they are determined. If different sedimentary environments in different strata in the work area or multiple provenance systems exist in the same stratum, they need to be divided into blocks or layers. Simulation (increases the workload), and the estimation results are greatly affected by the model, so the model must be accurate (when the deposition relationship is complex, a lot of workload will be increased)

Therefore, the amount of analysis and calculation is large, and sufficient hardware support is required.

3. Use foreign emerge software to combine the inverted wave impedance and the amplitude, frequency, phase and other attributes extracted from the seismic data through linear or nonlinear algorithms to fit curves that have a clear response to lithology, such as Sp, porosity, etc. Certain relevant conditions need to be met to obtain better results, and the resolution of seismic attributes is limited by seismic data, which will affect the fitting effect

However, because it follows the shape of the curve and the change of the absolute value in the inversion data at different places is often difficult to grasp, it is rarely used as a control parameter. When it is necessary to use the changing absolute value as a control to remove some unwanted relative changes In addition, when the reservoir in a certain position corresponds to the relatively high part of the inversion curve in the vertical direction, and other parts correspond to the relatively low part of the inversion curve, it is also difficult to predict the results. represented in the same section

To sum up, the above technical methods all have certain applicable conditions, and it is inconvenient to unconditionally meet the production requirements of reservoir prediction

Images