Search

CN-121995474-A - Seismic reflection coefficient inversion method, system and medium based on L0 norm constraint

CN121995474ACN 121995474 ACN121995474 ACN 121995474ACN-121995474-A

Abstract

The invention provides an inversion method, system and medium for seismic reflection coefficients based on L0 norm constraint, and belongs to the field of oil-gas geophysical exploration. The method comprises the steps of 100, 200, constructing an inversion equation based on L0 norm constraint, 300, solving an optimal solution for the constructed inversion equation based on L0 norm constraint, and obtaining a reflection coefficient. According to the invention, the constraint of L0 norm is increased during inversion of the reflection coefficient, namely, the reflection coefficient is 0 as much as possible, but the wavelet can be reconstructed, so that the signal-to-noise ratio of the reflection coefficient is high, the transverse continuity is good, the resolution capability of an earthquake is improved, and the precision of seismic exploration is enhanced by the earthquake frequency band.

Inventors

  • WANG XIAOPIN

Assignees

  • 中国石油化工股份有限公司
  • 中石化石油物探技术研究院有限公司

Dates

Publication Date
20260508
Application Date
20241104

Claims (10)

  1. 1. An inversion method of seismic reflection coefficients based on L0 norm constraint is characterized by comprising the following steps: step 100, acquiring post-stack seismic data; Step 200, constructing an inversion equation based on L0 norm constraint; And 300, solving an optimal solution for the constructed inversion equation based on the L0 norm constraint to obtain a reflection coefficient.
  2. 2. The method of claim 1, wherein the post-stack seismic data comprises a plurality of post-stack seismic sub-data, the post-stack seismic record data being denoted { d i }, i = 1, 2.
  3. 3. The method of claim 1, wherein constructing an inversion equation based on the L0 norm constraint in step 200 is: Wherein, beta is a constant, H is a binary function H (0) =0, the others are 1, lambda is a constant parameter, the general order of magnitude is 0.005, wr=d+n, w is a seismic wavelet matrix, R is a reflection coefficient, d is post-stack seismic data, R is an auxiliary parameter, R= { R i }.
  4. 4. A method according to claim 3, wherein in step 300, the constructed inversion equation based on the L0 norm constraint is solved for optimal solution to obtain the reflection coefficient, and the specific operations include: step 301, splitting a constructed inversion equation based on the L0 norm constraint, which specifically includes: because each component of the reflection coefficient r can be optimized individually, the inversion equation is written as: Splitting the formula (4) into 2 parts: step 302, obtaining an optimal solution of the auxiliary parameter, which specifically includes: Inputting an initial reflection coefficient, and obtaining an optimal solution of the auxiliary parameter by adopting the following formula: wherein i=1, 2, the term, N; step 303, calculating to obtain a reflection coefficient by adopting an optimal solution of the auxiliary parameter, which comprises the following specific operations: Adopting a formula (5), and solving the bias derivative of r i to obtain the following formula: substituting the obtained optimal solution of the auxiliary parameter into a formula (9), and calculating to obtain the reflection coefficient.
  5. 5. The method of claim 4, wherein the optimizing the constructed inversion equation based on the L0 norm constraint in step 300 to obtain the reflection coefficient further comprises: Step 304, iterative optimization is performed to obtain a final reflection coefficient, and the specific operations include: and (3) taking the reflection coefficient calculated in the step (303) as input, repeating the steps (302, 303), and performing iterative optimization for N times to obtain a final reflection coefficient.
  6. 6. An inversion system for seismic reflection coefficients based on an L0 norm constraint, comprising: an acquisition unit for acquiring post-stack seismic data; The construction unit is used for constructing an inversion equation based on the L0 norm constraint; and the solving unit is used for solving an optimal solution for the constructed inversion equation based on the L0 norm constraint to obtain the reflection coefficient.
  7. 7. The system of claim 6, wherein the post-stack seismic data acquired in the acquisition unit comprises a plurality of post-stack seismic sub-data, the post-stack seismic record data being denoted { d i }, i = 1, 2.
  8. 8. The system of claim 6, wherein the inversion equation based on the L0 norm constraint constructed in the construction unit is: Wherein, beta is another constant, H is a binary function H (0) =0, the others are 1, lambda is a constant parameter, the general order of magnitude is 0.005, wr=d+n, w is a seismic wavelet matrix, R is a reflection coefficient, d is post-stack seismic data, and R is an auxiliary parameter.
  9. 9. The system according to claim 8, wherein the solving unit solves the constructed inversion equation based on the L0 norm constraint for an optimal solution to obtain the reflection coefficient, specifically performs the following operations: step 301, splitting a constructed inversion equation based on the L0 norm constraint, which specifically includes: because each component of the reflection coefficient r can be optimized individually, the inversion equation is written as: Splitting the formula (4) into 2 parts: step 302, obtaining an optimal solution of the auxiliary parameter, which specifically includes: Inputting an initial reflection coefficient, and obtaining an optimal solution of the auxiliary parameter by adopting the following formula: wherein i=1, 2, the term, N; step 303, calculating to obtain a reflection coefficient by adopting an optimal solution of the auxiliary parameter, which comprises the following specific operations: Adopting a formula (5), and solving the bias derivative of r i to obtain the following formula: Substituting the obtained optimal solution of the auxiliary parameter into a formula (9), and calculating to obtain a reflection coefficient; Step 304, iterative optimization is performed to obtain a final reflection coefficient, and the specific operations include: and (3) taking the reflection coefficient calculated in the step (303) as input, repeating the steps (302, 303), and performing iterative optimization for N times to obtain a final reflection coefficient.
  10. 10. A computer-readable storage medium storing at least one program executable by a computer, the at least one program when executed by the computer causing the computer to perform the steps in the seismic reflection coefficient inversion method based on the L0 norm constraint according to any one of claims 1-5.

