Viscoelastic compensation parameter field modeling method and device for curved wide-line reflection seismic data

Abstract

The application provides a viscoelastic compensation parameter field modeling method and device for curved wide-line reflection seismic data, and relates to the technical field of earthquakes, and comprises the following steps: determining an inverse Q value offset gather according to seismic data corresponding to a curved wide-line work area and a preset Q value; coupling determination of a target equivalent Q value and a high cutoff frequency corresponding to an analysis window based on a preset analysis window on an inverse Q value offset profile, taking an acceptable noise level threshold of the curved wide-line work area as a constraint, and the analysis window is determined based on the banded curved distribution characteristics of common midpoint; and performing interpolation processing on the target equivalent Q value and the high cutoff frequency based on the banded curved distribution characteristics to construct a viscoelastic compensation parameter field corresponding to the curved wide-line work area. The application can solve the core problems that the existing equivalent Q value modeling method based on surface reflection data cannot match the banded distribution and structural characteristics of the curved wide-line CMP due to the adaptation of straight-line measurement, resulting in inaccurate Q value picking and Q field modeling distortion.

Description

Technical Field

[0001] This invention relates to the field of earthquake technology, and in particular to a method and apparatus for modeling viscoelastic compensation parameter fields of flexural width line reflection seismic data. Background Technology

[0002] In oil and gas seismic exploration, the curved-width line acquisition technique has become a core acquisition solution in complex areas due to its ability to flexibly adapt to complex surface obstacles (such as mountains and loess plateaus) and to construct a wide-coverage observation system through multiple receiver lines, thereby improving the signal-to-noise ratio while controlling construction costs. However, the viscoelasticity of the subsurface medium leads to attenuation of high-frequency components and phase distortion of seismic waves, severely reducing the imaging resolution of deep reservoirs. Furthermore, the common center point (CMP) of the curved-width line exhibits an irregular, banded distribution, which is fundamentally different from the uniform straight distribution of traditional straight-line survey lines, posing a challenge to Q (Quality Factor) value field modeling based on reflection data.

[0003] The Q-value field is the core foundation of viscoelastic compensation (inverse Q-filtering, viscoelastic migration), and accurate modeling directly determines imaging resolution. Traditional Q-value estimation relies heavily on VSP (Vertical Seismic Profile) or inter-well data. These types of data are scarce and have limited coverage, making it difficult to meet the non-uniform Q-value modeling requirements of large-scale 3D / bend-width line survey areas. However, the equivalent Q-value modeling method based on surface reflection data has achieved mature application in straight-line survey areas due to its significant advantages: First, it only requires surface reflection seismic data and does not rely on scarce VSP or inter-well data, making it suitable for large-scale industrial exploration; Second, through multi-channel spectrum stacking and spectral envelope analysis, it effectively avoids the reference layer selection difficulties and thin-layer tuning interference faced by traditional spectral ratio methods, resulting in higher Q-value picking stability; Third, the equivalent Q-value changes smoothly without abrupt changes, which can significantly reduce the scanning range, improve picking accuracy, and can complete unscanned areas through correlation interpolation with the velocity field, efficiently constructing the Q-value field for the entire survey area. However, this equivalent Q-value modeling method assumes that the CMP (Common Midpoint) is distributed along a straight line. Its core design logic is based on straight-line topology optimization, which cannot adapt to the banded distribution and structural bending characteristics of the curved line. This results in inaccurate Q-value picking and distorted Q-field modeling when applied directly. It is urgent to optimize it specifically to adapt to the unique needs of curved line reflection seismic data. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a viscoelastic compensation parameter field modeling method and apparatus for curved-width line reflection seismic data, which can solve the core problem of existing equivalent Q-value modeling methods based on surface reflection data being unable to fit the zonal distribution and structural characteristics of curved-width lines (CMP) due to their adaptability to straight survey lines. This results in inaccurate Q-value picking and distorted Q-field modeling.

