Dual-complicated area true surface velocity fusion modeling method

Abstract

The application provides a true ground surface velocity fusion modeling method for double complex regions, which comprises the following steps: step 1, determining a near-ground surface model depth reliable range; step 2, obtaining a high-speed top interface and a floating reference surface; step 3, performing floating reference surface correction amount obtaining and seismic data correction; step 4, performing middle-deep layer reflected wave velocity analysis and tomographic modeling; step 5, performing true ground surface velocity fusion starting surface unification and velocity trend analysis; step 6, performing velocity gradient correction and inverse distance weighted fusion modeling; and step 7, outputting seismic data and fused true ground surface velocity model. The true ground surface velocity fusion modeling method for double complex regions can effectively make up for the defects of insufficient near-ground surface layer precision of reflection tomographic modeling, improve pre-stack depth migration velocity modeling precision, improve migration imaging quality and reduce well-seismic error.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of oilfield development, and particularly relates to a true surface velocity fusion modeling method for double complex regions. BACKGROUND

[0002] With the gradual development of pre-stack depth migration technology in major oilfields, good processing results have been achieved in general. Seismic data in the mountain front zone in western China has low signal-to-noise ratio, poor consistency, and severe lateral variation in velocity, sometimes accompanied by velocity inversion. The precision of the traditional velocity modeling method is not enough, especially in the shallow layer, which results in that it is difficult to achieve ideal results in pre-stack depth migration in this type of area. Therefore, how to establish a true surface velocity model and improve the precision of pre-stack depth migration velocity modeling has become the key to solving the imaging quality of migration in complex mountain front zones.

[0003] Many scholars have studied the velocity modeling problem in this type of area, mainly including: the travel time and path inversion of the seismic first arrival wave can obtain the surface medium velocity structure; Wang Xiaohua et al. proposed a multi-information constrained tomographic inversion technology, and Gao Lidong et al. proposed a near-surface modeling technology based on constrained first arrival wave tomographic inversion. The reflection wave carries the information of the underground interface and can reflect the structure and parameter distribution of different depths and different positions in the underground in a more fine scale, in which Zhang Kai and Qin Ning et al. proposed a tomographic velocity inversion method of reflection wave under double complex conditions. The tomographic inversion technology based on a single wave has been improved day by day, but in actual application, the near-surface velocity model obtained based on the first arrival wave tomographic inversion is mainly used in the complex surface static correction link and rarely used in the pre-stack depth migration modeling process. The tomographic modeling method based on reflection wave faces the problem of insufficient precision in the shallow layer.

[0004] The surface of the mountain front zone in western China is rugged, low, and the thickness and velocity of the low-velocity zone are complex and variable, and the structure is complex. The current conventional pre-stack depth migration idea for seismic data in the mountain front zone is: using the first arrival tomographic inversion of the large cannon to obtain the static correction amount of the shot point and the geophone of the datum surface, and performing a large smoothing radius smoothing on the actual surface elevation to obtain the smoothed surface (floating datum surface), and decomposing the correction amount to the floating surface by replacing the velocity. The single shot data is corrected to the floating surface, and the reflection wave tomographic modeling method is used to establish a pre-stack depth migration velocity model. This modeling method has a large artificial correction on the near-surface and a large modification on the near-surface layer, which results in insufficient velocity precision of the near-surface layer and affects the subsequent pre-stack depth migration imaging effect.

