A well control constraint-based surface grid model conversion and correction method

Abstract

The application provides a well control constraint-based surface grid model conversion and correction method and belongs to the technical field of oil seismic exploration. The application converts network model data surfaces from continuous grid model samples, calculates micro-logging point grid model information, and obtains difference information of discrete micro-logging and grid model points. The data surfaces include thickness surfaces and equivalent velocity surfaces. According to the difference information of discrete micro-logging and grid model points, difference information distribution is carried out based on well control constraints, and the difference information corresponding to each grid model point is obtained. The grid model is corrected based on a similarity coefficient and the difference information. The application solves the problem that micro-logging layered surface results and vertical continuous grid sample data of the grid model cannot be directly compared, solves the problem that the discrete distributed micro-logging is difficult to correct the grid model body, and makes the corrected grid model fully matched with the micro-logging while having strong similarity with the grid model before correction.

Description

Technical Field

[0001] This invention relates to the field of petroleum seismic exploration technology, and in particular to a method for converting and correcting surface grid models based on well control constraints. Background Technology

[0002] Surface mesh models typically refer to near-surface tomography models. Tomography, in this context, means imagining an indivisible object as a series of thin slices and providing an image of the object on each slice. Seismic tomography refers to the reconstruction of images of the Earth's internal structure using various seismic wavefield information, such as travel time or amplitude variations. Near-surface tomography models are specifically used to reconstruct models of the near-surface internal structure.

[0003] Micro-logging is currently the most direct and accurate method for surface surveying. It determines the velocity and thickness of the near-surface layer by sequentially drilling at different depths within the well. Generally, the near-surface layer has a multi-layered structure, which can be roughly divided into a low-velocity layer, a decreasing-velocity layer, and a high-velocity layer. Normal drilling should penetrate to the top plate of the high-velocity layer without needing to penetrate the bottom plate. Therefore, only the velocity of the last layer is recorded, not its thickness.

[0004] Near-surface tomography models are surface structure inversion models completed using ray tracing methods, relying on first arrival time data and initial model data. The model results often do not match the measured surface survey data such as micrologging, raising doubts about the reliability of the results. Especially when processing contiguous seismic data, near-surface tomography models inverted separately for multiple blocks are difficult to close, which brings difficulties to subsequent static correction closure and ultimately affects the quality of seismic data imaging.

[0005] Existing methods mainly use the surface model of micrologging interpolation as the initial surface model for inversion. Even with some constraint measures, this method is difficult to achieve closure between the inverted tomographic model and the micrologging results, and it is difficult to fundamentally solve the problem of closure of the continuous tomographic model.

[0006] The existing technology has the following shortcomings:

[0007] 1. It is impossible to achieve closure between the inverted tomographic model and the micrologging results;

[0008] 2. It cannot fundamentally solve the problem of closure in continuous chromatography models. Summary of the Invention

[0009] To address the problems existing in the prior art, this invention provides a method for surface grid model conversion and correction based on well control constraints. The method converts a network model data surface from continuous grid model samples, calculates the grid model information of micrologging points, and obtains the difference information between discrete micrologging points and grid model points. The data surface includes a thickness surface and an equivalent velocity surface. Based on the difference information between discrete micrologging points and grid model points, the difference information is allocated according to well control constraints to calculate the difference information corresponding to each grid model point. The grid model is then corrected based on the similarity coefficient and the difference information. This invention solves the problem that micrologging layered surface results cannot be directly compared with the vertical continuous grid sample data of the grid model, and also solves the problem that discretely distributed micrologging is difficult to correct the grid model volume. This ensures that the corrected grid model perfectly matches the micrologging results while also maintaining strong similarity to the original grid model.

[0010] This invention provides a method for converting and correcting surface mesh models based on well control constraints, comprising the following steps:

[0011] The network model data surface is converted from the continuous grid model sample, and the grid model information of the micro-logging point is calculated to obtain the difference information between the discrete micro-logging point and the grid model point; the data surface includes: thickness surface and equivalent velocity surface;

[0012] Based on the difference information between discrete micro-logging and grid model points, the difference information is allocated according to well control constraints, and the difference information corresponding to each grid model point is calculated.

[0013] The mesh model is corrected based on similarity coefficient and difference information.

[0014] Preferably, the process of converting the network model data surface from the continuous grid model sample and calculating the micro-logging point grid model information specifically includes:

[0015] The model thickness and equivalent velocity of the continuous grid are calculated based on the surface elevation and the top interface elevation of the high-velocity grid sample, forming the model thickness surface and the equivalent velocity surface.

[0016] The surface velocity model consists of a low-velocity layer and a deceleration layer. The first layer of the surface velocity model is called the low-velocity layer, and all layers from the second layer onwards are called deceleration layers. The high-velocity top interface is the top surface of the first medium layer that can refract and propagate seismic waves after the shallow surface low-velocity medium is stripped away. The propagation speed of seismic waves in this medium layer is the high-velocity top velocity. The low-deceleration zone is the entire surface model medium.

[0017] Based on the micrologging coordinates, the mesh model thickness and equivalent velocity results corresponding to the micrologging points are extracted.

[0018] Preferably, the process of obtaining the difference information between discrete micro-logging and grid model points includes the following steps:

[0019] The formula for calculating the total thickness and equivalent velocity of the micrologging results is as follows:

[0020]

[0021]

[0022]

[0023] in,

[0024] m represents the total number of floors;

[0025] h i The thickness of the i-th layer representing the results of micrologging;

[0026] v i Represents the velocity of the i-th layer;

[0027] h 总 It is the total thickness;

[0028] t 总 This is the total time;

[0029] v 平 It is the equivalent velocity in the low-velocity zone of the actual micro-logging.

[0030] The low-velocity band thickness surface, the equivalent velocity surface, and the high-velocity top boundary velocity surface are extracted from the mesh model.

[0031] Based on the coordinates of the microlog, the grid model data corresponding to the microlog location is extracted from the low-velocity zone thickness surface, the equivalent velocity surface, and the high-velocity top boundary velocity surface.

[0032] The difference and quotient of the total thickness of the micrologging results and the thickness of the corresponding grid model at the point are respectively taken as the thickness difference Δh and the thickness change rate Δh. p ;

[0033] The obtained Δh and Δh p The difference information belonging to this micro-logging point;

[0034] The difference and quotient of the equivalent velocity from the micrologging results and the equivalent velocity from the corresponding grid model at the point are respectively taken as the equivalent velocity difference ΔV. A and equivalent rate of change of velocity ΔV Ap The obtained ΔV A and ΔV Ap The difference information belonging to this micro-logging point;

[0035] The difference and quotient between the high-velocity top velocity of the micrologging results and the high-velocity top velocity of the corresponding grid model at the point are respectively taken as the high-velocity top velocity difference ΔV. T and the rate of change of high-speed top velocity ΔVTp The obtained ΔV T With ΔV Tp It refers to the difference information at that micro-logging point.

[0036] Preferably, the difference information between discrete micrologging and grid model points is used to allocate the difference information based on well control constraints. The difference information corresponding to each grid model point is obtained by using an adaptive optimization interpolation method based on inverse distance weighted interpolation to allocate the difference information of micrologging points to each grid model point, thus completing the allocation of difference information for well control constraints based on micrologging results.

[0037] Preferably, the adaptive optimization interpolation method based on inverse distance weighted interpolation is as follows: on the basis of inverse distance weighted interpolation, the angular relationship information between the interpolation target point and each control point is used as the interpolation control weight.

[0038] Preferably, inverse distance weighted interpolation is a weighted average interpolation method that uses the reciprocal of the distance as the interpolation coefficient.

[0039] Preferably, the adaptive optimization interpolation method is as follows:

[0040] Let the interpolation target point be a "light source", and each control point have a circular area that is opaque in the center and gradually becomes transparent towards the edges. Then the part of the circular area facing away from the light source has a "shadow".

[0041] The formula for the angular relationship weight A between the control points in the "shadow," including those that can see all "light sources," those that cannot see any "light sources," and those that can see some "light sources," is as follows:

[0042]

[0043] in,

[0044] θ1 is the angle at which the control point, which is not visible as a "light source," enters the "shadow" of the interpolation target point;

[0045] θ2 is the angle at which the control point, which is completely invisible from the light source, enters the "shadow" of the interpolation target point;

[0046] α is the angle at which the control points that can see all the "light sources" enter the "shadow" of the interpolation target point.