[0005] In a first aspect, the present invention provides a viscoelastic compensation parameter field modeling method for seismic data with a flexural width line reflection, comprising: Based on the seismic data corresponding to the curve width line work area and the preset Q value, the inverse Q value migration gather is determined. The inverse Q value migration gather includes the inverse Q value migration profile corresponding to the preset Q value. Based on the pre-set analysis window on the inverse Q-value offset profile, and constrained by the acceptable noise level threshold of the bend width line work area, the target equivalent Q-value and high cutoff frequency corresponding to the analysis window are coupled and determined. The analysis window is determined based on the strip-shaped bend distribution characteristics of the common center point. Interpolation processing based on the characteristics of the strip-shaped bending distribution is performed on the target equivalent Q value and high cutoff frequency to construct the viscoelastic compensation parameter field corresponding to the bending width line work area. The viscoelastic compensation parameter field includes the target equivalent Q value field and the target high cutoff frequency field.

[0006] In one implementation, the inverse Q-value migration gather is determined based on the seismic data corresponding to the bend width line work area and a preset Q value: Identify high-frequency bad paths in the seismic data corresponding to the bend width line work area, so as to remove high-frequency bad paths from the seismic data; Based on the preset Q value, the effective seismic traces after removing high-frequency bad traces in the seismic data are subjected to inverse Q-value filtering and pre-stack time migration processing to obtain the inverse Q-value migration trace set.

[0007] In one implementation, identifying high-frequency bad channels in the seismic data corresponding to the bend width line work area includes: Fourier transform processing is performed on all original seismic traces contained in the seismic data corresponding to the bend width line work area to obtain the dominant frequency of the seismic data. Based on the dominant frequency, the highest frequency to be compensated, and the lowest frequency of the seismic data, the high frequency band and the low frequency band are divided. For any original seismic trace, determine the ratio between the total energy of the original seismic trace in the high-frequency band and the total energy in the low-frequency band to obtain the energy ratio corresponding to the original seismic trace; The bad track identification threshold is determined based on the energy ratio corresponding to all original seismic traces. For any original seismic trace, if the energy ratio corresponding to the original seismic trace is higher than the bad trace identification threshold, the original seismic trace is determined to be a high-frequency bad trace.

[0008] In one implementation, a bad trace identification threshold is determined based on the energy ratio corresponding to all original seismic traces, including: The energy ratios corresponding to all original seismic traces are compiled into an energy ratio sequence to determine the median of the energy ratio sequence; Determine the absolute deviation between the energy ratios contained in the energy ratio sequence and the median of the energy ratio sequence, and take the median of the absolute deviation sequence composed of all absolute deviations as the median absolute deviation; The bad sector identification threshold is determined based on the median and median absolute deviation of the energy ratio sequence.

[0009] In one implementation, based on a preset Q-value, the effective seismic traces in the seismic data after removing high-frequency bad channels are subjected to inverse Q-value filtering and pre-stack time migration processing to obtain an inverse Q-value migration gather, including: Perform the following operation on any valid seismic trace after removing high-frequency bad traces from the seismic data: Based on the preset Q value, the effective seismic trace is subjected to inverse Q-value filtering to obtain the inverse Q-value filtered data corresponding to the effective seismic trace. For all inverse Q-value filtered data of the effective seismic trace, pre-stack time migration processing is performed in batches to obtain the imaging results corresponding to the effective seismic trace. Based on the preset Q value, the imaging results corresponding to all valid seismic traces are classified into the corresponding data buffers. The imaging results stored in the data buffers with the preset Q value are sorted, repositioned and processed into gathers to obtain inverse Q migration gathers.

[0010] In one implementation, based on a preset analysis window on the inverse Q-value offset profile, and constrained by an acceptable noise level threshold for the bend width line work area, the target equivalent Q-value and high cutoff frequency corresponding to the analysis window are coupled and determined, including: Perform the following operation on any preset analysis window on the inverse Q-value offset profile: Based on the in-phase axis tilt angle within the analysis window, a local fitting algorithm is used to perform signal-to-noise separation on the inverse Q-value offset profile, and the noise level distribution of the inverse Q-value offset profile within the analysis window is obtained. If the noise level distribution of the inverse Q-value offset profile within the analysis window is less than or equal to the acceptable noise level threshold of the bend width line work area, the preset Q value corresponding to the inverse Q-value offset profile is used as the candidate equivalent Q value. With the goal of maximizing the effective spectral width of the inverse Q-value offset profile, the target equivalent Q-value corresponding to the analysis window is determined from the candidate equivalent Q-values, and the high cutoff frequency is determined based on the frequency points in the inverse Q-value offset profile corresponding to the target equivalent Q-value.

