Seismic forward modeling method based on least square conjugate gradient iteration of frequency domain

A technique of least squares and conjugate gradient, applied in the field of seismic exploration, to achieve the effects of stable calculation, loose iteration conditions, and reliable simulation of seismic response characteristics

Pending Publication Date: 2021-11-19
SOUTHWEST PETROLEUM UNIV
View PDF1 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to aim at the defects of the prior art, and provide a seismic forward modeling method based on least squares conjugate gradient iteration in the frequency domain, which can maintain considerable Reduce memory consumption and improve calculation efficiency under the condition of accuracy. Compared with the forward modeling method based on BI-CGSTAB iterative solution, it can obtain more relaxed use conditions and better calculation stability when the calculation efficiency is close

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
  • Seismic forward modeling method based on least square conjugate gradient iteration of frequency domain
  • Seismic forward modeling method based on least square conjugate gradient iteration of frequency domain
  • Seismic forward modeling method based on least square conjugate gradient iteration of frequency domain

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0112] Choose a grid size of N x =N z = 100 of the homogeneous medium model, its discretized coefficient matrix structure is as follows figure 2 As shown, it can be seen that its coefficient matrix is ​​highly sparse, with a size of 10000*10000. The specific structure of its 9-point difference can be seen from the partially enlarged part of the figure.

Embodiment 2

[0114] choose as image 3 The two-dimensional layered model shown, its model mesh size is: N x =N z =300, space interval dx=dz=5m, setting parameters include:

[0115] The forward modeling time is 1s, the number of PML absorption layers is 10, the source coordinates are (150, 150), the main frequency of the source is 30Hz, the preset error is 0.01%, and the initial vector is x 0 to zero.

[0116] According to the forward modeling method of the present invention, after the relative error satisfies the preset error range, the output 30Hz frequency slice is as attached Figure 5 As shown, the wave field snapshot at the time of 312ms is attached Figure 6 As shown, it can be seen that: Figure 5 The characteristics of the single-frequency data sheet are clear, the upper low-speed layer has more wave numbers than the lower high-speed layer, and the wave number relationship between the two conforms to the theoretical relationship of the velocity model. The velocity interface is...

Embodiment 3

[0118] choose as Figure 4 The Marmousi model shown (Marmousi-1 model), its model grid size is: N x =737,N z =751, resample 1 / 4 sample point horizontally and vertically, so the model grid size becomes N x =184,N z =187, space interval dx=12.5m, dz=10m, setting parameters include:

[0119] The forward modeling time is 2s, the number of PML absorption layers is 50, the source coordinates are (50, 50), the main frequency of the source is 15Hz, the receiving point is located at z=49 above the shot point, the preset error is 0.01%, and the initial vector x 0 to zero.

[0120] According to the forward modeling method of the present invention, after the relative error satisfies the range of preset error, the output 15Hz frequency slice is as attached Figure 7 As shown, the wave field snapshot at the time of 625ms is attached Figure 8 As shown, it can be seen that: Figure 7 The high-speed belt on the right side of the single-frequency data slice model has a certain display o...

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 discloses a seismic forward modeling method based on least square conjugate gradient iteration of a frequency domain. The method comprises the following steps: inputting a time-harmonic wave equation after Fourier transform to the frequency domain; setting corresponding forward modeling simulation parameters; and performing iterative solution on the time-harmonic wave equation through a least square conjugate gradient algorithm according to the forward modeling simulation parameters. Compared with a forward modeling method based on LU direct solution, the method provided by the invention can reduce memory consumption and improve calculation efficiency under the condition of keeping the same precision, and compared with a forward modeling method based on BI-CGSTAB iterative solution, the method provided by the invention can obtain more loose use conditions and better calculation stability under the condition of approximate calculation efficiency.

Description

technical field [0001] The invention relates to the technical field of seismic exploration. Background technique [0002] As a basic geophysical exploration method, seismic exploration is the most effective method to solve oil and gas exploration problems. Forward modeling based on the wave equation is the basic work of seismic exploration, and plays an important role in the acquisition, processing, and interpretation stages of seismic exploration. [0003] Seismic forward modeling methods in the prior art can be divided into time domain and frequency domain according to the calculation domain. The time domain algorithm has been widely used because of its simple implementation and high precision. However, under the requirements of full waveform inversion, it needs Based on the forward modeling, iteratively solve the wave equation repeatedly. If the forward modeling is performed in the time domain, it will bring a huge amount of calculation and consume a lot of memory and ti...

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 Applications(China)
IPC IPC(8): G01V1/36
CPCG01V1/282G01V1/36
Inventor 刘文革周觅路尹成牟其松涂文茂李崇文
Owner SOUTHWEST PETROLEUM 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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap