Depth domain phase-control inversion method

Abstract

The invention provides a depth domain phase-control inversion method. The depth domain phase-control inversion method comprises the steps of: A, carrying out seismic phase analysis; B, matching sample wells having similar waveforms; C, filtering in a range higher than a seismic frequency band to filter out a high frequency band, so as to find cut-off frequency of the similar waveforms having common structures; D, performing interpolation to establish an initial model, and carrying out phase control by combining with a seismic phase analysis result during the interpolation process; E, performing frequency analysis on seismic data, determining a low-pass frequency and a high-pass frequency, and filtering the initial model to filter out the low-pass frequency and the high-pass frequency, so as to obtain a band-notched initial model; F, combining with colored inversion, and merging an intermediate-frequency portion of colored inversion with low-frequency and high-frequency portions of the band-notched initial model by utilizing a frequency domain to obtain an absolute wave impedance or velocity invertomer, thereby completing depth domain reservoir inversion. The depth domain phase-control inversion method overcomes the high-frequency information loss caused by resampling during the well-logging curve time-depth conversion process, and effectively guarantees the high resolution feature of the seismic data.

Description

technical field [0001] The invention belongs to the field of oil and gas seismic exploration and investigation, and more specifically relates to a method for seismic interpretation and reservoir inversion in the depth domain. Background technique [0002] The comprehensive analysis process in seismic exploration technology has gone through three stages of development: "2D to 3D, post-stack to pre-stack, time domain to depth domain". Time-domain inversion is quite mature, but depth-domain inversion is a frontier topic. Although it has been discussed and researched, there are still no relatively perfect theories and methods so far. The inversion based on the convolution model mostly stays in converting the wavelet in the time domain to the depth domain for convolution, and fails to achieve the ultimate goal of eliminating the time-depth conversion. [0003] Although the current research on depth domain is still relatively small, and some special software has not been develope...

AI Technical Summary

Problems solved by technology

Therefore, it is difficult to use a unified spatial wavelet (depth domain wavelet) to construct a convolution model.

Images