[0011] In one implementation, the target equivalent Q value and high cutoff frequency are interpolated based on the characteristics of the banded bending distribution to construct the viscoelastic compensation parameter field corresponding to the bending width line work area, including: Perform the following operation on any common center point where analysis windows are set up: By performing time-linear interpolation on the target equivalent Q value and high cutoff frequency corresponding to the analysis window set at the common center point, the equivalent Q value sequence and high cutoff frequency sequence continuously distributed along the time depth direction of the common center point are obtained; The cumulative chord length difference algorithm is used to perform lateral interpolation of the equivalent Q value sequence and high cutoff frequency sequence corresponding to the common center point, so as to obtain the initial equivalent Q value field and initial high cutoff frequency field of the common center point along the strip-shaped bending distribution characteristics. The initial equivalent Q-value field and the initial high-frequency cutoff field are smoothed to obtain the viscoelastic compensation parameter field corresponding to the bending width line work area.

[0012] Secondly, the present invention also provides a viscoelastic compensation parameter field modeling device for flexural width line reflection seismic data, comprising: The gather construction module is used to determine the inverse Q-value migration gather based on the seismic data corresponding to the curve width line work area and the preset Q value. The inverse Q-value migration gather includes the inverse Q-value migration profile corresponding to the preset Q value. The Q-value and high cutoff frequency determination module is used to determine the target equivalent Q-value and high cutoff frequency corresponding to the analysis window based on the preset analysis window on the inverse Q-value offset profile, with the acceptable noise level threshold of the bend width line work area as a constraint. The analysis window is determined based on the strip-shaped bend distribution characteristics of the common center point. The parameter field construction module is used to perform interpolation processing on the target equivalent Q value and high cutoff frequency based on the characteristics of the strip-shaped bending distribution, so as to construct the viscoelastic compensation parameter field corresponding to the bending width line work area. The viscoelastic compensation parameter field includes the equivalent Q value field and the high cutoff frequency field.

[0013] Thirdly, the present invention also provides an electronic device including a processor and a memory, the memory storing computer-executable instructions executable by the processor, the processor executing the computer-executable instructions to implement any of the methods provided in the first aspect.

[0014] Fourthly, the present invention also provides a computer-readable storage medium storing computer-executable instructions, which, when invoked and executed by a processor, cause the processor to implement any of the methods provided in the first aspect.

[0015] This invention provides a method and apparatus for modeling a viscoelastic compensation parameter field from reflection seismic data along a bend-width line. First, based on the seismic data corresponding to the bend-width line work area and a preset Q-value, an inverse Q-value migration gather is determined. The inverse Q-value migration gather includes an inverse Q-value migration profile corresponding to the preset Q-value. Then, based on a preset analysis window on the inverse Q-value migration profile, and constrained by an acceptable noise level threshold for the bend-width line work area, the target equivalent Q-value and high cutoff frequency corresponding to the analysis window are coupled and determined. The analysis window is determined based on the banded bend distribution characteristics of common midpoints. Finally, interpolation processing based on the banded bend distribution characteristics is performed on the target equivalent Q-value and high cutoff frequency to construct the viscoelastic compensation parameter field corresponding to the bend-width line work area. The viscoelastic compensation parameter field includes a target equivalent Q-value field and a target high cutoff frequency field. This invention addresses the core problem of existing equivalent Q-value modeling methods based on surface reflection data. These methods are adapted to straight survey lines but cannot accurately match the zonal distribution and structural features of curved lines (CMP), leading to inaccurate Q-value picking and distorted Q-field modeling. This invention retains the core advantages of equivalent Q-value modeling and achieves accurate modeling through targeted optimization of curved lines.

[0016] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained through the structures particularly pointed out in the description and the drawings.