[0047] Preferably, the adaptive optimization interpolation method of inverse distance weighted interpolation is as follows:

[0048] Given n points, the inverse distance weighted interpolation function is:

[0049]

[0050] By incorporating the angular relationship weights between the control points and the target points into the inverse distance weighted interpolation formula, we obtain an adaptive optimization interpolation formula based on inverse distance weighted interpolation:

[0051]

[0052] in,

[0053] (x i ,y i Let be the planar coordinates of the i-th point, and let i be the label of the i-th point;

[0054] j is the label of the j-th point, j = 1, 2, ..., n;

[0055] z i Let i be the value of the i-th point, where i = 1, 2, ..., n;

[0056] d j It is the point (x,y) to (x j ,y j ) horizontal distance of points;

[0057] p is a weighted power exponent greater than 0;

[0058] A j Let be the angle relation weight of the j-th point.

[0059] Preferably, the mesh model is corrected based on similarity coefficient and difference information, including:

[0060] The similarity coefficient K is defined as the degree of similarity between the constrained mesh model and the unconstrained mesh model, and its range is [-1, 1].

[0061] When K equals 0, it means that the shape of the constrained mesh model is unrelated to that of the unconstrained mesh model.

[0062] A K value of 1 indicates a strong correlation; a K value greater than 0 and less than 1 indicates a weak correlation; a K value of -1 indicates an inverse correlation.

[0063] The similarity coefficient K is used to control the amount of correction of the mesh model by the difference information, and finally the mesh model correction is completed.

[0064] Preferably, it also includes at least one of the following in the network model after correction based on model grid point difference information and similarity coefficient control: high-speed top boundary morphology, low-speed reduction zone grid velocity, and grid velocity below the high-speed top boundary.

[0065] Preferably, the high-speed top boundary morphology of the mesh model after correction based on model mesh point difference information and similarity coefficient control includes:

[0066] Substituting the difference in low-velocity zone thickness and the rate of change in thickness from the model grid point difference information into the similarity coefficient calculation formula, we obtain the low-velocity zone thickness after the model grid point correction.

[0067] Subtract the thickness of the low-velocity zone from the surface elevation at that point to obtain the corrected elevation of the high-speed top interface.

[0068] The high-speed top boundary morphology is adjusted by adjusting the similarity coefficient.

[0069] Preferably, the mesh velocity in the low-deceleration zone of the mesh model is corrected based on the model mesh point difference information and similarity coefficient, including:

[0070] Substituting the equivalent velocity difference and equivalent velocity change rate in the low-speed-deceleration zone from the model grid point difference information into the similarity coefficient calculation formula, we obtain the equivalent velocity of the low-speed-deceleration zone after the model grid point correction.

[0071] The ratio of the change in the thickness of the low-speed-deceleration zone is obtained by quoting the corrected thickness of the low-speed-deceleration zone with the original thickness.

[0072] Calculate the average velocity for each corrected grid corresponding to the original grid and fill it into the corresponding corrected grid.

[0073] Then, calculate the rate of change of the equivalent velocity based on the equivalent velocity of the low deceleration zone before and after the correction;

[0074] All mesh velocities in the low-decay zone of the new mesh model are multiplied by the equivalent velocity change rate to complete the velocity correction of the low-decay zone of the mesh model.

[0075] Preferably, the mesh velocity below the high-speed top boundary of the mesh model is corrected based on the model mesh point difference information and similarity coefficient, including:

[0076] Substituting the high-speed top velocity difference and the high-speed top velocity change rate from the model grid point difference information into the similarity coefficient calculation formula, we obtain the high-speed top velocity after the model grid point correction;

[0077] Find the mesh whose velocity is closest to the high-speed top velocity of the modified mesh model in the mesh model before modification. Using this mesh as the starting point, stitch this mesh and all subsequent meshes together in sequence below the low-velocity zone mesh of the modified mesh model.

[0078] If the number of grid cells after correction is less than the number before correction, copy the bottommost grid cell and continue stitching until the number of grid cells before correction is the same; if the number of grid cells after correction is more than the number before correction, trim the excess grid cells.

[0079] Preferably, it further includes correcting the mesh model velocity based on the surface delay time, including: if the surface delay time is accurate, correcting the mesh model velocity in the low-deceleration zone region based on the thickness of the low-deceleration zone of the mesh model.

[0080] Preferably, the correction of the mesh model velocity based on surface delay specifically includes:

[0081] First, extract the equivalent velocity surface and thickness surface of the low-velocity zone from the mesh model;

[0082] Then, based on the location of the shot receiver, the thickness of the low velocity reduction zone in the corresponding grid of the shot receiver is extracted from the thickness surface;

[0083] The equivalent velocity of the shot receiver is obtained by dividing the thickness of the low velocity reduction zone in the grid corresponding to the shot receiver by the surface delay of the shot receiver.

[0084] An adaptive optimization interpolation method based on inverse distance weighted interpolation is used to assign the discrete shot-receiver point equivalent velocity to each grid point.

[0085] Divide the equivalent velocity assigned to each grid point through the shot receiver point by the equivalent velocity of the low-velocity zone extracted from the grid model to obtain the rate of change of the equivalent velocity of the low-velocity zone.

[0086] The mesh velocity at each mesh point in the low-deceleration zone is multiplied by the equivalent velocity change rate to correct the mesh model velocity for surface delay.

[0087] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0088] (1) This invention converts complex grid model volume data into two data surfaces: model thickness and equivalent velocity, which solves the problem that the results of micro-logging layered surface layer and the vertical continuous grid sample data of grid model cannot be directly compared. This allows the results of discrete micro-logging to be compared with grid model volume data, providing usable data for subsequent constraint algorithms.

[0089] (2) The present invention uses an adaptive optimization interpolation method based on inverse distance weighted interpolation to distribute the difference information of discrete micro-logging to each grid model point. It can "tailor-make" a control point that plays a role for each grid model point, minimize the error caused by the interpolation algorithm to the result, and solve the problem that it is difficult to correct the grid model body of discrete micro-logging.

[0090] (3) This invention uses similarity coefficients to process the difference information of each grid point to complete the grid model correction. It can not only maintain the high-speed top shape similar to the grid model before correction, the velocity change law of the low-speed zone similar, and the grid velocity below the top interface consistent, so that the corrected grid model can be completely matched with the micro-logging, but also make the grid models with similar but not closed shapes close by setting different similarity coefficients to constrain the grid model after closing, providing strong support for subsequent static correction calculation of the closed grid. Attached Figure Description

[0091] Figure 1 Flow chart of the surface grid model conversion and correction method based on well control constraint for an embodiment of the present invention;

[0092] Figure 2 Flow chart of the surface grid model conversion and correction method based on well control constraint for another embodiment of the present invention;

[0093] Figure 3 Flow chart of the surface grid model conversion and correction method based on well control constraint for another embodiment of the present invention;

[0094] Figure 4 Schematic diagram of the continuous grid model for an embodiment of the present invention;

[0095] Figure 5 Schematic diagram of the thickness surface converted for an embodiment of the present invention;

[0096] Figure 6 Schematic diagram of the equivalent velocity surface converted for an embodiment of the present invention;

[0097] Figure 7 Schematic diagram of the adaptive optimization principle provided by an embodiment of the present invention; the unshaded part in the figure is the part that is completely transparent, that is, except for the partially transparent and completely opaque parts, it is all transparent;

[0098] Figure 8 (a)-(d) Schematic diagrams of the role of the correlation coefficient provided by an embodiment of the present invention; where (a) is K = 0, uncorrelated, (b) is 0 < K < 1, weakly correlated, (c) is K = 1, strongly correlated, and (d) is K = -1, inversely correlated;

[0099] Figure 9 Schematic diagram of correcting the grid velocity of the low-velocity reduction zone provided by an embodiment of the present invention; the first column shows the grid situation before correction, the second column represents the reduction of the low-velocity reduction zone grid after correction, and the third column represents the increase of the low-velocity reduction zone grid after correction;

[0100] Figure 10 Schematic diagram of correcting the grid velocity below the high-velocity top boundary provided by an embodiment of the present invention; the first column is the network model before correction, the second column shows the case where the number of grids after correction is more than that before correction, and the redundant grids are trimmed; the third column shows the case where the number of grids after correction is less than that before correction, and the bottommost grid is copied and continued to be spliced until the number of grids is the same as that before correction.

[0101] Figure 11 Flow schematic diagram of the grid model velocity correction based on accurate surface delay time provided by an embodiment of the present invention;

[0102] Figure 12 This is a schematic diagram of the velocity of a mesh model based on surface delay, provided as an embodiment of the present invention. The first column is the mesh model before correction, and the second column is the mesh model after correction. Detailed Implementation

[0103] The following is in conjunction with the appendix Figure 1-12 The specific embodiments of the present invention will be described in detail below.

