CN108957554B - Seismic inversion method in geophysical exploration - Google Patents

Abstract

The invention discloses a seismic inversion method in geophysical exploration, which comprises the steps of obtaining logging data in a seismic work area, and collecting and editing force and moment data of corresponding points of each seismic layer; layer-by-layer interpretation is carried out on each layer of seismic data, and interpretation result data are gridded to obtain a geological framework model based on depth; determining a logging curve according to the determined geological framework model, and performing fractal random interpolation on a fractal parameter body of the logging curve to obtain an initial logging curve; and determining an actual logging curve according to the seismic data, correcting the initial logging curve according to the comparison result, and determining a seismic inversion model according to the corrected curve. The method comprises the steps of acquiring logging data in a seismic work area, collecting and editing force and moment data of corresponding points of each seismic layer, introducing various parameters of force and moment, establishing an exploration model by combining a specific soil environment, and determining a final inversion model by correcting a final value.

Description

Seismic inversion method in geophysical exploration

Technical Field

The invention relates to the technical field of seismic inversion, in particular to a seismic inversion method in geophysical exploration.

Background

In the oil and gas industry, seismic inversion is used as a core technology for reservoir prediction, the high-resolution advantage in the longitudinal direction of logging and the dense sampling advantage in the transverse direction of earthquake can be closely combined, characteristic information such as lithology, physical property and the like in the transverse direction of a stratum can be estimated, and the method is widely applied to various stages of oil exploration and development at present. However, the conventional time domain seismic inversion method has poor identification capability for thin reservoirs, and a great deal of valuable high-level information is lost in the process of converting the well logging curve from a depth domain to a time domain, so that the method has great limitations.

In recent years, the most common practice for seismic inversion reservoir prediction has been depth domain seismic data inversion based on convolution models. The method has the first problem of wavelet acquisition, namely, the time domain wavelet is converted into the depth domain wavelet through time-depth conversion, and then the convolution effect of the wavelet is eliminated according to the convolution and inverse wavelet of the seismic channel, so that an approximate stratum reflection coefficient series is obtained for inverting a high-resolution stratum type section.

The logging constraint seismic inversion technology plays an important role in lithology identification and reservoir prediction, the seismic data with intensive spatial sampling and the logging data with high vertical resolution are combined, the interface reflection type seismic data are inverted to reflect the wave impedance data volume of the stratum, the identification precision of the reservoir is improved, the thickness change condition of the reservoir can be more finely described, particularly, in the aspect of thin reservoir identification, the resolution is improved through logging constraint, and thin layers which cannot be identified by conventional earthquakes can be identified.

The existing inversion method is mainly based on an amplitude inversion method, namely the change of the seismic amplitude strength is considered to reflect the spatial change of a reservoir stratum, most of the inversion methods are based on a model, the building process of the model is mainly based on the spatial interpolation and extrapolation of logging attributes under the constraint of a seismic interpretation horizon, and the change condition of a geologic body in the transverse direction is not considered in the whole modeling process, so that although the model-based inversion method can improve the identification capability of a thin layer, errors exist in the identification of a spatial pinch-off point of the reservoir stratum.

Therefore, the seismic inversion method in the prior art carries out inversion according to the model and waveform identification, and does not consider the actual physical factors and stress factors.

Disclosure of Invention

The invention aims to provide a seismic inversion method in geophysical exploration, which aims to solve the technical problem that stress factors are not considered in a model.

In order to achieve the purpose, the invention provides the following technical scheme: the method comprises the following steps:

step a, acquiring logging data in a seismic work area, and acquiring and editing force and moment data of corresponding points of each seismic layer;

b, performing layer-by-layer interpretation on each layer of seismic data, and gridding interpretation result data to obtain a geological frame model based on depth;

step c, determining a logging curve according to the determined geological framework model, and performing fractal random interpolation on a fractal parameter body of the logging curve to obtain an initial logging curve;

d, determining an actual logging curve according to the seismic data, correcting the initial logging curve according to a comparison result, and determining a seismic inversion model according to the corrected curve;

in the step a, a plurality of sampling points of each seismic layer are established according to the following model: one of the sampling points is modeled as:

wherein,

the zero offset values of the detection force and the moment of the seismic layer sampling point,

is the gravity and gravity moment of the sampling point,

and

external force and external moment applied to the sampling point are included, wherein the moment is the moment of the sampling point relative to the central line of the seismic layer cavity;

setting the vector from the origin of the central coordinate system of the seismic layer cavity to the centroid of the sampling point as

Then

The detailed expression of equation (1) in the survey coordinate system based on the center of the seismic layer void is:

wherein,

is a transformation matrix from the world coordinate system to the central coordinate system of the seismic interval cavity, determined by the initial position of the bytes, x_c,y_c,z_c；

When the sampling point is positioned on the upper surface layer of the seismic layer cavity:

detecting the existence relation between force and external force:

F₁representing the detected value of force, m, at the position of the upper layer_loadRepresenting the real-time gravity of the geology itself, e.g. the weight of the sand, gravel, sand itself, f_extExternal force, F_0x,F_0y,F_0zIndicating the resultant force, T, in each direction of the position_0x,T_0y,T_0zIndicating the resultant moment, x, in each direction of the position_c,y_c,z_cRepresenting the coordinates at the location, enabling the location of the sampling point to be uniquely determined;

when the sampling point is positioned on the lower surface of the seismic layer cavity:

detecting the existence relation between force and external force:

when the sampling point is positioned in the middle layer of the seismic layer cavity:

detecting the existence relation between force and external force:

through collecting the force detection measurement data in the motion process, the force detection zero offset and the self load weight parameter to be identified can be synchronously obtained:

finally determine x by the above equation (9)_c,y_c,z_cAnd the coordinates of the positions can uniquely determine the stress and the moment of the sampling points and the positions of the corresponding sampling points.

Further, in the step d, the correction is performed according to the following formula (10) in accordance with the correction of each selected point;

i_m＝ρ×i₀ (10)

wherein i_mRepresents the instantaneous value of the sample point after correction, rho represents the correction factor, i₀Representing the instantaneous value of the sample point; the correction coefficient p is calculated by the following formula (11),

where ρ represents a correction coefficient, i₀₁And i₀₂The instantaneous values of two points at the position of deviation when the deviation occurs are shown, N represents the number of sampling times, and k represents the sampling sequence.

Furthermore, in the step b, a plurality of sets of reflection positions are interpreted and tracked for the seismic data, the quality of the interpreted positions is checked and modified, and the positions are subjected to full three-dimensional tracking interpretation, so that the position interpretation result is fine, and each earthquake has a fine position interpretation result.

Further, in the step c, calculating a fractal parameter of a logging curve model of a single well, and converting the fractal parameter of each seismic channel into a fractal parameter body of the logging curve; respectively calculating fractal parameters of a logging curve depth domain model of a single well, calibrating the seismic fractal parameters of each seismic channel into a fractal parameter body of the logging curve by methods of normalization, proportion weighting and the like, thus obtaining the fractal parameters of each seismic channel of the logging curve, and then calculating the random component funine of fractal random interpolation of each seismic channel according to the following formula:

k is a calibration coefficient, H is a fractal dimension parameter reflecting details and roughness of the hierarchical domain post-stack seismic data, sigma is the variance of the hierarchical domain post-stack seismic data normal distribution, G is a Gauss random variable obeying the standard normal distribution, Delta X is the distance between a known point and an interpolation point, and a is a one-dimensional coordinate of the interpolation point.

Compared with the prior art, the invention has the beneficial effects that: a seismic inversion method in geophysical exploration comprises the steps of acquiring logging data in a seismic work area, collecting and editing force and moment data of corresponding points of each seismic layer, introducing various parameters of force and moment, establishing an exploration model by combining a specific soil environment, and determining a final inversion model by correcting a final value.

Further, the calibration method of the embodiment of the invention comprises the following steps: the identification time is short, only force detection data need to be collected, and a simulated exploration-based earthquake stress model is determined by adopting detection force, moment, stress position and combining external force and gravity information of different soil textures.

The invention ensures the stability of the final value by adopting a correction mode and comparing and correcting the empirical value and the real-time value.

Drawings

FIG. 1 is a logic flow diagram of a method of seismic inversion in geophysical exploration according to the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Referring to fig. 1, an embodiment of the invention provides a seismic inversion method in geophysical exploration, which specifically includes:

step a, acquiring logging data in a seismic work area, and acquiring and editing force and moment data of corresponding points of each seismic layer;

b, performing layer-by-layer interpretation on each layer of seismic data, and gridding interpretation result data to obtain a geological frame model based on depth;

step c, determining a logging curve according to the determined geological framework model, and performing fractal random interpolation on a fractal parameter body of the logging curve to obtain an initial logging curve;

and d, determining an actual logging curve according to the seismic data, correcting the initial logging curve according to the comparison result, and determining a seismic inversion model according to the corrected curve.

In this embodiment, in step a, a plurality of sampling points of each seismic layer are established according to the following model: one of the sampling points is modeled as:

wherein,

the zero offset values of the detection force and the moment of the seismic layer sampling point,

is the gravity and gravity moment of the sampling point,

and

the external force and external moment applied to the sampling point are shown, wherein the moment is the moment of the sampling point relative to the central line of the seismic layer cavity.

Setting the vector from the origin of the central coordinate system of the seismic layer cavity to the centroid of the sampling point as

Then

The detailed expression of equation (1) in the survey coordinate system based on the center of the seismic layer void is:

wherein,

is a transformation matrix from the world coordinate system to the central coordinate system of the seismic interval cavity, determined by the initial position of the bytes, x_c,y_c,z_c。

When the sampling point is positioned on the upper surface layer of the seismic layer cavity:

detecting the existence relation between force and external force:

F₁representing the detected value of force, m, at the position of the upper layer_loadRepresenting the real-time gravity of the geology itself, e.g. the weight of the sand, gravel, sand itself, f_extExternal force, F_0x,F_0y,F_0zIndicating the resultant force, T, in each direction of the position_0x,T_0y,T_0zIndicating the resultant moment, x, in each direction of the position_c,y_c,z_cRepresenting the coordinates at that location, the location of the sample point can be uniquely determined.

When the sampling point is positioned on the lower surface of the seismic layer cavity:

detecting the existence relation between force and external force:

when the sampling point is positioned in the middle layer of the seismic layer cavity:

detecting the existence relation between force and external force:

through collecting the force detection measurement data in the motion process, the force detection zero offset and the self load weight parameter to be identified can be synchronously obtained:

finally determine x by the above equation (9)_c,y_c,z_cAnd the coordinates of the positions can uniquely determine the stress and the moment of the sampling points and the positions of the corresponding sampling points.

The calibration method of the embodiment of the invention comprises the following steps: the identification time is short, only force detection data need to be collected, and a simulated exploration-based earthquake stress model is determined by adopting detection force, moment, stress position and combining external force and gravity information of different soil textures.

B, performing layer-by-layer interpretation on each layer of seismic data, and gridding interpretation result data to obtain a geological frame model based on depth; the method comprises the steps of interpreting and tracking a plurality of sets of reflection positions on seismic data, checking and modifying the quality of the interpreted positions, and performing full three-dimensional tracking interpretation on the positions, so that the position interpretation result is fine, each earthquake has a fine position interpretation result, and particularly some unconformities and standard reflection layers need to be interpreted.

Specifically, on the basis of horizon interpretation, well logging curves of a target interval are analyzed in combination with well drilling layering, curve types such as sound waves, density, gamma, porosity, resistivity, lithology and the like are comprehensively analyzed, electrical characteristics, elasticity and radioactivity characteristics of a reservoir are analyzed, lithology and oil-gas content of the reservoir are analyzed, and geological deposition characteristics of the target interval are analyzed, which is not repeated.

And c, determining a logging curve according to the determined geological framework model, and performing fractal random interpolation on a fractal parameter body of the logging curve to obtain an initial logging curve.

Specifically, there may be various embodiments for calculating the fractal parameters of the logging curve model of a single well and converting the fractal parameters of each seismic trace into the fractal parameter body of the logging curve. Respectively calculating fractal parameters of a logging curve depth domain model of a single well, calibrating the seismic fractal parameters of each seismic channel into a fractal parameter body of the logging curve by methods of normalization, proportion weighting and the like, thus obtaining the fractal parameters of each seismic channel of the logging curve, and then calculating the random component funine of fractal random interpolation of each seismic channel according to the following formula:

k is a calibration coefficient, H is a fractal dimension parameter reflecting details and roughness of the hierarchical domain post-stack seismic data, sigma is the variance of the hierarchical domain post-stack seismic data normal distribution, G is a Gauss random variable obeying the standard normal distribution, Delta X is the distance between a known point and an interpolation point, and a is a one-dimensional coordinate of the interpolation point.

There are various ways to find the variance σ and the fractal parameter H. For example, in one embodiment: on the level domain post-stack seismic data volume or the derivative attribute volume thereof, converting one-dimensional seismic channel data into volume data in the longitudinal direction, sequentially establishing a seismic data cube, namely a subvolume, by taking a certain point in the longitudinal direction as a center and a certain distance in the transverse direction as a radius in a geological frame model constraint range, and normalizing the subvolume data.

And calculating the fractal parameters and the variance of each subvolume, so that the fractal parameters and the variance of the seismic data of each seismic channel can be calculated. In the same way, the fractal dimension parameter and variance of the logging curve to be inverted can be calculated, and further the fractal dimension parameter ratio and variance ratio of the logging curve and the seismic data or the derivative attribute body thereof on the well point in a certain interval range can be calculated.

And d, determining an actual logging curve according to the seismic data, correcting the initial logging curve according to the comparison result, and determining a seismic inversion model according to the corrected curve.

Specifically, in the present embodiment, each point selected is corrected according to the following formula (10);

i_m＝ρ×i₀ (10)

wherein i_mRepresents the instantaneous value of the sample point after correction, rho represents the correction factor, i₀Representing the instantaneous value of the sample point; the correction coefficient p is calculated by the following formula (11),

where ρ represents a correction coefficient, i₀₁And i₀₂The instantaneous values of two points at the position of deviation when the deviation occurs are shown, N represents the number of sampling times, and k represents the sampling sequence.

The use of the correction mode ensures the stability of the final value by comparing and correcting the empirical value and the real-time value.

In the specific implementation process, the first-stage reactor,

and respectively extracting each well of all the inversion wells as a verification well, performing depth domain reservoir seismic inversion on the unextracted inversion wells, comparing the inversion result of the unextracted inversion wells with the logging curve of the verification well, and calculating the inversion error between the inversion result of the unextracted inversion wells and the logging curve of the verification well.

Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.

Claims (4)

1. A method of seismic inversion in geophysical exploration, comprising:

step a, acquiring logging data in a seismic work area, and acquiring and editing force and moment data of corresponding points of each seismic layer;

b, performing horizon interpretation on each layer of seismic data, and gridding interpretation result data to obtain a geological frame model based on depth;

step c, determining a logging curve according to the determined geological framework model, and performing fractal random interpolation on a fractal parameter body of the logging curve to obtain an initial logging curve;

d, determining an actual logging curve according to the seismic data, correcting the initial logging curve according to a comparison result, and determining a seismic inversion model according to the corrected curve;

in the step a, a plurality of sampling points of each seismic layer are established according to the following model: the establishing model of any sampling point is as follows:

wherein,

the zero offset values of the detected force and moment of the sampling points,

is the gravity and gravity moment of the sampling point,

and

external force and external moment applied to the sampling point; the moment is the moment of the sampling point relative to the central line of the seismic layer cavity;

the vector from the origin of the measurement coordinate system at the center of the seismic layer cavity to the centroid of the sampling point is

Wherein x is_c,y_c,z_cRepresenting coordinates; then

The detailed expression of equation (1) in the survey coordinate system based on the center of the seismic layer void is:

wherein,

a transformation matrix which is a measurement coordinate system from the world coordinate system to the center of the seismic interval cavity;

when the sampling point is positioned on the upper surface layer of the seismic layer cavity:

detecting the existence relation between force and external force:

F₁representing the detected value of force, m, at the position of the upper layer_loadRepresenting the real-time gravity of the geological formation itself, f_extDenotes an external force, F_0x,F_0y,F_0zIndicating the resultant force, T, in each direction_0x,T_0y,T_0zRepresenting resultant moments in various directions;

when the sampling point is located on the lower surface of the seismic layer cavity:

detecting the existence relation between force and external force:

when the sampling point is positioned in the middle layer of the seismic layer cavity:

detecting the existence relation between force and external force:

by collecting force detection measurement data in the movement process, force detection zero offset and self load weight parameters to be identified are synchronously obtained:

finally determine x by the above equation (9)_c,y_c,z_cAnd acquiring the stress and the moment of the sampling point and the position of the corresponding sampling point by the coordinates at the position.

2. The seismic inversion method in geophysical exploration according to claim 1, wherein in said step d, each point selected is corrected according to the following formula (10);

i_m＝ρ×i₀ (10)

wherein i_mRepresents the instantaneous value of the sample point after correction, rho represents the correction factor, i₀Representing the instantaneous value of the sample point; the correction coefficient p is calculated by the following formula (11),

where ρ represents a correction coefficient, i₀₁And i₀₂The instantaneous values of two points at the position of deviation when the deviation occurs are shown, N represents the number of sampling times, and k represents the sampling sequence.

3. The method of seismic inversion in geophysical exploration according to claim 1, wherein in step b, interpretation and tracking of sets of reflection horizons are performed on the seismic data; and checking and modifying the quality of the interpreted horizon, and performing full three-dimensional tracking interpretation on the horizon to ensure that the horizon interpretation result is fine, so that each seismic channel has a fine horizon interpretation result.

4. The seismic inversion method in geophysical exploration according to claim 1, wherein in step c, a single-well log model fractal parameter is calculated, and the fractal parameter of each seismic trace is converted into a log fractal parameter body; respectively calculating fractal parameters of a logging curve depth domain model of a single well, calibrating the seismic fractal parameters of each seismic channel into a fractal parameter body of the logging curve by a normalization and proportion weighting method, further obtaining the fractal parameters of each seismic channel of the logging curve, and then calculating the random component funine of fractal random interpolation of each seismic channel according to the following formula:

k is a calibration coefficient, H is a fractal dimension parameter reflecting details and roughness of the hierarchical domain post-stack seismic data, sigma is the variance of the hierarchical domain post-stack seismic data normal distribution, G is a Gauss random variable obeying the standard normal distribution, Delta X is the distance between a known point and an interpolation point, and a is a one-dimensional coordinate of the interpolation point.

Images