[0017] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0018] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 A flowchart illustrating a viscoelastic compensation parameter field modeling method for seismic data with a flexural width line reflection provided in an embodiment of the present invention; Figure 2 A flowchart illustrating another viscoelastic compensation parameter field modeling method for reflection seismic data with a bend width line provided in an embodiment of the present invention; Figure 3A schematic diagram of the CMP point distribution for a bend width line provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of a conventional pre-stack time-migrated local imaging profile provided in an embodiment of the present invention; Figure 5 This invention provides an equivalent Q selection window and a schematic diagram of the Q value and high cutoff frequency F3 picking results. Figure 6 A schematic diagram of the equivalent Q-value field of a bend width line provided in an embodiment of the present invention; Figure 7 A schematic diagram of a two-dimensional bend width line viscoelastic pre-stack time-migrating local imaging profile based on equivalent Q value is provided for an embodiment of the present invention. Figure 8 This is a schematic diagram of a local imaging profile spectrum comparison provided in an embodiment of the present invention; Figure 9 This is a magnified comparison image of the imaging profile details provided in an embodiment of the present invention; Figure 10 This is a schematic diagram of a viscoelastic compensation parameter field modeling device for flexural width line reflection seismic data provided in an embodiment of the present invention; Figure 11 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Existing technologies mainly fall into two categories: one is the equivalent Q-value field modeling method for straight survey lines based on surface reflection data, and the other is the conventional data processing method for curved and wide survey lines. While the equivalent Q-value modeling method for straight survey lines possesses the aforementioned advantages in straight survey areas, it suffers from adaptability defects when applied to curved and wide survey lines. (1) High-frequency noise control is not adapted to the characteristics of the curve width line data: The observation geometry of different offset groups of the curve width line is large and the noise level distribution is more complex. Existing methods do not finely remove high-frequency bad channels, which easily leads to high-frequency noise channels remaining, affecting the accuracy and stability of the equivalent Q value scan.

[0022] (2) Poor compatibility between CMP points and window selection: The existing method arranges CMP points at equal intervals in a straight line, and the horizontal range of the window is based on the design of the straight line survey line. However, the curved line CMP is distributed in a strip-shaped curved distribution. The straight CMP points cannot fit the structural trend. The curved section with large structural changes is prone to the window crossing the fault / structure, resulting in the superposition of multiple spectrums and the mixing of invalid signals, which greatly reduces the weakening effect of thin-layer tuning.

[0023] (3) Q field interpolation does not match the topology of the curved line: The existing method uses straight-line topology interpolation to fill the Q value in the unscanned area, but the center line of the curved line CMP is curved. Straight-line interpolation will force the Q value to change along a straight line, which is out of sync with the actual absorption characteristics of the underground strata, resulting in a distorted Q field characterization and failing to reflect the local absorption differences in the curved section.

[0024] Furthermore, there is currently no publicly available technology that combines the mature advantages of equivalent Q-value modeling for straight test lines with the distribution characteristics of reflection data from curved and wide lines, and there is a lack of high-precision Q-value field modeling schemes that are adapted to the topology and structure of curved and wide lines.

[0025] Based on this, the present invention provides a viscoelastic compensation parameter field modeling method and apparatus for curved-width line reflection seismic data, which can solve the core problem of existing equivalent Q-value modeling methods based on surface reflection data being unable to fit the zonal distribution and structural characteristics of curved-width lines (CMP) due to their adaptability to straight survey lines. This results in inaccurate Q-value picking and distorted Q-field modeling.

[0026] To facilitate understanding of this embodiment, a viscoelastic compensation parameter field modeling method for flexural width line reflection seismic data disclosed in this embodiment of the invention will first be described in detail. (See [link to relevant documentation]). Figure 1 The diagram shows a flowchart of a viscoelastic compensation parameter field modeling method for flexural width line reflection seismic data. The method mainly includes the following steps S102 to S106: Step S102: Determine the inverse Q-value migration gather based on the seismic data corresponding to the curve width line work area and the preset Q value.

[0027] Among them, the inverse Q-value migration gather includes the inverse Q-value migration profile corresponding to the preset Q-value. The Q-value is also the quality factor value. The inverse Q-value migration profile refers to the two-dimensional reflection imaging result obtained after performing pre-stack time migration processing on all inverse Q-filtered seismic traces under the same preset equivalent quality factor Q-value, and completing the repositioning and stacking based on the common center point coordinates.

[0028] In one implementation, seismic data corresponding to the bend width line work area is acquired, high-frequency bad channels in the seismic data are extracted, and then the effective seismic traces after removing high-frequency bad channels in the seismic data are subjected to inverse Q-value filtering and pre-stack time migration processing according to a preset Q-value to generate a series of inverse Q-value migration profiles related to the preset Q-value, which constitute an inverse Q-value migration set.

[0029] Step S104: Based on the preset analysis window on the inverse Q-value offset profile, and constrained by the acceptable noise level threshold of the bend width line work area, the target equivalent Q-value and high cutoff frequency corresponding to the analysis window are coupled and determined.