[0104] This invention provides a method for converting and correcting surface mesh models based on well control constraints, comprising the following steps:

[0105] The network model data surface is extracted from the continuous grid model samples. The grid model information of the micro-logging points is calculated to obtain the difference information between the discrete micro-logging points and the grid model points. The data surface includes: a thickness surface and an equivalent velocity surface; such as... Figure 4 As shown, the model thickness and equivalent velocity (average velocity) of the continuous mesh are extracted to form the model thickness surface. Figure 5 ), equivalent velocity surface ( Figure 6 )

[0106] Based on the difference information between discrete micro-logging and grid model points, the difference information is allocated according to well control constraints, and the difference information corresponding to each grid model point is calculated.

[0107] The mesh model is corrected based on similarity coefficient and difference information.

[0108] According to a specific embodiment of the present invention, the conversion of network model data surfaces from continuous grid model samples and the calculation of micro-logging point grid model information specifically include:

[0109] The model thickness and equivalent velocity of the continuous grid are calculated based on the surface elevation and the top interface elevation of the high-velocity grid sample, forming the model thickness surface and the equivalent velocity surface.

[0110] The surface velocity model consists of a low-velocity layer and a deceleration layer. The first layer of the surface velocity model is called the low-velocity layer, and all layers from the second layer onwards are called deceleration layers. The high-velocity top interface is the top surface of the first medium layer that can refract and propagate seismic waves after the shallow surface low-velocity medium is stripped away. The propagation speed of seismic waves in this medium layer is the high-velocity top velocity. The low-deceleration zone is the entire surface model medium.

[0111] The low-speed reduction zone refers to the overall surface model medium, and the high-speed top interface is the bottom interface of the surface model, representing the end of the low-speed surface medium.

[0112] Based on the micrologging coordinates, the mesh model thickness and equivalent velocity results corresponding to the micrologging points are extracted.

[0113] According to a specific embodiment of the present invention, the process of obtaining the difference information between discrete micro-logging points and grid model points includes the following steps:

[0114] The formula for calculating the total thickness and equivalent velocity of the micrologging results is as follows:

[0115]

[0116]

[0117]

[0118] in,

[0119] m represents the total number of floors;

[0120] h i The thickness of the i-th layer representing the results of micrologging;

[0121] v i Represents the velocity of the i-th layer;

[0122] h 总 It is the total thickness;

[0123] t 总 This is the total time;

[0124] v 平 It is the equivalent velocity in the low-velocity zone of the actual micro-logging.

[0125] The low-velocity band thickness surface, the equivalent velocity surface, and the high-velocity top boundary velocity surface are extracted from the mesh model.

[0126] Based on the coordinates of the microlog, the grid model data corresponding to the microlog location is extracted from the low-velocity zone thickness surface, the equivalent velocity surface, and the high-velocity top boundary velocity surface.

[0127] The difference and quotient of the total thickness of the micrologging results and the thickness of the corresponding grid model at the point are respectively taken as the thickness difference Δh and the thickness change rate Δh. p ;

[0128] The obtained Δh and Δh p The difference information belonging to this micro-logging point;

[0129] The difference and quotient of the equivalent velocity from the micrologging results and the equivalent velocity from the corresponding grid model at the point are respectively taken as the equivalent velocity difference ΔV. A and equivalent rate of change of velocity ΔV Ap The obtained ΔV A and ΔV Ap The difference information belonging to this micro-logging point;

[0130] The difference and quotient between the high-velocity top velocity of the micrologging results and the high-velocity top velocity of the corresponding grid model at the point are respectively taken as the high-velocity top velocity difference ΔV. T and the rate of change of high-speed top velocity ΔV Tp The obtained ΔV T With ΔV Tp It refers to the difference information at that micro-logging point.

[0131] According to a specific embodiment of the present invention, the difference information between discrete micrologging and grid model points is used to allocate the difference information based on well control constraints. Calculating the difference information corresponding to each grid model point involves using an adaptive optimization interpolation method based on inverse distance weighted interpolation to allocate the difference information of the micrologging points to each grid model point, thus completing the allocation of difference information based on well control constraints using micrologging results. Here, well control constraints refer to using micrologging results to constrain and control the grid model results, ensuring that the surface grid model results calculated through methods such as tomographic inversion match the measured micrologging results.

[0132] According to a specific embodiment of the present invention, the adaptive optimization interpolation method based on inverse distance weighted interpolation is as follows: on the basis of inverse distance weighted interpolation, the angular relationship information between the interpolation target point and each control point is used as the interpolation control weight.

[0133] According to one embodiment of the invention, inverse distance weighted interpolation is a weighted average interpolation method using the reciprocal of the distance as the interpolation coefficient. This method can perform interpolation in an exact or smooth manner. The power parameter controls how the weight coefficients decrease with increasing distance from a grid node. For a larger power, closer data points are given a higher weight share; for a smaller power, the weights are distributed more evenly among the data points.

[0134] According to a specific embodiment of the present invention, the adaptive optimization interpolation method is as follows:

[0135] Let the interpolation target point be a "light source", and each control point have a circular area that is opaque in the center and gradually becomes transparent towards the edges. Then the part of the circular area facing away from the light source has a "shadow".

[0136] The formula for the angular relationship weight A between the control points in the "shadow," including those that can see all "light sources," those that cannot see any "light sources," and those that can see some "light sources," is as follows:

[0137]

[0138] in,

[0139] θ1 is the angle at which the control point, which is not visible as a "light source," enters the "shadow" of the interpolation target point;

[0140] θ2 is the angle at which the control point, which is completely invisible from the light source, enters the "shadow" of the interpolation target point;

[0141] α is the angle at which the control points that can see all the "light sources" enter the "shadow" of the interpolation target point.

[0142] According to a specific embodiment of the present invention, the adaptive optimization interpolation method of inverse distance weighted interpolation is as follows:

[0143] Given n points, the inverse distance weighted interpolation function is:

[0144]

[0145] By incorporating the angular relationship weights between the control points and the target points into the inverse distance weighted interpolation formula, we obtain an adaptive optimization interpolation formula based on inverse distance weighted interpolation:

[0146]

[0147] in,

[0148] (x i ,y i Let be the planar coordinates of the i-th point, and let i be the label of the i-th point;

[0149] j is the label of the j-th point, j = 1, 2, ..., n;

[0150] z i Let i be the value of the i-th point, where i = 1, 2, ..., n;

[0151] d j It is the point (x,y) to (x j ,y j ) horizontal distance of points;

[0152] p is a weighted power exponent greater than 0;

[0153] A j Let be the angle relation weight of the j-th point.

[0154] According to a specific embodiment of the present invention, the mesh model is corrected based on similarity coefficient and difference information, including:

[0155] The similarity coefficient K is defined as the degree of similarity between the constrained mesh model and the unconstrained mesh model, and its range is [-1, 1].

[0156] When K equals 0, it means that the shape of the constrained mesh model is unrelated to that of the unconstrained mesh model.

[0157] A K value of 1 indicates a strong correlation; a K value greater than 0 and less than 1 indicates a weak correlation; a K value of -1 indicates an inverse correlation.

[0158] The similarity coefficient K is used to control the amount of correction of the mesh model by the difference information, and finally the mesh model correction is completed.

[0159] According to a specific embodiment of the present invention, it further includes at least one of the following in the network model after correction based on model grid point difference information and similarity coefficient control: high-speed top boundary morphology, low-speed reduction zone grid velocity, and grid velocity below the high-speed top boundary.

[0160] According to a specific embodiment of the present invention, the high-speed top boundary morphology of the mesh model after correction based on model mesh point difference information and similarity coefficient control includes:

[0161] Substituting the difference in low-velocity zone thickness and the rate of change in thickness from the model grid point difference information into the similarity coefficient calculation formula, we obtain the low-velocity zone thickness after the model grid point correction.

[0162] Subtract the thickness of the low-velocity zone from the surface elevation at that point to obtain the corrected elevation of the high-speed top interface.

[0163] The high-speed top boundary morphology is adjusted by adjusting the similarity coefficient.

[0164] According to a specific embodiment of the present invention, the mesh velocity in the low-deceleration zone of the mesh model is corrected based on the model mesh point difference information and similarity coefficient, including:

[0165] Substituting the equivalent velocity difference and equivalent velocity change rate in the low-speed-deceleration zone from the model grid point difference information into the similarity coefficient calculation formula, we obtain the equivalent velocity of the low-speed-deceleration zone after the model grid point correction.

[0166] The ratio of the change in the thickness of the low-speed-deceleration zone is obtained by quoting the corrected thickness of the low-speed-deceleration zone with the original thickness.

[0167] Calculate the average velocity for each corrected grid corresponding to the original grid and fill it into the corresponding corrected grid.