[0005] In the Chinese patent application No. 201410203486.9, a shallow-middle-deep layer velocity fusion method in seismic modeling is involved, and the specific implementation process is as follows: (1) taking a near-surface velocity model; obtaining a middle-deep layer velocity model; (2) selecting an appropriate window, and fusing the near-surface velocity model and the middle-deep layer velocity model into a unified shallow-middle-deep layer velocity model; (3) outputting the unified shallow-middle-deep layer velocity model. In the process (2), the specific calculation method of how to fuse the two velocities is mainly divided into 9 steps. Three standards are given for the fused velocity of the near-surface and the middle-deep layer: ① the fused velocity model should be well transitioned with the middle-deep layer velocity model outside the fusion area to avoid the generation of a pseudo interface; ② the fused velocity model value should be ensured to be within the range determined by the near-surface velocity and the middle-deep layer velocity value; ③ the fused velocity model can reduce the iteration times of migration as much as possible. The patent mainly introduces the algorithm of how to fuse the near-surface and the middle-deep layer velocity values, and good processing effect is achieved as a whole, but there are still several problems to be further solved: (1) in step 1, the near-surface velocity obtained by the big gun first arrival tomographic inversion is on the undulating elevation surface (true surface), and the reflected wave tomographic velocity is on the surface after a large radius smoothing of the undulating elevation surface (floating surface), and the two surfaces are quite different. (2) in step 1, the velocity model obtained by the big gun first arrival tomographic inversion is limited by the ray density, and with the increase of depth, the velocity accuracy is obviously reduced below the bottom interface depth covered by the ray density, and the depth range is not suitable, which greatly affects the velocity fusion. (3) in step 2, the algorithm of the fusion process of the near-surface and the middle-deep layer velocity values is relatively complicated, and the geophysical significance is not very clear.

[0006] In the Chinese patent application No. CN201410203486.9, a shallow-middle-deep layer velocity fusion method in seismic modeling is involved, including: obtaining a near-surface velocity model; obtaining a middle-deep layer velocity model; selecting an appropriate window, and fusing the near-surface velocity model and the middle-deep layer velocity model into a unified shallow-middle-deep layer velocity model; when the near-surface and the middle-deep layer velocity functions overlap in the depth domain during fusion, the shallow-middle-deep layer velocity model is fused by a weighting function; when there is a vacancy area of the two velocity functions in the depth domain, the vacancy is filled by interpolation to establish a unified full velocity function; and outputting the unified shallow-middle-deep layer velocity model. The patent application has poor adaptability in the selection of the fusion surface for the seismic data with a particularly complex surface and a large elevation difference, and the well data is not used in the velocity fusion process, and a higher precision initial velocity model can be theoretically established based on the logging curve.

[0007] In the Chinese patent application No. CN201910216730.8, a multi-information fusion seismic velocity modeling method is involved, which comprises: inputting information required for modeling; smoothing acoustic logging velocity; setting a number of virtual wells and constructing virtual wells; filling a velocity model along a structural trend with each logging data respectively and weighting and summing the filling results to obtain a logging velocity interpolation model; fusing the logging interpolation velocity with prior velocity and adjusting based on the prior velocity as a middle-deep layer velocity; determining a fusion bottom surface and a fusion area based on ray coverage of a near-surface model, and using near-surface velocity above the fusion bottom surface; obtaining a unified full velocity field by weighting and summing the near-surface velocity and the middle-deep layer velocity in the fusion area and counting well-seismic velocity error. The invention patent is mainly aimed at eastern seismic data, without considering the existence of undulating elevation difference, and the method is not suitable for complex seismic data in the west.

[0008] Therefore, we have invented a new double-complex-area true surface velocity fusion modeling method, which solves the above technical problems. SUMMARY

[0009] The purpose of the present application is to provide a double-complex-area true surface velocity fusion modeling method capable of providing a unified shallow-middle-deep layer velocity model and ensuring the accuracy of the migration velocity field and the undulating surface pre-stack depth migration result.

[0010] The purpose of the present application can be achieved by the following technical measures: a double-complex-area true surface velocity fusion modeling method, which comprises: step 1, determining a near-surface model depth reliable range; step 2, obtaining a high-speed top interface and a floating reference surface; step 3, performing floating reference surface correction amount obtaining and seismic data correction; step 4, performing middle-deep layer reflected wave velocity analysis and tomographic modeling; step 5, performing true surface velocity fusion starting surface unification and velocity trend analysis; step 6, performing velocity gradient correction and inverse distance weighted fusion modeling; and step 7, outputting seismic data and fused true surface velocity model.