[0030] The analysis window is determined based on the banded curvature distribution characteristics of the common center point (CMP). The banded curvature distribution characteristics refer to the fact that the common center point (CMP) of the curve width line presents a composite spatial distribution pattern in the plane space, which is both a horizontally distributed banded structure (formed by multiple receiving line widths) and a continuous longitudinal curvature along the obstacle avoidance path on the ground. This feature is reflected in the fact that the CMP point group has non-zero horizontal width, non-zero curvature and monotonic spatial continuity along the survey line direction, which is different from the single linear distribution of straight survey lines, as well as discrete point clusters or regular grid distributions.

[0031] In one example, based on the inverse Q-value migration profile, an analysis window is selected that matches the common center point banded curvature distribution characteristics of the bend width line. Based on the coupling characteristics of the equivalent Q-value and the high cutoff frequency F3, and using the acceptable noise level of the bend width line work area as a constraint, the target equivalent Q-value and high cutoff frequency F3 corresponding to each analysis window are simultaneously determined, combined with the spectral width and imaging quality of the inverse Q-value migration profile. This embodiment of the invention improves imaging resolution while ensuring the signal-to-noise ratio of seismic imaging, providing parameter basis for the subsequent construction of the equivalent quality factor Q field and the high cutoff frequency F3 field.

[0032] Step S106: The target equivalent Q value and high cutoff frequency are interpolated based on the characteristics of the banded bending distribution to construct the viscoelastic compensation parameter field corresponding to the bending width line work area.

[0033] The viscoelastic compensation parameter field includes the target equivalent Q-value field and the target high-frequency cutoff field. In one example, using the target equivalent Q-value and high-frequency cutoff determined by each analysis window as discrete nodes, an interpolation method that conforms to the banded bending distribution characteristics of the bend width line (CMP) is adopted. Time-direction interpolation, bend width line lateral interpolation, and spatial smoothing are performed sequentially to finally form a continuously distributed target equivalent Q-value field and target high-frequency cutoff field for the bend width line work area, providing continuous parameter support for the viscoelastic compensation processing of bend width line reflection seismic data.

[0034] The viscoelastic compensation parameter field modeling method for curved-width line reflection seismic data provided in this invention determines the analysis window based on the characteristics of the zonal curved distribution, determines the target equivalent Q value and high cutoff frequency by coupling and constraining the noise level, and constructs the target equivalent Q value field and target high cutoff frequency field by using an interpolation method that fits the characteristics of the zonal curved distribution. This solves the core problem of existing equivalent Q value modeling methods based on surface reflection data, which are adapted to straight survey lines and cannot fit the zonal distribution and structural characteristics of curved-width lines (CMP), resulting in inaccurate Q value picking and distorted Q field modeling. This invention retains the core advantages of equivalent Q value modeling and achieves accurate modeling through targeted optimization of curved-width lines.

[0035] For ease of understanding, this invention provides a specific implementation method for modeling the viscoelastic compensation parameter field of flexural width line reflection seismic data. See [link to relevant documentation]. Figure 2 The flowchart shown is a different method for modeling the viscoelastic compensation parameter field of reflection seismic data with a curved width line, including the following steps 1 to 5: Step 1: Acquire pre-stack reflection seismic data along the bend width line, perform migration preprocessing, acquire the migration velocity field, and generate a conventional pre-stack time migration profile; Step 2: Perform high-frequency bad path removal on the pre-stack reflection seismic data of the bend width line, and set the scan range and number of preset Q values; Step 3: Based on the pre-stack reflection seismic data of the curved width line after removing high-frequency bad channels and the preset Q value, inverse Q filtering is performed sequentially on each valid seismic trace, and pre-stack time migration is performed in batches to generate inverse Q migration trace sets; Step 4: Based on the conventional pre-stack time migration profile and inverse Q migration gather, select the analysis window along the common midpoint of the bend width line in the strip-shaped bend path, and couple and determine the target equivalent Q value and high cutoff frequency F3 corresponding to each analysis window; Step 5: Using the target equivalent Q value and high cutoff frequency F3 determined by each analysis window as discrete nodes, an interpolation algorithm (such as cumulative chord length interpolation algorithm) that fits the CMP strip curved path is used to construct a continuously distributed target equivalent Q value field and target high cutoff frequency field throughout the entire work area.

[0036] While retaining the advantages of equivalent Q-value modeling—no well data required, stable picking, and efficient modeling—this invention achieves precise adaptation to the zonal bending distribution characteristics of the bend-width line CMP. The constructed target equivalent Q-value field has a significantly improved matching degree with the absorption characteristics of the subsurface medium, providing reliable support for viscoelastic migration and is suitable for high-resolution seismic imaging of bend-width line work areas under complex surface conditions.

[0037] To address the applicability issue of the viscoelastic pre-stack time migration method based on equivalent Q-values ​​with two-dimensional bend width line data, and to achieve accurate viscoelastic compensation under complex bend width line conditions, the specific implementation process of the viscoelastic compensation parameter field modeling method for bend width line reflection seismic data is as follows: Step 1: Acquire pre-stack reflection seismic data along the bend width line, perform migration preprocessing, obtain the migration velocity field, and generate a conventional pre-stack time migration profile.

[0038] Specifically, pre-processing of 2D curved-width pre-stack seismic data migration is performed, encompassing raw data decoding, observation system definition, static correction, pre-stack denoising, amplitude compensation, and deconvolution. The core of observation system definition is the accurate positioning of the common center point (CMP): a virtual CMP centerline is automatically calculated based on the coordinates of the shot and receiver points, and manual adjustments are made in sections with significant line curvature, taking into account geological structural trends, to ensure that this centerline closely matches the center of the reflected wave energy scattering points. Pre-stack denoising employs FK filtering and coherent noise attenuation methods to suppress surface waves and random interference while preserving the amplitude and phase information of effective reflected waves. Amplitude compensation corrects amplitude differences between different receiver lines and different offset groups through surface consistency correction. Deconvolution is used to compress seismic wavelet lengths and improve the longitudinal resolution of the data.

[0039] In the velocity analysis phase, CMP gathers are extracted at preset intervals, and dynamic correction velocity picking and spatial smoothing are performed to establish an initial migration velocity field. Subsequently, based on the pre-stack time migration results, velocity parameters are iteratively optimized according to the in-phase axis flatness to obtain a high-precision migration velocity field. Simultaneously, based on the stretching characteristics of the migration gathers at each CMP point, the wavelet stretching rate at different time depths is statistically analyzed, and stretching cut parameters are set accordingly. Based on the optimized migration velocity field, the Kirchhoff pre-stack time migration algorithm is used to calculate travel times based on the geometric topological characteristics of the bend width line observation system, and weighted stacking is performed to generate a conventional pre-stack time migration profile.

[0040] Step 2: Perform high-frequency bad path removal on the pre-stack reflection seismic data of the bend width line, and set the scan range and number of preset Q values.

[0041] Wide-beam seismic data acquisition often traverses complex terrains such as mountains and loess plateaus. It is affected by surface obstacles, lateral heterogeneity of near-surface low-velocity zones, and the zonal distribution of the observation system, resulting in a low signal-to-noise ratio of seismic data. Surface waves, scattered waves, and other interference waves have strong energy and complex spatial distribution. After being superimposed with random noise and multiple waves, the effective reflection signal is significantly suppressed, which increases the difficulty of identifying high-frequency bad channels and accurately estimating the equivalent quality factor Q value.

[0042] First, identify high-frequency bad channels in the seismic data corresponding to the bend-width line work area, and remove these bad channels from the seismic data. Specifically, this includes: (1) Perform Fourier transform processing on all original seismic traces contained in the seismic data corresponding to the bend width line work area to obtain the dominant frequency of the seismic data. Based on the dominant frequency, the highest frequency to be compensated, and the lowest frequency of the seismic data, divide the high-frequency band and the low-frequency band. For example, the high-frequency band is the dominant frequency. Up to the highest frequency of proposed compensation The frequency range, with the low-frequency band representing the lowest frequency of seismic data. To main frequency The frequency range.

[0043] (2) For any original seismic trace, determine the ratio between the total energy in the high-frequency band and the total energy in the low-frequency band of the original seismic trace to obtain the energy ratio corresponding to the original seismic trace. The formula for calculating the energy ratio is shown below: ; in, Indicates the first The energy ratio of the high-frequency band to the low-frequency band of the seismic trace. Indicates the first The frequency corresponding to the earthquake trace is The amplitude of the amplitude spectrum. Exponential weighting can amplify the energy ratio difference between the high-frequency noise channel and the effective channel. Under low signal-to-noise ratio, the high-frequency noise energy of the high-frequency bad channel is abnormally prominent. The effective channel is significantly larger than the effective channel, while the effective channel is relatively stable due to low-frequency reflections. Keep it within a reasonable range.

[0044] (3) Determine the bad trace identification threshold based on the energy ratios corresponding to all original seismic traces. In this embodiment of the invention, a robust statistical method based on the median Med(r) and median absolute deviation (MAD) is used to set the threshold. The median absolute deviation (MAD) is calculated for the energy ratio sequence of all seismic traces. Robust statistics that are simple to calculate and resistant to extreme noise interference are suitable for low signal-to-noise ratio data characteristics of curved lines. Specifically, they include: (3.1) Compile the energy ratios corresponding to all original seismic traces into an energy ratio sequence to determine the median of the energy ratio sequence. Specifically, compile the energy ratios of all seismic traces... , ,..., (n is the total number of seismic traces), sorted in ascending order; if n is odd, take the value of the (n+1) / 2th position after sorting; if n is even, take the average of the values ​​of the (n / 2)th and (n / 2)+1th positions after sorting, which is the median Med(r) of the energy ratio sequence.

[0045] (3.2) Determine the absolute deviation between the energy ratios included in the energy ratio sequence and the median of the energy ratio sequence, and take the median of the absolute deviation sequence consisting of all absolute deviations as the median absolute deviation. Specifically, for each energy ratio... Calculate the absolute difference between it and Med(r), i.e. , to obtain the absolute deviation sequence Sort the above absolute deviation sequence in ascending order, and take the value of the middle position again (take the middle value for odd positions and the average of the middle two for even positions). This value is the median absolute deviation (MAD).

[0046] (3.3) Determine the bad sector identification threshold based on the median and median absolute deviation of the energy ratio sequence. For example, the expression for the bad sector identification threshold is as follows: .

[0047] (4) For any original seismic trace, if the energy ratio corresponding to the original seismic trace is higher than the bad trace identification threshold, the original seismic trace is determined to be a high-frequency bad trace. Specifically, if the energy ratio of a certain seismic trace is higher than the bad trace identification threshold, the original seismic trace is determined to be a high-frequency bad trace. If the bad trace exceeds the bad trace identification threshold, it is identified as a high-frequency bad trace and removed. Seismic traces that do not exceed the bad trace identification threshold are retained and recorded as valid seismic traces.

[0048] It should be noted that, under the technical premise of adapting to the noise characteristics of the bend width line, the robust statistical method composed of the median and the median absolute deviation (MAD) can be replaced by other statistical methods with equivalent robustness, as long as the method still aims to characterize the noise distribution characteristics of the bend width line seismic data, it falls within the protection scope of this invention.

[0049] Second, set the scan range and number of preset Q values: Before generating the inverse Q-offset gather and performing Q-value picking, the Q-value scanning range and number must be determined. The Q-value scanning range needs to be adapted to the low signal-to-noise ratio characteristics of the curved-width line and the formation absorption pattern: the scanning range is set to 50-500, with 50-150 covering shallow high-absorbing strata (such as loess and fractured zones), and 150-500 covering medium-deep low-absorbing strata (such as tight sandstone and carbonate rocks), ensuring that the true Q-value distribution of strata at different depths is covered and avoiding Q-value picking deviations due to low signal-to-noise ratio; the number of scans is set to 10-15, uniformly selected at 1 / Q intervals, which can ensure the Q-value scanning density to avoid missing the optimal compensation Q-value, and will not increase the computational load due to too many scans, providing a reasonable parameter basis for subsequent inverse Q-filtering.

[0050] Step 3: Based on the pre-stack reflection seismic data with curved width lines after removing high-frequency bad channels and the preset Q value, inverse Q filtering is sequentially performed on each valid seismic trace, and pre-stack time migration is performed in batches to generate inverse Q migration gathers. This can also be described as: based on the preset Q value, inverse Q-value filtering and pre-stack time migration are performed on the valid seismic traces in the seismic data after removing high-frequency bad channels to obtain inverse Q migration gathers.