[0168] Then, calculate the rate of change of the equivalent velocity based on the equivalent velocity of the low deceleration zone before and after the correction;

[0169] All mesh velocities in the low-decay zone of the new mesh model are multiplied by the equivalent velocity change rate to complete the velocity correction of the low-decay zone of the mesh model.

[0170] According to a specific embodiment of the present invention, the mesh velocity below the high-speed top boundary of the mesh model is corrected based on model mesh point difference information and similarity coefficient, including:

[0171] Substituting the high-speed top velocity difference and the high-speed top velocity change rate from the model grid point difference information into the similarity coefficient calculation formula, we obtain the high-speed top velocity after the model grid point correction;

[0172] Find the mesh whose velocity is closest to the high-speed top velocity of the modified mesh model in the mesh model before modification. Using this mesh as the starting point, stitch this mesh and all subsequent meshes together in sequence below the low-velocity zone mesh of the modified mesh model.

[0173] If the number of grid cells after correction is less than the number before correction, copy the bottommost grid cell and continue stitching until the number of grid cells before correction is the same; if the number of grid cells after correction is more than the number before correction, trim the excess grid cells.

[0174] Typically, a low-speed reduction zone at a given point will occupy multiple vertical grids. When stitching, the process involves finding the velocity grid in the grid of that point before correction that is closest to the last grid of the low-speed reduction zone after correction, and then stitching all the grids below that grid sequentially after the last grid of the low-speed reduction zone after correction.

[0175] Compare the number of grid cells before and after correction. Assuming a certain point's grid model consists of 10 cells vertically, and after correction, the point representing the end of the low-velocity zone is the 4th cell, and the 7th cell in the original grid for that point has the closest velocity to the 4th cell after correction, then append the 8th to 10th cells from the original grid to the 4th cell in the corrected grid for that point. This results in only 8 cells vertically at that point, a decrease in the number of grid cells after correction compared to before. Conversely, the number of grid cells increases after correction. If a point only has 8 vertical grid cells after correction, then the 8th cell should be copied and pasted as the 9th and 10th cells to maintain consistency with the original grid number. (Reference) Figure 10 This is a comparison of other grid numbers before and after correction in two different cases.

[0176] According to a specific embodiment of the present invention, the method further includes correcting the mesh model velocity based on the surface delay time, including: if the surface delay time is accurate, correcting the mesh model velocity in the low-deceleration zone region based on the thickness of the low-deceleration zone of the mesh model.

[0177] According to a specific embodiment of the present invention, correcting the mesh model velocity based on surface delay specifically includes:

[0178] First, extract the equivalent velocity surface and thickness surface of the low-velocity zone from the mesh model;

[0179] Then, based on the location of the shot receiver, the thickness of the low velocity reduction zone in the corresponding grid of the shot receiver is extracted from the thickness surface;

[0180] The equivalent velocity of the shot receiver is obtained by dividing the thickness of the low velocity reduction zone in the grid corresponding to the shot receiver by the surface delay of the shot receiver.

[0181] An adaptive optimization interpolation method based on inverse distance weighted interpolation is used to assign the discrete shot-receiver point equivalent velocity to each grid point.

[0182] Divide the equivalent velocity assigned to each grid point through the shot receiver point by the equivalent velocity of the low-velocity zone extracted from the grid model to obtain the rate of change of the equivalent velocity of the low-velocity zone.

[0183] The mesh velocity at each mesh point in the low-deceleration zone is multiplied by the equivalent velocity change rate to correct the mesh model velocity for surface delay.

[0184] Example 1

[0185] According to a specific embodiment of the present invention, in conjunction with the appendix Figure 1 The following is a detailed description of the surface mesh model conversion and correction method based on well control constraints of the present invention.

[0186] This invention provides a method for converting and correcting surface mesh models based on well control constraints, comprising the following steps:

[0187] The network model data surface is converted from the continuous grid model sample, and the grid model information of the micro-logging point is calculated to obtain the difference information between the discrete micro-logging point and the grid model point; the data surface includes: thickness surface and equivalent velocity surface;

[0188] Based on the difference information between discrete micro-logging and grid model points, the difference information is allocated according to well control constraints, and the difference information corresponding to each grid model point is calculated.

[0189] The mesh model is corrected based on similarity coefficient and difference information.

[0190] Example 2

[0191] According to a specific embodiment of the present invention, in conjunction with the appendix Figure 2 The following is a detailed description of the surface mesh model conversion and correction method based on well control constraints of the present invention.

[0192] This invention provides a method for converting and correcting surface mesh models based on well control constraints, comprising the following steps:

[0193] The network model data surface is converted from the continuous grid model sample, and the grid model information of the micro-logging point is calculated to obtain the difference information between the discrete micro-logging point and the grid model point; the data surface includes: thickness surface and equivalent velocity surface;

[0194] The process of converting network model data surfaces from continuous grid model samples and calculating micro-logging point grid model information specifically includes:

[0195] The model thickness and equivalent velocity of the continuous grid are calculated based on the surface elevation and the top interface elevation of the high-velocity grid sample, forming the model thickness surface and the equivalent velocity surface.

[0196] The surface velocity model consists of a low-velocity layer and a deceleration layer. The first layer of the surface velocity model is called the low-velocity layer, and all layers from the second layer onwards are called deceleration layers. The high-velocity top interface is the top surface of the first medium layer that can refract and propagate seismic waves after the shallow surface low-velocity medium is stripped away. The propagation speed of seismic waves in this medium layer is the high-velocity top velocity. The low-deceleration zone is the entire surface model medium.

[0197] Based on the micrologging coordinates, the mesh model thickness and equivalent velocity results corresponding to the micrologging points are extracted.

[0198] The process of obtaining the difference information between discrete micro-logging and grid model points includes the following steps:

[0199] The formula for calculating the total thickness and equivalent velocity of the micrologging results is as follows:

[0200]

[0201]

[0202]

[0203] in,

[0204] m represents the total number of floors;

[0205] h i The thickness of the i-th layer representing the results of micrologging;

[0206] v i Represents the velocity of the i-th layer;

[0207] h 总 It is the total thickness;

[0208] t 总 This is the total time;

[0209] v 平 It is the equivalent velocity in the low-velocity zone of the actual micro-logging.

[0210] The low-velocity band thickness surface, the equivalent velocity surface, and the high-velocity top boundary velocity surface are extracted from the mesh model.

[0211] Based on the coordinates of the microlog, the grid model data corresponding to the microlog location is extracted from the low-velocity zone thickness surface, the equivalent velocity surface, and the high-velocity top boundary velocity surface.

[0212] The difference and quotient of the total thickness of the micrologging results and the thickness of the corresponding grid model at the point are respectively taken as the thickness difference Δh and the thickness change rate Δh. p The obtained Δh and Δh p The difference information belonging to this micro-logging point;

[0213] The difference and quotient of the equivalent velocity from the micrologging results and the equivalent velocity from the corresponding grid model at the point are respectively taken as the equivalent velocity difference ΔV. A and equivalent rate of change of velocity ΔV Ap The obtained ΔV A and ΔV Ap The difference information belonging to this micro-logging point;

[0214] The difference and quotient between the high-velocity top velocity of the micrologging results and the high-velocity top velocity of the corresponding grid model at the point are respectively taken as the high-velocity top velocity difference ΔV. T and the rate of change of high-speed top velocity ΔV Tp The obtained ΔV T With ΔV Tp It refers to the difference information at that micro-logging point.

[0215] Based on the difference information between discrete micro-logging points and grid model points, an adaptive optimization interpolation method based on inverse distance weighted interpolation is used to allocate the difference information of micro-logging points to each grid model point, thereby completing the allocation of difference information based on well control constraints and obtaining the difference information corresponding to each grid model point.

[0216] The mesh model is corrected based on similarity coefficient and difference information, including:

[0217] The similarity coefficient is defined as the degree of similarity between the constrained mesh model and the unconstrained mesh model, and its range is [-1, 1].

[0218] When K equals 0, it means that the shape of the constrained mesh model is unrelated to that of the unconstrained mesh model.

[0219] A K value of 1 indicates a strong correlation; a K value greater than 0 and less than 1 indicates a weak correlation; a K value of -1 indicates an inverse correlation.

[0220] The similarity coefficient K is used to control the amount of correction of the mesh model by the difference information, and finally the mesh model correction is completed.

[0221] Example 3

[0222] According to a specific embodiment of the present invention, in combination with Figure 2 and Figure 4-8 The following is a detailed description of the surface mesh model conversion and correction method based on well control constraints of the present invention.

[0223] This invention provides a method for converting and correcting surface mesh models based on well control constraints, comprising the following steps:

[0224] The network model data surface is extracted from the continuous grid model samples. The grid model information of micrologging points is calculated to obtain the difference information between the discrete micrologging points and the grid model points. The data surface includes: a thickness surface and an equivalent velocity surface. The continuous network model is as follows: Figure 4 As shown.