[0011] The purpose of the present application can also be achieved by the following technical measures:

[0012] In step 1, ray density analysis is used by big cannon first arrival tomographic inversion, the bottom interface of the ray density is taken as an envelope surface A, and the depth difference from the surface elevation to the envelope surface A is the depth range of the near-surface model.

[0013] In step 2, a high-speed top interface B is selected above the envelope surface, and the high-speed top interface B is appropriately smoothed; the surface elevation interface is smoothed with a larger radius to obtain a floating surface C.

[0014] In step 3, the replacement velocity V revlaoe(BC)Calculate the correction amount T_BC from the high-speed top interface B to the floating surface C, and correct the seismic single-shot data to the floating surface C.

[0015] In step 3, the formulas for calculating the floating reference surface correction and correcting seismic data are: T float(BC) =T0+C source(BC) +C deteot(BC) Where T0 represents the original single-shot record, and C source(BC) Indicates the replacement speed V revlace(BC) The calculated correction amount from the actual surface of the shot point to the floating surface, C detect(BC) Indicates the replacement speed V revlace(BC) The calculated correction amount from the receiver point to the floating surface, T. float(BC) This represents the new shot record after the correction is applied to the floating surface C.

[0016] In step 4, reflection wave velocity analysis is performed on the seismic data on the floating surface C, and time-depth conversion is carried out to obtain the depth migration velocity of the middle and deep layers.

[0017] In step 5, after the analysis in steps 1-4 above, the seismic data and the velocity of the intermediate-deep reflected waves are on the floating surface C, and the near-surface velocity is on the true ground surface. Next, the seismic data and velocity will be unified onto the same surface.

[0018] In step 5, when unifying seismic data and velocities onto the same surface, the correction T_BC applied to the seismic data in step 2 is first removed. The true surface is then smoothed with a small radius to obtain a smooth surface D. The correction T_D from the true surface to the smooth surface D is calculated and applied to the seismic data, thereby correcting the data to the smooth surface D. Similarly, the near-surface and intermediate-deep velocities are corrected to the smooth surface D. The gap between the smooth surface D and the true surface is filled using velocity interpolation based on the nearest proximity principle.

[0019] In step 5, the seismic data is unified onto a floating surface D with a small smoothing radius on the actual Earth's surface. The specific formula is as follows:

[0020] T float(BD) =T float(BC) -C source(BC) -C detect(BC) +C source(BD) +C deteot(BD) In the formula, the first three terms on the right side represent the process of returning the seismic record to the true representative surface, C soure(BD) Indicates the replacement speed V revlace(BC) The calculated correction amount, C, is the actual surface correction from the shot point to the small smooth surface T_D. detect(BD) Indicates the replacement speed V revlace(BC) The calculated correction amount of the receiver point to the small smooth ground surface T_D, Tfloat(BD) New shot record after applying the new small smooth surface D to the correction amount.

[0021] In step 6, the small smooth surface D is selected to the high-speed top surface B as a velocity fusion window, the near-surface velocity and the middle-deep layer velocity trend in the window are analyzed, the near-surface velocity and the middle-deep layer velocity trend are corrected according to the logging velocity trend by using a gradient method, and then a reverse distance weighting function is selected for velocity fusion.

[0022] In step 7, the depth migration velocity model of the small smooth surface is output, and then the velocity is updated iteratively according to the pre-stack depth migration method.

[0023] The true surface velocity fusion modeling method in the double complex area in the application fuses the near-surface velocity and the middle-deep layer velocity obtained by reflection tomography together to establish an integrated pre-stack depth migration velocity model of the true surface, makes up for the defects of insufficient precision of the near-surface layer in reflection tomography modeling, and obtains a pre-stack depth migration velocity model with higher precision, thereby improving the imaging effect of the seismic data in the mountain front zone. The true surface velocity fusion modeling method for the double complex area is researched in the application, which is the inheritance and development of the existing patents, can establish a more accurate shallow-middle-deep pre-stack depth migration velocity model, improves the modeling precision, finally improves the pre-stack depth migration imaging quality, has wide applicability and application value. In the application, the near-surface velocity model is obtained by using the first arrival of the big gun to constrain the tomographic inversion, the near-surface velocity credible depth range is obtained according to the ray density bottom boundary depth, the appropriate elevation smoothing surface is selected as the starting surface of the near-surface and middle-deep layer velocity, the near-surface and middle-deep layer velocity trend is corrected according to the acoustic logging velocity trend, the reverse distance weighting method is used for fusion modeling, and a pre-stack depth migration initial velocity model with higher precision is obtained. The integrated modeling method of “taking, selecting, correcting and fusing” is innovated, and the quality and effect of the seismic data migration imaging are improved. Compared with the existing method of only relying on reflection for depth modeling, the application mainly has the following advantages:

[0024] The elevation changes greatly in the double complex area, the low and deceleration zone has thick and complex and changeable velocity, the traditional velocity modeling method has insufficient precision, especially the error of the shallow layer is large, the modeling method of the application comprehensively utilizes the first arrival and reflection fusion modeling, can effectively make up for the defects of insufficient precision of the near-surface layer in reflection tomography modeling, improve the pre-stack depth migration velocity modeling precision, improve the migration imaging quality, reduce the well-seismic error, and has great technical advantages compared with the traditional method. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 It is a schematic view of the smoothing surface, the surface and the ray density envelope surface in an embodiment of the application.

[0026] Figure 2 Figure 9 is a schematic diagram of a reflected wave velocity model in an embodiment of the present application;

[0027] Figure 3 Figure 10 is a schematic diagram of seismic data and velocity correction to a small smooth surface in an embodiment of the present application;

[0028] Figure 4 Figure 11 is a schematic diagram of a comparison of acoustic logging curve, near-surface and middle-deep layer longitudinal velocity trend in an embodiment of the present application;

[0029] Figure 5 Figure 12 is a schematic diagram of a comparison of fused velocity models in different data correction methods in an embodiment of the present application;

[0030] Figure 6 Figure 13 is a schematic diagram of a comparison of models in different mathematical algorithms for velocity fusion in an embodiment of the present application;

[0031] Figure 7 Figure 14 is a flowchart of a specific embodiment of a true surface velocity fusion modeling method in a double complex area of the present application. DETAILED DESCRIPTION

[0032] It should be noted that the following detailed description is merely exemplary in nature and is intended to provide further description of the application. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0033] It is also important to note that the terms "comprises", "comprising", "includes", "including", "contains", "containing" or variations thereof herein, do not specify an exhaustive or complete list of elements or steps as they can be subject to revision by the applicant.

[0034] The true surface velocity fusion modeling method in a double complex area of the present application includes the following steps:

[0035] Step 1, determination of the depth range of the near-surface model; mainly using the ray density analysis of the big gun first arrival tomography inversion, taking the ray density bottom boundary as the envelope surface A, and the depth difference from the surface elevation to the envelope surface A as the depth range of the near-surface model;

[0036] Step 2, determination of the high-speed top boundary and the floating reference surface; selecting the high-speed top boundary B above the envelope surface, and appropriately smoothing the high-speed top boundary B. Smoothing the surface elevation boundary with a large radius to obtain the floating surface C.

[0037] Step 3, calculation of correction amount of floating datum and correction of seismic data; using the replacement velocity V replace(BC) The correction amount T_BC from high velocity top surface B to floating surface C is calculated, and the seismic single shot data is corrected to the surface elevation floating surface C.

[0038] The calculation of correction amount of floating datum and correction of seismic data can be expressed as: T float(BC) = T0+C souroe(BC) +C deteot(BC) , wherein T0 represents the original single shot record, C souroe(BC) represents the correction amount of the shot point true surface to the floating surface calculated by the replacement velocity V replace(BC) , C detet(BC) represents the correction amount of the receiver point true surface to the floating surface calculated by the replacement velocity V replace(BC) , and T float(BC) represents the new shot record after the correction amount is applied to the floating surface C.

[0039] Step 4, middle-deep layer reflection wave velocity analysis and tomographic modeling; reflection wave velocity analysis is carried out for the seismic data on the floating surface C, and time-depth conversion is carried out to obtain the depth migration velocity of the middle-deep layer.

[0040] Step 5, unification of true surface velocity fusion starting surface and velocity trend analysis; after the analysis of the previous steps (1) (2) (3) (4), the seismic data and the middle-deep layer reflection wave velocity are on the floating surface C, and the near-surface velocity is on the true surface. Next, the seismic data and the velocity are unified to the same surface. The specific method is: remove the correction amount T_BC applied to the seismic data in (2), perform small radius smoothing on the true surface to obtain a smoothed surface D, and calculate the correction amount T_D of the true surface to the small smoothed surface D. Apply T_D on the seismic data to correct the data to the small smoothed surface D. Similarly, the near-surface and middle-deep layer velocity are corrected to the small smoothed surface D, and the gap between the small smoothed surface D and the true surface is filled by velocity interpolation according to the nearest principle.

[0041] The seismic data is unified to the true surface after small smoothing radius on the floating surface D, and the specific formula can be expressed as:

[0042] T float(BD) = T float(BC) -C source(BC) -C deteot(BC) +C souroe(BD) +C deteot(BD) , wherein the first three terms on the right side of the formula represent returning the seismic record to the true representative surface, C soure(BD) represents the correction amount of the shot point true surface to the small smoothed surface T_D calculated by the replacement velocity V revlaoe(BC) , C deteot(BD) represents the correction amount of the receiver point true surface to the small smoothed surface calculated by the replacement velocity Vrevlace(BC) The calculated correction amount of the receiver point to the small smooth ground surface T_D, T float(BD) This indicates the new shot record after applying the correction to a small, smooth surface D.

[0043] Step 6, velocity gradient correction and inverse distance weighted fusion modeling; select the small smooth surface D to the high-velocity top surface B as the velocity fusion window, analyze the near-surface velocity and medium-deep velocity trends within the window, and use the well logging velocity trend as a basis to perform gradient correction on the near-surface velocity and medium-deep velocity trends, and then select the inverse distance weighted function for velocity fusion.

[0044] Step 7: Output of seismic data and the fused true surface velocity model. Output the depth migration velocity model for a small, smooth surface. Subsequent velocity updates and iterations will follow the conventional pre-stack depth migration technique.

[0045] In a specific embodiment of the application of the present invention, such as Figure 7 As shown, Figure 7 The flowchart of the true surface velocity fusion modeling method for dual complex regions of the present invention includes the following steps:

[0046] (1) Determination of the reliable depth range of the near-surface model; mainly using the first arrival tomography inversion ray density analysis of the gun, with the bottom boundary of the ray density as the envelope surface A (see Figure 1 The depth difference from the surface elevation to the envelope surface A is the depth range of the near-surface model;

[0047] (2) Determination of the high-speed top interface and the floating reference surface; Select the high-speed top interface B above the envelope surface, and smooth the high-speed top interface B appropriately. Smooth the surface elevation interface with a large radius to obtain the floating surface C, see... Figure 1 .

[0048] (3) Determination of floating reference surface correction and seismic data correction; using the replacement velocity V_replacement in the work area, calculate the correction T_BC from the high-speed top surface B to the floating surface C, and correct the seismic single-shot data to the floating surface C of the ground elevation, see Figure 1 ;

[0049] (4) Mid-deep reflection wave velocity analysis and tomographic modeling: For seismic data on floating surface C, reflection wave velocity analysis is carried out and time-depth conversion is performed to obtain the mid-deep depth migration velocity.

[0050] (5) True surface velocity fusion starting surface unification and velocity trend analysis; after the analysis of the previous (1) (2) (3) (4) steps, the seismic data and the middle-deep reflection wave velocity are on the floating surface C, the near-surface velocity is on the true surface, and then the seismic data and the velocity are unified to the same surface. The specific method is: remove the correction amount T_BC applied to the seismic data in (2), perform small radius smoothing on the true surface to obtain a smoothing surface D, and calculate the correction amount T_D from the true surface to the small smoothing surface D. Apply T_D to the seismic data to correct the data to the small smoothing surface D. Similarly, the near-surface and middle-deep layer velocity is corrected to the small smoothing surface D, and the gap between the small smoothing surface D and the true surface is filled by velocity interpolation according to the nearest principle.

