Plane seismic exploration signal decomposition method based on Kriging method

Abstract

A plane seismic exploration signal decomposition method based on a Kriging method belongs to the field of exploration geophysics. The method comprises the following steps: acquiring a seismic exploration signal, setting the basic lag distance and the maximum lag distance, and calculating a two-dimensional experimental variation function of a seismic attribute slice to obtain a two-dimensional variation function graph; determining ellipse information corresponding to target information from the variation function graph, extracting experimental variation function curves of long-axis and short-axis directions (respectively corresponding to primary and secondary directions of the target information), and using a theoretical model to perform one-dimensional second-order nested fitting; determining the initial parameters, including sill, long-axis range, ellipse direction and the ratio of long axis to short axis, of the two-dimensional variation function according to the parameters of the separated variation function curves of the primary and secondary directions, tuning parameters, selecting different models (spherical model, exponential model, Gauss model), and making the fitted two-dimensional variation function more consistent with a target signal experimental variation function; and determining a two-dimensional theoretical variation function, substituting the function into a Kriging interpolation equation, and decomposing to obtain the target information.

Description

technical field [0001] The invention relates to a signal decomposition method for plane seismic exploration based on kriging method, belongs to the field of exploration geophysics, is a signal decomposition method for plane seismic exploration based on kriging method, and is an important tool for plane seismic attribute analysis and data processing, Lay the groundwork for subsequent reservoir predictions. Background technique [0002] Plane seismic exploration signals (usually plane seismic attributes) refer to seismic attribute slices extracted along stratigraphic horizons or isochronous slices (time windows) on a 3D seismic data volume, and are mainly used for quantitative seismic facies analysis and reservoir prediction. Due to the influence of the seismic acquisition and observation system, the influence of the quality of the original 3D seismic data and the accuracy of the interpretation horizon, and the constraints of the attribute data extraction method, there are oft...

AI Technical Summary

Problems solved by technology

Moreover, in the Steerable Pyramid decomposition elbow, whether it is random noise or correlated noise, it will reduce the accuracy of direction estimation.

[0007] The above-mentioned multi-scale and multi-directional image theory method in the transform domain transforms the spatial frequency domain and completes the decomposition according to the spectral characteristics of the signal. Although there is an indirect correspondence to the spatial structure relationship, the mathematical transformation method of the discrete signal often produces false images, directly Analyzing the spatial structure relationship is definitely the best way to decompose a certain structural information component of the signal

Images