[0225] The process of converting network model data surfaces from continuous grid model samples and calculating micro-logging point grid model information specifically includes:

[0226] The model thickness and equivalent velocity of the continuous mesh are calculated based on the surface elevation and the top interface elevation of the high-velocity surface of the continuous mesh model sample, forming the model thickness surface and the equivalent velocity surface; the thickness surface is as follows: Figure 5 As shown, the equivalent velocity surface is as follows Figure 6 As shown.

[0227] The surface velocity model consists of a low-velocity layer and a deceleration layer. The first layer of the surface velocity model is called the low-velocity layer, and all layers from the second layer onwards are called deceleration layers. The high-velocity top interface is the top surface of the first medium layer that can refract and propagate seismic waves after the shallow surface low-velocity medium is stripped away. The propagation speed of seismic waves in this medium layer is the high-velocity top velocity. The low-deceleration zone is the entire surface model medium.

[0228] Based on the micrologging coordinates, the mesh model thickness and equivalent velocity results corresponding to the micrologging points are extracted.

[0229] The process of obtaining the difference information between discrete micro-logging and grid model points includes the following steps:

[0230] The formula for calculating the total thickness and equivalent velocity of the micrologging results is as follows:

[0231]

[0232]

[0233]

[0234] in,

[0235] m represents the total number of floors;

[0236] h i The thickness of the i-th layer representing the results of micrologging;

[0237] v i Represents the velocity of the i-th layer;

[0238] h 总 It is the total thickness;

[0239] t 总 This is the total time;

[0240] v 平 It is the equivalent velocity in the low-velocity zone of the actual micro-logging.

[0241] The low-velocity band thickness surface, the equivalent velocity surface, and the high-velocity top boundary velocity surface are extracted from the mesh model.

[0242] Based on the coordinates of the microlog, the grid model data corresponding to the microlog location is extracted from the low-velocity zone thickness surface, the equivalent velocity surface, and the high-velocity top boundary velocity surface.

[0243] The difference and quotient of the total thickness of the micrologging results and the thickness of the corresponding grid model at the point are respectively taken as the thickness difference Δh and the thickness change rate Δh. p ;

[0244] The obtained Δh and Δh p The difference information belonging to this micro-logging point;

[0245] The difference and quotient of the equivalent velocity from the micrologging results and the equivalent velocity from the corresponding grid model at the point are respectively taken as the equivalent velocity difference ΔV. A and equivalent rate of change of velocity ΔV Ap The obtained ΔV A and ΔV Ap The difference information belonging to this micro-logging point;

[0246] The difference and quotient between the high-velocity top velocity of the micrologging results and the high-velocity top velocity of the corresponding grid model at the point are respectively taken as the high-velocity top velocity difference ΔV. T and the rate of change of high-speed top velocity ΔV Tp The obtained ΔV T With ΔV Tp It refers to the difference information at that micro-logging point.

[0247] Based on the difference information between discrete micro-logging points and grid model points, an adaptive optimization interpolation method based on inverse distance weighted interpolation is used to allocate the difference information of micro-logging points to each grid model point, thereby completing the allocation of difference information based on well control constraints and obtaining the difference information corresponding to each grid model point.

[0248] The adaptive optimization interpolation method based on inverse distance weighted interpolation is as follows: on the basis of inverse distance weighted interpolation, the angular relationship information between the interpolation target point and each control point is used as the interpolation control weight. Inverse distance weighted interpolation is a weighted average interpolation method with the reciprocal of the distance as the interpolation coefficient.

[0249] like Figure 7 As shown, the adaptive optimization interpolation method is as follows:

[0250] Let the interpolation target point be a "light source", and each control point have a circular area that is opaque in the center and gradually becomes transparent towards the edges. Then the part of the circular area facing away from the light source has a "shadow".

[0251] The formula for the angular relationship weight A between the control points in the "shadow," including those that can see all "light sources," those that cannot see any "light sources," and those that can see some "light sources," is as follows:

[0252]

[0253] in,

[0254] θ1 is the angle at which the control point, which is not visible as a "light source," enters the "shadow" of the interpolation target point;

[0255] θ2 is the angle at which the control point, which is completely invisible from the light source, enters the "shadow" of the interpolation target point;

[0256] α is the angle at which the control points that can see all the "light sources" enter the "shadow" of the interpolation target point;

[0257] The adaptive optimization interpolation method of inverse distance weighted interpolation is as follows:

[0258] Given n points, the inverse distance weighted interpolation function is:

[0259]

[0260] By incorporating the angular relationship weights between the control points and the target points into the inverse distance weighted interpolation formula, we obtain an adaptive optimization interpolation formula based on inverse distance weighted interpolation:

[0261]

[0262] in,

[0263] (x i ,y i Let be the planar coordinates of the i-th point, and let i be the label of the i-th point;

[0264] j is the label of the j-th point, j = 1, 2, ..., n;

[0265] z i Let i be the value of the i-th point, where i = 1, 2, ..., n;

[0266] d j It is the point (x,y) to (x j ,y j ) horizontal distance of points;

[0267] p is a weighted power exponent greater than 0;

[0268] A j Let be the angle relation weight of the j-th point.

[0269] The mesh model is corrected based on similarity coefficient and difference information, including:

[0270] The similarity coefficient is defined as the degree of similarity between the constrained mesh model and the unconstrained mesh model, and its range is [-1, 1].

[0271] When K equals 0, it means that the shape of the constrained mesh model is unrelated to that of the unconstrained mesh model.

[0272] A K value of 1 indicates a strong correlation; a K value greater than 0 and less than 1 indicates a weak correlation; a K value of -1 indicates an inverse correlation.

[0273] The similarity coefficient K controls the amount of correction applied to the mesh model using the difference information, ultimately completing the mesh model correction. The role of the correlation coefficient is as follows: Figure 8 As shown in (a)-(d), the correlation coefficients can be adjusted according to the understanding of the actual data and the needs.

[0274] Example 4

[0275] According to a specific embodiment of the present invention, the surface mesh model conversion and correction method based on well control constraints of the present invention will be described in detail below.

[0276] This invention provides a method for converting and correcting surface mesh models based on well control constraints, comprising the following steps:

[0277] The network model data surface is converted from the continuous grid model sample, and the grid model information of the micro-logging point is calculated to obtain the difference information between the discrete micro-logging point and the grid model point; the data surface includes: thickness surface and equivalent velocity surface;

[0278] The process of converting network model data surfaces from continuous grid model samples and calculating micro-logging point grid model information specifically includes:

[0279] The model thickness and equivalent velocity of the continuous grid are calculated based on the surface elevation and the top interface elevation of the high-velocity grid sample, forming the model thickness surface and the equivalent velocity surface.

[0280] The surface velocity model consists of a low-velocity layer and a deceleration layer. The first layer of the surface velocity model is called the low-velocity layer, and all layers from the second layer onwards are called deceleration layers. The high-velocity top interface is the top surface of the first medium layer that can refract and propagate seismic waves after the shallow surface low-velocity medium is stripped away. The propagation speed of seismic waves in this medium layer is the high-velocity top velocity. The low-deceleration zone is the entire surface model medium.

[0281] Based on the micrologging coordinates, the mesh model thickness and equivalent velocity results corresponding to the micrologging points are extracted.

[0282] The process of obtaining the difference information between discrete micro-logging and grid model points includes the following steps:

[0283] The formula for calculating the total thickness and equivalent velocity of the micrologging results is as follows:

[0284]

[0285]

[0286]

[0287] in,

[0288] m represents the total number of floors;

[0289] h i The thickness of the i-th layer representing the results of micrologging;

[0290] v i Represents the velocity of the i-th layer;

[0291] h 总 It is the total thickness;

[0292] t 总 This is the total time;

[0293] v 平 It is the equivalent velocity in the low-velocity zone of the actual micro-logging.

[0294] The low-velocity band thickness surface, the equivalent velocity surface, and the high-velocity top boundary velocity surface are extracted from the mesh model.

[0295] Based on the coordinates of the microlog, the grid model data corresponding to the microlog location is extracted from the low-velocity zone thickness surface, the equivalent velocity surface, and the high-velocity top boundary velocity surface.

[0296] The difference and quotient of the total thickness of the micrologging results and the thickness of the corresponding grid model at the point are respectively taken as the thickness difference Δh and the thickness change rate Δh. p ;

[0297] The obtained Δh and Δh p The difference information belonging to this micro-logging point.