[0051] (6) Velocity gradient correction and inverse distance weighted fusion modeling; select the small smoothing surface D to the high-speed top surface B as the velocity fusion window, analyze the near-surface velocity and the middle-deep layer velocity trend in the window, and correct the near-surface velocity and the middle-deep layer velocity trend according to the well logging velocity trend. Then, select the inverse distance weighted function for velocity fusion.

[0052] (7) Seismic data and fused true surface velocity model output. Output the depth migration velocity model of the small smoothing surface, and then update and iterate the velocity according to the conventional prestack depth migration technology.

[0053] In a specific embodiment of the application, Figure 1 is a schematic diagram of the smoothing surface, the surface and the ray density envelope surface in an embodiment of the application. Figure 1 As can be seen from the figure, the part above the envelope surface A has higher ray coverage times and more uniform ray density, and the inverted near-surface model has higher reliability. Figure 1 The selection of the true surface, the floating surface C and the high-speed top B in the figure is the basis for subsequent velocity fusion surface unification and data correction.

[0054] Figure 2 is a schematic diagram of the reflection wave velocity model in an embodiment of the application; after the seismic data is corrected to the floating surface C with the replacement velocity, the velocity analysis is carried out according to the conventional processing method, the time domain velocity is converted into the depth domain, and the reflection wave depth migration velocity model is as shown in Figure 2 .

[0055] Figure 3 is a schematic diagram of the seismic data and the velocity correction to the small smoothing surface in an embodiment of the application; using the formula in the figure, the near-surface velocity and the reflection wave velocity are unified to the small smoothing surface D, and the velocity fusion can be seamlessly spliced to better realize the organic fusion of the two.

[0056] Figure 4It is the schematic diagram of the contrast analysis of the acoustic logging curve, near-surface and middle-deep longitudinal velocity trend in an embodiment of the present application; the left drawing is the acoustic logging curve, the middle drawing is the near-surface velocity, reflected wave velocity, fused velocity curve schematic diagram, and the right drawing is the fused area display.

[0057] Figure 5 It is the fused velocity model contrast schematic diagram in different data correction methods in an embodiment of the present application.

[0058] Figure 6 It is the model contrast schematic diagram in different mathematical algorithms for velocity fusion in an embodiment of the present application.

[0059] The true near-surface velocity fusion modeling method in double complex areas of the present application, aiming at the double complex work area, uses the near-surface velocity model obtained by the big gun first arrival tomographic inversion, and further develops the constrained tomographic inversion by using the micro-logging data to improve the inversion precision of the near-surface velocity model. The near-surface velocity credible depth range is selected according to the ray density bottom boundary depth. The seismic data is corrected, the reflected wave velocity analysis is carried out, and the middle-deep velocity model is obtained. The starting surface of the velocity and the seismic data is unified, the near-surface and middle-deep velocity trend is analyzed by using the logging data, and is corrected, finally the inverse distance weighted fusion modeling is used to obtain the higher precision pre-stack depth migration initial velocity model, so as to improve the quality and effect of the migration imaging.

[0060] Finally, it should be noted that: the above only describes the preferred embodiments of the present application, and is not used to limit the present application, although the present application is described in detail by referring to the foregoing embodiments, for those skilled in the art, the technical solutions recorded in the foregoing embodiments can be modified, or some technical features can be replaced. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application should be included in the protection scope of the present application.

[0061] In addition to the technical features described in the specification, they are known technologies for professional technicians.