Description

Seismic reflection coefficient inversion method, system and medium based on L0 norm constraint Technical Field The invention belongs to the field of oil and gas geophysical exploration, and particularly relates to an inversion method, system and medium for seismic reflection coefficients based on L0 norm constraint. Background The problem of seismic exploration resolution is an important problem throughout the whole process of seismic acquisition, processing and interpretation, and the traditional deconvolution method based on the deconvolution model is difficult to adapt to the actual requirements of the current high-precision seismic exploration. In order to solve the practical problems in exploration production, researchers research and develop a seismic nonlinear inversion method and a corresponding technology capable of improving the resolution of seismic exploration by absorbing and referencing the development theory and results of other subjects. Tang Yuyuan et al propose a seismic reflection coefficient method (2013, fifth annual meeting of hydrocarbon geophysics) based on L1-2 norm constraint, introduce L1-2 norm sparse regularization into seismic deconvolution, solve an objective function by adopting pDCA algorithm, realize seismic reflection coefficient inversion, verify the advantages of L1-2 norms through numerical simulation data testing, have better effect performance, better noise immunity and can well protect the relative position and amplitude of the reflection coefficient when the seismic data contain noise. Zhu Xiangyu and the like propose to improve the resolution (2022, petroleum geophysical prospecting) of the seismic data based on reflection coefficient inversion, firstly, according to a compressed sensing theory, an objective function under the constraint of a sparse rule operator is constructed by utilizing sparsity of the reflection coefficient and compressibility of the seismic data, then the reflection coefficient is inverted through a rapid soft threshold iteration method, the frequency spectrum of the seismic data is compensated by utilizing low-frequency and high-frequency components in the frequency spectrum of the reflection coefficient, the reconstruction of the frequency spectrum of the seismic data is realized, and finally, the purpose of bidirectionally widening a frequency band to improve the resolution of the seismic data is achieved. Chinese patent publication CN115327624A discloses an inversion method and an inversion system for seismic wavelets and reflection coefficients, which assume that the seismic wavelets have tight support and smoothness, assume that the reflection coefficients are relatively sparse, and construct the optimization problem of corresponding inversion seismic wavelets and reflection coefficient sequences, split the joint inversion problem of the seismic wavelets and the reflection coefficients based on the tight smoothness and the relative sparsity into wavelet inversion sub-problems and reflection coefficient inversion sub-problems by using alternate iteration, and solve the two sub-problems by a near-end algorithm. Compared with the existing seismic wavelet and reflection coefficient inversion method, the method has the advantages that the optimal parameters are easy to select, the inversion obtained reflection coefficient has good transverse continuity, and the like. The reflection coefficient obtained by the existing method is disordered in the transverse direction, the signal to noise ratio is low, and the transverse continuity is poor and has a certain gap from the actual geological condition. Disclosure of Invention The invention aims to solve the problems in the prior art and provide a seismic reflection coefficient inversion method, a system and a medium based on L0 norm constraint, which improve the inversion precision of reflection coefficients of post-stack seismic data and the transverse continuity of sections, expand the seismic frequency band and enhance the seismic exploration precision. The invention is realized by the following technical scheme: In a first aspect of the present invention, there is provided a seismic reflection coefficient inversion method based on an L0 norm constraint, comprising: step 100, acquiring post-stack seismic data; Step 200, constructing an inversion equation based on L0 norm constraint; And 300, solving an optimal solution for the constructed inversion equation based on the L0 norm constraint to obtain a reflection coefficient. The invention further improves that: the post-stack seismic data includes a plurality of post-stack seismic sub-data, the post-stack seismic record data being recorded as { d i }, i=1, 2. The invention further improves that: the inversion equation based on the L0 norm constraint is constructed in step 200 as follows: Wherein, beta is another constant, H is a binary function H (0) =0, the others are 1, lambda is a constant parameter, the general order of magnitude is 0.005, w