[0298] The difference and quotient of the equivalent velocity from the micrologging results and the equivalent velocity from the corresponding grid model at the point are respectively taken as the equivalent velocity difference ΔV. A and equivalent rate of change of velocity ΔV Ap The obtained ΔV A and ΔV Ap The difference information belonging to this micro-logging point;

[0299] The difference and quotient between the high-velocity top velocity of the micrologging results and the high-velocity top velocity of the corresponding grid model at the point are respectively taken as the high-velocity top velocity difference ΔV. T and the rate of change of high-speed top velocity ΔV Tp The obtained ΔV T With ΔV Tp It refers to the difference information at that micro-logging point.

[0300] Based on the difference information between discrete micro-logging points and grid model points, an adaptive optimization interpolation method based on inverse distance weighted interpolation is used to allocate the difference information of micro-logging points to each grid model point, thereby completing the allocation of difference information based on well control constraints and obtaining the difference information corresponding to each grid model point.

[0301] The adaptive optimization interpolation method based on inverse distance weighted interpolation is as follows: on the basis of inverse distance weighted interpolation, the angular relationship information between the interpolation target point and each control point is used as the interpolation control weight. Inverse distance weighted interpolation is a weighted average interpolation method with the reciprocal of the distance as the interpolation coefficient.

[0302] The adaptive optimization interpolation method is as follows:

[0303] Let the interpolation target point be a "light source", and each control point have a circular area that is opaque in the center and gradually becomes transparent towards the edges. Then the part of the circular area facing away from the light source has a "shadow".

[0304] The formula for the angular relationship weight A between the control points in the "shadow," including those that can see all "light sources," those that cannot see any "light sources," and those that can see some "light sources," is as follows:

[0305]

[0306] in,

[0307] θ1 is the angle at which the control point, which is not visible as a "light source," enters the "shadow" of the interpolation target point;

[0308] θ2 is the angle at which the control point, which is completely invisible from the light source, enters the "shadow" of the interpolation target point;

[0309] α is the angle at which the control points that can see all the "light sources" enter the "shadow" of the interpolation target point;

[0310] The adaptive optimization interpolation method of inverse distance weighted interpolation is as follows:

[0311] Given n points, the inverse distance weighted interpolation function is:

[0312]

[0313] By incorporating the angular relationship weights between the control points and the target points into the inverse distance weighted interpolation formula, we obtain an adaptive optimization interpolation formula based on inverse distance weighted interpolation:

[0314]

[0315] in,

[0316] (x i ,y i Let be the planar coordinates of the i-th point, and let i be the label of the i-th point;

[0317] j is the label of the j-th point, j = 1, 2, ..., n;

[0318] z i Let i be the value of the i-th point, where i = 1, 2, ..., n;

[0319] d j It is the point (x,y) to (x j ,y j ) horizontal distance of points;

[0320] p is a weighted power exponent greater than 0;

[0321] A j Let be the angle relation weight of the j-th point.

[0322] The mesh model is corrected based on similarity coefficient and difference information, including:

[0323] The similarity coefficient is defined as the degree of similarity between the constrained mesh model and the unconstrained mesh model, and its range is [-1, 1].

[0324] When K equals 0, it means that the shape of the constrained mesh model is unrelated to that of the unconstrained mesh model.

[0325] A K value of 1 indicates a strong correlation; a K value greater than 0 and less than 1 indicates a weak correlation; a K value of -1 indicates an inverse correlation.

[0326] The similarity coefficient K is used to control the amount of correction of the mesh model by the difference information, and finally the mesh model correction is completed.

[0327] It also includes the high-speed top boundary morphology of the mesh model after correction based on model mesh point difference information and similarity coefficient control, including:

[0328] Substituting the difference in low-velocity zone thickness and the rate of change in thickness from the model grid point difference information into the similarity coefficient calculation formula, we obtain the low-velocity zone thickness after the model grid point correction.

[0329] Subtract the thickness of the low-velocity zone from the surface elevation at that point to obtain the corrected elevation of the high-speed top interface.

[0330] The high-speed top boundary morphology is adjusted by adjusting the similarity coefficient.

[0331] It also includes correcting the mesh velocity in the low-deceleration zone of the mesh model based on the model mesh point difference information and similarity coefficient, including:

[0332] Substituting the equivalent velocity difference and equivalent velocity change rate in the low-speed-deceleration zone from the model grid point difference information into the similarity coefficient calculation formula, we obtain the equivalent velocity of the low-speed-deceleration zone after the model grid point correction.

[0333] The ratio of the change in the thickness of the low-speed-deceleration zone is obtained by quoting the corrected thickness of the low-speed-deceleration zone with the original thickness.

[0334] Calculate the average velocity for each corrected grid corresponding to the original grid and fill it into the corresponding corrected grid.

[0335] Then, calculate the rate of change of the equivalent velocity based on the equivalent velocity of the low deceleration zone before and after the correction;

[0336] All mesh velocities within the low-decay zone of the new mesh model are multiplied by the equivalent rate of velocity change to complete the velocity correction for the low-decay zone of the mesh model. The corrected mesh velocities for the low-decay zone are as follows: Figure 9 As shown, the corrected velocity variation law of the low-speed band grid is equivalent to the compression or stretching of the velocity variation law before the correction, and the trend of change is consistent with that before the correction.

[0337] It also includes correcting the mesh velocity below the high-speed top boundary of the mesh model based on the model mesh point difference information and similarity coefficient, including:

[0338] Substituting the high-speed top velocity difference and the high-speed top velocity change rate from the model grid point difference information into the similarity coefficient calculation formula, we obtain the high-speed top velocity after the model grid point correction;

[0339] Find the mesh whose velocity is closest to the high-speed top velocity of the modified mesh model in the mesh model before modification. Using this mesh as the starting point, stitch this mesh and all subsequent meshes together in sequence below the low-velocity zone mesh of the modified mesh model.

[0340] If the number of grid cells after correction is less than the number before correction, copy the bottommost grid cell and continue stitching until the number of grid cells before correction is the same; if the number of grid cells after correction is more than the number before correction, trim the excess grid cells.

[0341] Example 5

[0342] According to a specific embodiment of the present invention, in conjunction with the appendix Figure 3-12The following is a detailed description of the surface mesh model conversion and correction method based on well control constraints of the present invention.

[0343] This invention provides a method for converting and correcting surface mesh models based on well control constraints, comprising the following steps:

[0344] The network model data surface is extracted from the continuous grid model samples. The grid model information of micrologging points is calculated to obtain the difference information between the discrete micrologging points and the grid model points. The data surface includes: a thickness surface and an equivalent velocity surface. The continuous network model is as follows: Figure 4 As shown;

[0345] The process of converting network model data surfaces from continuous grid model samples and calculating micro-logging point grid model information specifically includes:

[0346] The model thickness and equivalent velocity of the continuous mesh are calculated based on the surface elevation and the top interface elevation of the high-velocity surface of the continuous mesh model sample, forming the model thickness surface and the equivalent velocity surface; the thickness surface is as follows: Figure 5 As shown, the equivalent velocity surface is as follows Figure 6 As shown.

[0347] The surface velocity model consists of a low-velocity layer and a deceleration layer. The first layer of the surface velocity model is called the low-velocity layer, and all layers from the second layer onwards are called deceleration layers. The high-velocity top interface is the top surface of the first medium layer that can refract and propagate seismic waves after the shallow surface low-velocity medium is stripped away. The propagation speed of seismic waves in this medium layer is the high-velocity top velocity. The low-deceleration zone is the entire surface model medium.

[0348] Based on the micrologging coordinates, the mesh model thickness and equivalent velocity results corresponding to the micrologging points are extracted.

[0349] The process of obtaining the difference information between discrete micro-logging and grid model points includes the following steps:

[0350] The formula for calculating the total thickness and equivalent velocity of the micrologging results is as follows:

[0351]

[0352]

[0353]

[0354] in,

[0355] m represents the total number of floors;

[0356] h i The thickness of the i-th layer representing the results of micrologging;

[0357] v iRepresents the velocity of the i-th layer;

[0358] h 总 It is the total thickness;

[0359] t 总 This is the total time;

[0360] v 平 It is the equivalent velocity in the low-velocity zone of the actual micro-logging.

[0361] The low-velocity band thickness surface, the equivalent velocity surface, and the high-velocity top boundary velocity surface are extracted from the mesh model.

[0362] Based on the coordinates of the microlog, the grid model data corresponding to the microlog location is extracted from the low-velocity zone thickness surface, the equivalent velocity surface, and the high-velocity top boundary velocity surface.

[0363] The difference and quotient of the total thickness of the micrologging results and the thickness of the corresponding grid model at the point are respectively taken as the thickness difference Δh and the thickness change rate Δh. p ;

[0364] The obtained Δh and Δh p The difference information belonging to this micro-logging point.

