Time domain multi-scale full waveform inversion method based on nonlinear high-order frequency extension

A full-waveform inversion and time-domain technology, which is applied to instruments, geophysical surveys, and measurement devices, can solve the problems of large amount of calculation in waveform inversion, affecting the construction of long-wavelength components, and complex attenuation parameter extraction. Weekly problems, easy to operate, and avoid dependence effects

Inactive Publication Date: 2019-08-02
JILIN UNIV
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Laplace waveform inversion has a large amount of calculation, and the extraction of attenuation parameters is complicated
Envelope inversion is accompanied by a wide source frequency band, which will introduce a lot of high-frequency information and affect the construction of long-wavelength components

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Time domain multi-scale full waveform inversion method based on nonlinear high-order frequency extension
  • Time domain multi-scale full waveform inversion method based on nonlinear high-order frequency extension
  • Time domain multi-scale full waveform inversion method based on nonlinear high-order frequency extension

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0092] The overall process of the present invention is as figure 1 shown.

[0093] use as figure 2 As can be seen from the spectrum analysis results of the shown Reck wavelet, the main frequency of the wavelet is 20Hz, and the low-frequency information below 8Hz is missing. The high-order low-frequency reconstruction method of the present invention is adopted. Reconstruction results using n=1, 2, 3, 11 are as follows image 3 (a)-3(d). It can be seen that the waveform of the seismic wave becomes narrower after the high-order power operation. image 3 The spectrum analysis results of (a)-3(d) are as follows image 3 (e)-3(h). It can be seen that when n>2, the energy below 8Hz is restored to varying degrees. When n is larger, the reconstructed record is more and more close to the broadband impulse function. The reconstructed low-frequency components can be extracted by low-pass filtering.

[0094]The behavior of the objective function is analyzed below. The real model...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention relates to a non-linear high order frequency expanding based time domain multi-scale full waveform inversion method. The method includes calculating a high power of seismic data; performing low pass filtering on the seismic data subjected to high power operation, and obtaining a gradient through zero delay cross-correlation operation of a positive wave field and a reverse wave field;updating a model through methods like a conjugate gradient method, taking the model as an initial model, performing data quadratic inversion, and obtaining a final inversion result after the end of the iteration. The method utilizes low frequency data reconstructed with different orders, so that multi-scale full waveform inversion can be achieved, the cycle skipping problem of full waveform inversion can be effectively overcome, and the dependence on data original bands can be avoided. The provided waveform compression and data stripping method are actually based on energy, and thus, regardless of the length of the seismic record, low frequency reconstruction and data stripping can be performed on each single shot record through the high power operation, and the loss problem of partial shot information caused by time delay can be solved.

Description

[0001] Technical field: [0002] The invention relates to a method for reconstructing missing low frequency and performing multi-scale full waveform inversion by using a nonlinear high-order frequency extension method. Nonlinear high-order operations expand the frequency band of limited-bandwidth seismic data through nonlinear methods, and extract and utilize reconstructed low-frequency information, which can effectively overcome the cycle-skipping problem of full waveform inversion. [0003] Background technique: [0004] Full waveform inversion drives velocity model updates by minimizing residuals between observed and simulated data. According to the Born approximation, this strongly nonlinear problem can be linearized based on the weak scattering approximation and thus can be solved by a local optimization algorithm. This requires the inversion to have an initial model that is close enough to the true model. However, in actual processing, conventional velocity analysis met...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G01V1/28G01V1/30C08G8/20
CPCC08G8/20G01V1/282G01V1/306G01V2210/62
Inventor 张盼韩立国胡勇许卓
Owner JILIN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products