Claims

1. A true surface velocity fusion modeling method for dual-complicated areas, characterized in that, The true surface velocity fusion modeling method of the double complex area comprises: Step 1, determining the depth reliable range of the near-surface model; Step 2, obtaining the high-speed top interface and the floating reference surface; Step 3, obtaining the floating reference surface correction amount and correcting the seismic data; Step 4, performing the reflection wave velocity analysis and tomographic modeling of the middle-deep layer; Step 5, performing the unification of the true surface velocity fusion starting surface and the velocity trend analysis; Step 6, performing the velocity gradient correction and inverse distance weighted fusion modeling; Step 7, outputting the seismic data and the fused true surface velocity model; In step 1, the ray density analysis is used to perform the tomographic inversion of the big cannon first arrival, the ray density bottom interface is taken as the envelope surface A, and the depth difference from the surface elevation to the envelope surface A is taken as the depth range of the near-surface model; In step 2, the high-speed top interface B is selected above the envelope surface, the high-speed top interface B is smoothed, the surface elevation interface is smoothed with a large radius to obtain the floating surface C; In step 3, the replacement velocity V of the work area is used to calculate the correction amount T_BC of the high-speed top interface B to the floating surface C, and the seismic single-shot data is corrected to the ground elevation floating surface C. replace(BC) , calculate the correction amount T_BC of the high-speed top interface B to the floating surface C, and correct the seismic single-shot data to the ground elevation floating surface C. In step 3, the formula for obtaining the floating reference surface correction amount and correcting the seismic data is: T float(BC) = T0 + C source(BC) + C detect(BC) where T0 represents the original single shot record, C source(BC) represents the shot point true surface to float surface correction calculated using the replacement velocity V replace(BC) represents the receiver point true surface to float surface correction calculated using the replacement velocity V detect(BC) represents the shot point true surface to float surface correction calculated using the replacement velocity V replace(BC) represents the receiver point true surface to float surface correction calculated using the replacement velocity V float(BC) represents the new shot record after the correction has been applied to the float surface C; In step 4, the reflection wave velocity analysis is performed on the seismic data on the floating surface C, and the time-depth conversion is performed to obtain the depth migration velocity of the middle-deep layer; In step 5, the seismic data and the velocity are unified to the same surface, first, the correction amount T_BC applied to the seismic data in step 3 is removed, the true surface is smoothed with a small radius to obtain the smoothed surface D, and the correction amount T_D from the true surface to the small smoothed surface D is calculated, T_D is applied to the seismic data to correct the data to the small smoothed surface D; similarly, the near-surface and middle-deep layer velocities are corrected to the small smoothed surface D, and the gap between the small smoothed surface D and the true surface is filled by using the nearest principle velocity interpolation.

2. The dual-complicated area true ground velocity fusion modeling method according to claim 1, characterized in that, In step 5, after the analysis of the previous steps 1-4, the seismic data and the middle-deep layer reflection wave velocity are on the floating surface C, and the near-surface velocity is on the true surface, then the seismic data and the velocity are unified to the same surface.

3. The dual-complicated area true ground velocity fusion modeling method according to claim 1, characterized in that, In step 5, the seismic data is unified to the true surface to the small smoothed surface D after the small smoothing radius, and the specific formula is: T float(BD) = T float(BC) - C source(BC) - C detect(BC) + C source(BD) + C detect(BD) The first three terms on the right side of the equation represent the return of the seismic record to the true representative surface, C source(BD) represents the correction amount of the shot point true surface to the small smooth surface T_D calculated by the replacement velocity V replace(BC) , C detect(BD) represents the correction amount of the receiver point true surface to the small smooth surface T_D calculated by the replacement velocity V replace(BC) , and T float(BD) represents the new shot record after the correction amount is applied to the small smooth surface D.

4. The dual-complicated area true ground velocity fusion modeling method according to claim 1, characterized in that, In step 6, the small smoothed surface D to the high-speed top surface B is selected as the velocity fusion window, the near-surface velocity and the middle-deep layer velocity trend in the window are analyzed, the well logging velocity trend is taken as the basis, the gradient method is used to correct the near-surface velocity and the middle-deep layer velocity trend, and then the inverse distance weighted function is selected for velocity fusion.

5. The dual-complicated area true ground velocity fusion modeling method according to claim 1, characterized in that, In step 7, the depth migration velocity model of the small smoothed surface is output, and then the velocity is updated and iterated according to the prestack depth migration method.

Images