[0365] The difference and quotient of the equivalent velocity from the micrologging results and the equivalent velocity from the corresponding grid model at the point are respectively taken as the equivalent velocity difference ΔV. A and equivalent rate of change of velocity ΔV Ap The obtained ΔV A and ΔV Ap The difference information belonging to this micro-logging point;

[0366] The difference and quotient between the high-velocity top velocity of the micrologging results and the high-velocity top velocity of the corresponding grid model at the point are respectively taken as the high-velocity top velocity difference ΔV. T and the rate of change of high-speed top velocity ΔV Tp The obtained ΔV T With ΔV Tp It refers to the difference information at that micro-logging point.

[0367] Based on the difference information between discrete micro-logging points and grid model points, an adaptive optimization interpolation method based on inverse distance weighted interpolation is used to allocate the difference information of micro-logging points to each grid model point, thereby completing the allocation of difference information based on well control constraints and obtaining the difference information corresponding to each grid model point.

[0368] The adaptive optimization interpolation method based on inverse distance weighted interpolation is as follows: on the basis of inverse distance weighted interpolation, the angular relationship information between the interpolation target point and each control point is used as the interpolation control weight. Inverse distance weighted interpolation is a weighted average interpolation method with the reciprocal of the distance as the interpolation coefficient.

[0369] The adaptive optimization interpolation method is as follows:

[0370] Let the interpolation target point be a "light source", and each control point have a circular area that is opaque in the center and gradually becomes transparent towards the edges. Then the part of the circular area facing away from the light source has a "shadow".

[0371] The formula for the angular relationship weight A between the control points in the "shadow," including those that can see all "light sources," those that cannot see any "light sources," and those that can see some "light sources," is as follows:

[0372]

[0373] in,

[0374] θ1 is the angle at which the control point, which is not visible as a "light source," enters the "shadow" of the interpolation target point;

[0375] θ2 is the angle at which the control point, which is completely invisible from the light source, enters the "shadow" of the interpolation target point;

[0376] α is the angle at which the control points that can see all the "light sources" enter the "shadow" of the interpolation target point;

[0377] like Figure 7 As shown, the adaptive optimization interpolation method of inverse distance weighted interpolation is as follows:

[0378] Given n points, the inverse distance weighted interpolation function is:

[0379]

[0380] By incorporating the angular relationship weights between the control points and the target points into the inverse distance weighted interpolation formula, we obtain an adaptive optimization interpolation formula based on inverse distance weighted interpolation:

[0381]

[0382] in,

[0383] (x i ,y i Let be the planar coordinates of the i-th point, and let i be the label of the i-th point;

[0384] j is the label of the j-th point, j = 1, 2, ..., n;

[0385] z i Let i be the value of the i-th point, where i = 1, 2, ..., n;

[0386] d j It is the point (x,y) to (x j ,yj ) horizontal distance of points;

[0387] p is a weighted power exponent greater than 0;

[0388] A j Let be the angle relation weight of the j-th point.

[0389] The mesh model is corrected based on similarity coefficient and difference information, including:

[0390] The similarity coefficient is defined as the degree of similarity between the constrained mesh model and the unconstrained mesh model, and its range is [-1, 1].

[0391] When K equals 0, it means that the shape of the constrained mesh model is unrelated to that of the unconstrained mesh model.

[0392] A K value of 1 indicates a strong correlation; a K value greater than 0 and less than 1 indicates a weak correlation; a K value of -1 indicates an inverse correlation.

[0393] The similarity coefficient K controls the amount of correction applied to the mesh model using the difference information, ultimately completing the mesh model correction. The role of the correlation coefficient is as follows: Figure 8 As shown in (a)-(d).

[0394] It also includes the high-speed top boundary morphology of the mesh model after correction based on model mesh point difference information and similarity coefficient control, including:

[0395] Substituting the difference in low-velocity zone thickness and the rate of change in thickness from the model grid point difference information into the similarity coefficient calculation formula, we obtain the low-velocity zone thickness after the model grid point correction.

[0396] Subtract the thickness of the low-velocity zone from the surface elevation at that point to obtain the corrected elevation of the high-speed top interface.

[0397] The high-speed top boundary morphology is adjusted by adjusting the similarity coefficient.

[0398] It also includes correcting the mesh velocity in the low-deceleration zone of the mesh model based on the model mesh point difference information and similarity coefficient, including:

[0399] Substituting the equivalent velocity difference and equivalent velocity change rate in the low-speed-deceleration zone from the model grid point difference information into the similarity coefficient calculation formula, we obtain the equivalent velocity of the low-speed-deceleration zone after the model grid point correction.

[0400] The ratio of the change in the thickness of the low-speed-deceleration zone is obtained by quoting the corrected thickness of the low-speed-deceleration zone with the original thickness.

[0401] Calculate the average velocity for each corrected grid corresponding to the original grid and fill it into the corresponding corrected grid.

[0402] Then, calculate the rate of change of the equivalent velocity based on the equivalent velocity of the low deceleration zone before and after the correction;

[0403] All mesh velocities within the low-decay zone of the new mesh model are multiplied by the equivalent rate of velocity change to complete the velocity correction for the low-decay zone of the mesh model. For example... Figure 9 A schematic diagram for correcting the grid velocity in the low-speed-deceleration zone.

[0404] It also includes correcting the mesh velocity below the high-speed top boundary of the mesh model based on the model mesh point difference information and similarity coefficient, including:

[0405] Substituting the high-speed top velocity difference and the high-speed top velocity change rate from the model grid point difference information into the similarity coefficient calculation formula, we obtain the high-speed top velocity after the model grid point correction;

[0406] Find the mesh whose velocity is closest to the high-speed top velocity of the modified mesh model in the mesh model before modification. Using this mesh as the starting point, stitch this mesh and all subsequent meshes together in sequence below the low-velocity zone mesh of the modified mesh model.

[0407] If the number of meshes after correction is less than before correction, copy the bottom mesh and continue stitching until the number of meshes is the same as before correction; if the number of meshes after correction is more than before correction, trim the excess meshes. Figure 10 The diagram shows the grid velocity below the high-speed top boundary of the corrected grid model. In fact, it is a "relocation" of the grid model below the high-speed top boundary before the correction. The high-speed top interface velocity after the correction is consistent with the micro-logging top interface velocity, and the grid model below the high-speed top boundary is completely consistent with the one before the correction.

[0408] It also includes, such as Figure 11 As shown, the mesh model velocity is corrected based on the surface delay time, including: assuming the surface delay time is accurate, the mesh model velocity in the low-velocity zone region is corrected based on the thickness of the low-velocity zone, specifically including:

[0409] First, extract the equivalent velocity surface and thickness surface of the low-velocity zone from the mesh model;

[0410] Then, based on the location of the shot receiver, the thickness of the low velocity reduction zone in the corresponding grid of the shot receiver is extracted from the thickness surface;

[0411] The equivalent velocity of the shot receiver is obtained by dividing the thickness of the low velocity reduction zone in the grid corresponding to the shot receiver by the surface delay of the shot receiver.

[0412] An adaptive optimization interpolation method based on inverse distance weighted interpolation is used to assign the discrete shot-receiver point equivalent velocity to each grid point.

[0413] Divide the equivalent velocity assigned to each grid point through the shot receiver point by the equivalent velocity of the low-velocity zone extracted from the grid model to obtain the rate of change of the equivalent velocity of the low-velocity zone.

[0414] Multiply the mesh velocity at each mesh point in the low-deceleration zone by the equivalent velocity change rate to correct the mesh model velocity for surface delay. For example... Figure 12 This is a schematic diagram illustrating the effect of correcting the mesh model velocity based on surface delay.

[0415] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of protection of the present invention.

