CN-116010761-B - Point spread function calculation method, point spread function calculation device, electronic equipment and medium

CN116010761BCN 116010761 BCN116010761 BCN 116010761BCN-116010761-B

Abstract

The application discloses a point spread function calculation method, a point spread function calculation device, electronic equipment and a medium. The method can include the steps of establishing an initial point spread function, setting constraint parameters, obtaining a final point spread function according to the initial point spread function and the constraint parameters, and calculating and solving. The application realizes the rapid solution of the point spread function based on the local neighborhood and the high-frequency approximation, better serves the property of the Hessian matrix and the seismic data Gao Weizheng research, provides a new technical idea for the high-resolution processing of the high-precision and high-efficiency seismic data in the future and provides a powerful technical support for practical production and application.

Inventors

TAO YONGHUI
BAI YINGZHE
XIANG CHEN

Assignees

中国石油化工股份有限公司
中国石油化工股份有限公司石油物探技术研究院

Dates

Publication Date: 20260508
Application Date: 20211021

Claims (4)

1. A method of calculating a point spread function, comprising: Establishing an initial point spread function; setting constraint parameters; Obtaining a final point spread function according to the initial point spread function and the constraint parameter, and calculating and solving; wherein establishing the initial point spread function includes: establishing a linear relation between seismic observation data and a speed model; Establishing an objective function by adopting a least square method, and further obtaining an underground model; determining the initial point spread function from the subsurface model; Wherein the objective function is: (1) wherein d represents seismic data, L represents a seismic wave propagation operator, and m represents a subsurface model; Wherein, the underground model is: (2) Wherein, the For the estimated subsurface model, H is the Hessian matrix; Wherein the initial point spread function is: (3) Wherein, the As a function of the green's function, Representing a wavelet of a seismic source, Representing green's function Is a companion operator to; the constraint parameters comprise a local neighborhood range, a background speed, an illumination angle range in an imaging observation system in the neighborhood range and imaging wavelets; wherein the final point spread function is: (4) Wherein, the Representing far field imaging wavelets corresponding to subsurface imaging points at high frequency approximations, A constraint function representing the illumination angle range, x 0 is the imaging point of the neighborhood, and v (x 0 ) is the velocity in the neighborhood.
2. A point spread function calculating apparatus, comprising: the initial module is used for establishing an initial point diffusion function; the constraint module is used for setting constraint parameters; The calculation module is used for obtaining a final point spread function according to the initial point spread function and the constraint parameter and calculating and solving; wherein establishing the initial point spread function includes: establishing a linear relation between seismic observation data and a speed model; Establishing an objective function by adopting a least square method, and further obtaining an underground model; determining the initial point spread function from the subsurface model; Wherein the objective function is: (1) wherein d represents seismic data, L represents a seismic wave propagation operator, and m represents a subsurface model; Wherein, the underground model is: (2) Wherein, the For the estimated subsurface model, H is the Hessian matrix; Wherein the initial point spread function is: (3) Wherein, the As a function of the green's function, Representing a wavelet of a seismic source, Representing green's function Is a companion operator to; the constraint parameters comprise a local neighborhood range, a background speed, an illumination angle range in an imaging observation system in the neighborhood range and imaging wavelets; wherein the final point spread function is: (4) Wherein, the Representing far field imaging wavelets corresponding to subsurface imaging points at high frequency approximations, A constraint function representing the illumination angle range, x 0 is the imaging point of the neighborhood, and v (x 0 ) is the velocity in the neighborhood.
3. An electronic device, the electronic device comprising: A memory storing executable instructions; a processor executing the executable instructions in the memory to implement the point spread function calculation method of claim 1.
4. A computer readable storage medium, characterized in that the computer readable storage medium stores a computer program which, when executed by a processor, implements the point spread function calculation method of claim 1.

Description

Point spread function calculation method, point spread function calculation device, electronic equipment and medium Technical Field The invention relates to the technical field of geophysical exploration, in particular to a point spread function calculation method, a point spread function calculation device, electronic equipment and a point spread function medium. Background The progressive refinement of the targets of hydrocarbon exploration places higher demands on the high resolution processing of seismic data. Conventional resolution enhancement processing techniques, such as deconvolution, anti-Q filtering, spectrum prolongation, etc., are typically based on one-dimensional assumptions, do not take into account the effect of real wave field propagation, do not fully take into account the fidelity of the processing, and do not enhance lateral resolution. In fact, high resolution imaging depends on the resolution of the spatial wavelet over the imaging profile. The higher the spatial wavelet resolution, the higher the corresponding imaging resolution. And the resolution of the spatial wavelet is determined by the Hessian matrix (model resolution matrix) of the imaging system. The Hessian matrix corresponds to the response function of a linear imaging system, consisting of a Point Spread Function (PSF) per point of the system (PSF for one point per row). For imaging systems, the PSF at each point of the model is different, so that the Hessian matrix acts on subsurface reflection coefficients equivalent to defining a space-variant convolution process. The least squares offset result mathematically differs from the normal offset result by the inverse of the Hessian matrix. Therefore, the least square imaging solving process is the inverse solving process of the Hessian matrix, and is also a process for eliminating adverse effects on the conventional offset imaging result caused by a seismic data observation system, underground medium speed change, seismic wave geometric diffusion, an offset imaging formula and the like. Because the Hessian matrix is very bulky, it is difficult to compute and store, and many scholars approximate it. The traditional Hessian matrix approximation method still takes implicit solution as a main principle, approximates the inverse of the Hessian matrix in a data domain in an inversion iteration mode, and realizes the successive approximation from an offset result to a real reflection coefficient. While explicit computation of the Hessian matrix in the imaging domain is easier to implement for object-oriented least squares offset imaging, huge computation and memory are still the biggest problems faced at present. In recent years, a point spread function solving algorithm based on a discrete point model is proposed for explicitly and approximately solving the Hessian, and the algorithm improves the solving efficiency to a certain extent, but still solves the problem based on a wave equation, so that the problem of huge calculation amount still faces to a large extent. Therefore, it is necessary to develop a point spread function calculation method, a device, an electronic apparatus and a medium based on local neighborhood high frequency approximation. The information disclosed in the background section of the invention is only for enhancement of understanding of the general background of the invention and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person skilled in the art. Disclosure of Invention The invention provides a point spread function calculation method, a device, electronic equipment and a medium, which are used for realizing the rapid solution of a point spread function based on local neighborhood and high-frequency approximation, so as to better serve the property of a Hessian matrix and the research of seismic data Gao Weizheng, provide a new technical idea for the high-resolution processing of high-precision and high-efficiency seismic data in the future and provide powerful technical support for practical production and application. In a first aspect, an embodiment of the present disclosure provides a method for calculating a point spread function, including: Establishing an initial point spread function; setting constraint parameters; and obtaining a final point spread function according to the initial point spread function and the constraint parameter, and calculating and solving. Preferably, establishing the diffusion function comprises: establishing a linear relation between seismic observation data and a speed model; Establishing an objective function by adopting a least square method, and further obtaining an underground model; The initial point spread function is determined from the subsurface model. Preferably, the objective function is: Q(m)=||Lm-d||2 (1) where d represents seismic data, L represents a seismic wave propagation operator, and m represents a subsurface model.