Claims

1. A method for converting and correcting surface mesh models based on well control constraints, characterized in that, Includes the following steps: The network model data surface is converted from the continuous grid model sample, and the grid model information of the micro-logging point is calculated to obtain the difference information between the discrete micro-logging point and the grid model point; the data surface includes: thickness surface and equivalent velocity surface; Based on the difference information between discrete micro-logging and grid model points, the difference information is allocated according to well control constraints, and the difference information corresponding to each grid model point is calculated. The mesh model is corrected based on similarity coefficient and difference information; The process of converting network model data surfaces from continuous grid model samples and calculating micro-logging point grid model information specifically includes: The model thickness and equivalent velocity of the continuous grid are calculated based on the surface elevation and the top interface elevation of the high-velocity grid sample, forming the model thickness surface and the equivalent velocity surface. The surface velocity model consists of a low-velocity layer and a deceleration layer. The first layer of the surface velocity model is called the low-velocity layer, and all layers from the second layer onwards are called deceleration layers. The high-velocity top interface is the top surface of the first medium layer that can refract and propagate seismic waves after the shallow surface low-velocity medium is stripped away. The propagation speed of seismic waves in this medium layer is the high-velocity top velocity. The low-deceleration zone is the entire surface model medium. Based on the micrologging coordinates, the mesh model thickness and equivalent velocity results corresponding to the micrologging points are extracted; The process of obtaining the difference information between discrete micro-logging and grid model points includes the following steps: The formula for calculating the total thickness and equivalent velocity of the micrologging results is as follows: in, m Represents the total number of floors; h i The first representative of micro-logging results i Layer thickness; v i Representing the i The speed of the layer; It is the total thickness; This is the total time; It is the equivalent velocity in the low-velocity zone of the actual micro-logging. The low-velocity band thickness surface, the equivalent velocity surface, and the high-velocity top boundary velocity surface are extracted from the mesh model. Based on the coordinates of the microlog, the grid model data corresponding to the microlog location is extracted from the low-velocity zone thickness surface, the equivalent velocity surface, and the high-velocity top boundary velocity surface. The difference and quotient of the total thickness of the micrologging results and the thickness of the corresponding mesh model at each point are respectively used as the thickness difference. and thickness change rate ; obtained and The difference information belonging to this micro-logging point; The difference and quotient of the equivalent velocity from the micrologging results and the equivalent velocity from the corresponding grid model at the location are used as the equivalent velocity difference. and equivalent rate of change , obtained and The difference information belonging to this micro-logging point; The difference and quotient between the high-velocity top velocity of the micrologging results and the high-velocity top velocity of the corresponding grid model at the point are respectively taken as the high-velocity top velocity difference. and the rate of change of high-speed top velocity , obtained and It refers to the difference information at that micro-logging point.

2. The surface mesh model conversion and correction method based on well control constraints according to claim 1, characterized in that, Using the difference information between discrete micrologging and grid model points, the difference information is allocated based on well control constraints. The difference information corresponding to each grid model point is obtained by using an adaptive optimization interpolation method based on inverse distance weighted interpolation to allocate the difference information of micrologging points to each grid model point, thus completing the allocation of difference information based on well control constraints based on micrologging results.

3. The surface mesh model conversion and correction method based on well control constraints according to claim 2, characterized in that, The adaptive optimization interpolation method based on inverse distance weighted interpolation is as follows: on the basis of inverse distance weighted interpolation, the angular relationship information between the interpolation target point and each control point is used as the interpolation control weight.

4. The surface mesh model conversion and correction method based on well control constraints according to claim 3, characterized in that, The adaptive optimization interpolation method is as follows: Let the interpolation target point be a "light source", and each control point have a circular area that is opaque in the center and gradually becomes transparent towards the edges. Then the part of the circular area facing away from the light source has a "shadow". The control points within the "shadow" include those that can see all "light sources," those that cannot see any "light sources," and those that can see some "light sources." The angular relationship weights of these control points relative to the target point are as follows: The formula is: ； in, The angle at which the control point, which is not visible as a "light source," enters the "shadow" of the interpolation target point; The angle at which the control point, from which the "light source" is completely invisible, enters the "shadow" of the interpolation target point; To be able to see the angle at which the control points of all "light sources" enter the "shadow" of the interpolation target point.

5. The surface mesh model conversion and correction method based on well control constraints according to claim 4, characterized in that, The adaptive optimization interpolation method of inverse distance weighted interpolation is as follows: It has For each point, the inverse distance weighted interpolation function is: ； By incorporating the angular relationship weights between the control points and the target points into the inverse distance weighted interpolation formula, we obtain an adaptive optimization interpolation formula based on inverse distance weighted interpolation: ； in, For the first i The planar coordinates of the points i For the first i The label of each point; j For the first j The label of each point, ; For the first i The value of each point, ; yes Click Horizontal distance between points; The weighted exponent is greater than 0; A j Let be the angle relation weight of the j-th point.

6. The surface mesh model conversion and correction method based on well control constraints according to claim 5, characterized in that, The mesh model is corrected based on similarity coefficient and difference information, including: The similarity coefficient is defined as the degree of similarity between the constrained mesh model and the unconstrained mesh model. K The range is [-1, 1]; K When the value is 0, it means that the shape of the constrained mesh model is unrelated to that of the unconstrained mesh model; K A value of 1 indicates a strong correlation. K A value greater than 0 and less than 1 indicates a weak correlation; K A value of -1 indicates an inverse correlation; Using similarity coefficients K The amount of correction applied to the mesh model by controlling the difference information is completed, and finally the mesh model correction is finished.

7. The surface mesh model conversion and correction method based on well control constraints according to claim 6, characterized in that, It also includes at least one of the following in the network model after correction based on model grid point difference information and similarity coefficient control: high-speed top boundary morphology, low-speed reduction zone grid velocity, and grid velocity below the high-speed top boundary.

8. The surface mesh model conversion and correction method based on well control constraints according to claim 7, characterized in that, The high-speed top boundary morphology of the mesh model after correction based on model mesh point difference information and similarity coefficient control includes: Substituting the difference in low-velocity zone thickness and the rate of change in thickness from the model grid point difference information into the similarity coefficient calculation formula, we obtain the low-velocity zone thickness after the model grid point correction. Subtract the thickness of the low-velocity zone from the surface elevation at that point to obtain the corrected elevation of the high-speed top interface. The high-speed top boundary morphology is adjusted by adjusting the similarity coefficient.

9. The surface mesh model conversion and correction method based on well control constraints according to claim 7, characterized in that, Based on the model grid point difference information and similarity coefficient, the grid velocity in the low-deceleration zone of the grid model is corrected, including: Substituting the equivalent velocity difference and equivalent velocity change rate in the low-speed-deceleration zone from the model grid point difference information into the similarity coefficient calculation formula, we obtain the equivalent velocity of the low-speed-deceleration zone after the model grid point correction. The ratio of the change in the thickness of the low-speed-deceleration zone is obtained by quoting the corrected thickness of the low-speed-deceleration zone with the original thickness. Calculate the average velocity for each corrected grid corresponding to the original grid and fill it into the corresponding corrected grid. Then, calculate the rate of change of the equivalent velocity based on the equivalent velocity of the low deceleration zone before and after the correction; All mesh velocities in the low-decay zone of the new mesh model are multiplied by the equivalent velocity change rate to complete the velocity correction of the low-decay zone of the mesh model.

10. The surface mesh model conversion and correction method based on well control constraints according to claim 7, characterized in that, Based on the model mesh point difference information and similarity coefficient correction, the mesh velocity below the high-speed top boundary of the mesh model is corrected, including: Substituting the high-speed top velocity difference and the high-speed top velocity change rate from the model grid point difference information into the similarity coefficient calculation formula, we obtain the high-speed top velocity after the model grid point correction; Find the mesh whose velocity is closest to the high-speed top velocity of the modified mesh model in the mesh model before modification. Using this mesh as the starting point, stitch this mesh and all subsequent meshes together in sequence below the low-velocity zone mesh of the modified mesh model. If the number of grid cells after correction is less than the number before correction, copy the bottommost grid cell and continue stitching until the number of grid cells before correction is the same; if the number of grid cells after correction is more than the number before correction, trim the excess grid cells.

11. The method for surface mesh model conversion and correction based on well control constraints according to any one of claims 1-10, characterized in that, It also includes correcting the mesh model velocity based on the surface delay time, including: when the surface delay time is accurate, correcting the mesh model velocity in the low-deceleration zone region based on the thickness of the low-deceleration zone of the mesh model.

12. The surface mesh model conversion and correction method based on well control constraints according to claim 11, characterized in that, The specific steps for correcting the mesh model velocity based on surface delay include: First, extract the equivalent velocity surface and thickness surface of the low-velocity zone from the mesh model; Then, based on the location of the shot receiver, the thickness of the low velocity reduction zone in the corresponding grid of the shot receiver is extracted from the thickness surface; The equivalent velocity of the shot receiver is obtained by dividing the thickness of the low velocity reduction zone in the grid corresponding to the shot receiver by the surface delay of the shot receiver. An adaptive optimization interpolation method based on inverse distance weighted interpolation is used to assign the discrete shot-receiver point equivalent velocity to each grid point. Divide the equivalent velocity assigned to each grid point through the shot receiver point by the equivalent velocity of the low-velocity zone extracted from the grid model to obtain the rate of change of the equivalent velocity of the low-velocity zone. The mesh velocity at each mesh point in the low-deceleration zone is multiplied by the equivalent velocity change rate to correct the mesh model velocity for surface delay.

Images