[0051] In this embodiment of the invention, based on the Q-value scanning range, number of scans, and seismic data after removing high-frequency bad channels determined in S102, a processing mode combining single-trace full-Q-value inverse Q-filtering and batch pre-stack time migration is adopted: inverse Q-filtering corresponding to each preset Q-value is sequentially performed on each valid seismic trace, and the filtering results are temporarily stored in memory; then, for all filtered data of the same seismic trace in memory, the basic parameters such as travel time, geometric weight, and geometric diffusion correction factor are shared, and batch pre-stack time migration is performed; thereby reducing data read / write operations and storage overhead, and ensuring that the basic parameters only need to be calculated once, finally generating an inverse Q-value migration trace set that corresponds one-to-one with each preset Q-value.

[0052] The core processing logic of this step is to iterate through all valid seismic traces within the program, and for each individual seismic trace to be processed, perform continuous processing of full Q-value filtering and batch migration in sequence. The specific process is as follows: Retrieve all valid seismic traces, enable single-trace looping within the program, and perform the following operations on the valid seismic traces: (1) Based on the preset Q value, the effective seismic trace is subjected to inverse Q-value filtering to obtain the inverse Q-value filtered data corresponding to the effective seismic trace.

[0053] Specifically, based on N preset Q values, inverse Q-filtering is performed sequentially for each preset Q value corresponding to the effective seismic trace. All filtering results are temporarily stored in memory, forming a single-trace inverse Q-filtered data set of size N. The inverse Q-value filtering process is as follows: If the Q value of the subsurface medium does not vary spatially, the seismic traces received at the surface can be obtained. The formula for the inverse Q filter: ; in, For earthquake channels The frequency domain signal after inverse Q filtering, For earthquake channels The spectrum, Angular frequency, For earthquake channels The main frequency. For the current effective seismic trace, N preset Q values ​​are substituted sequentially, and the amplitude attenuation compensation and phase dispersion correction are performed on the effective seismic trace using the inverse Q filtering formula to recover the high-frequency effective signal. All filtering results are temporarily stored in memory.

[0054] (2) Perform pre-stack time migration processing on all inverse Q-value filtered data of the effective seismic trace in batches to obtain the imaging results corresponding to the effective seismic trace.

[0055] Specifically, conventional pre-stack time migration calculations are performed in batches on the N inverse Q-filtered data sets of the valid seismic traces temporarily stored in memory. During this process, basic parameters such as travel time from the shot-receiver point to each imaging point, observation system geometric weights, and geometric spread correction factors only need to be calculated once, eliminating the need for repeated calculations for each preset Q value, thus significantly reducing the computational load. The N imaging results obtained from the batch migration are then assigned to the corresponding imaging result caches for their respective preset Q values. The pre-stack time migration process is as follows: For the N inverse Q-filtered frequency domain signals of this channel in memory, perform batch Kirchhoff pre-stack time migration imaging calculations. The basic parameters such as travel time and geometric weights are shared throughout the process. The pre-stack time migration result can be expressed as: ; in, Let i be the coordinates of the i-th CMP point. For the vertical two-way travel time of imaging point I, For the i-th CMP point at The imaging amplitude value at time 10:00. , These represent the travel times from the shot point and the receiver point to the imaging point, respectively. The frequency domain signal after inverse Q filtering The semi-derivative of .

[0056] (3) Based on the preset Q value, the imaging results corresponding to all valid seismic traces are classified into the corresponding data buffers. The imaging results stored in the data buffers with the preset Q value are sorted, repositioned and processed to obtain the inverse Q migration gather.

[0057] Following this logic, all valid seismic traces are covered in a loop. After the entire gather is processed, the imaging results of all seismic traces corresponding to each Q value are sorted, repositioned, and gathered according to the CMP point coordinates, ultimately resulting in a series of inverse Q-migrated gathers related to the preset Q value.

[0058] In this embodiment of the invention, the above processing mode is adapted to the processing efficiency requirements of curved-width seismic data. By temporarily storing intermediate data in memory, data I / O interaction is reduced. At the same time, relying on the parameter sharing design of batch migration, the amount of repeated calculations is greatly reduced. The generated inverse Q migration trace image features show obvious differences with the preset Q value, providing high-quality, one-to-one corresponding trace data support for subsequent Q value optimization based on imaging quality.

[0059] Step 4: Based on the conventional pre-stack time migration profile and inverse Q migration gather, select the analysis window along the common midpoint of the bend width line in the strip-shaped bend path, and couple and determine the target equivalent Q value and high cutoff frequency F3 corresponding to each analysis window.

[0060] In this embodiment of the invention, based on conventional pre-stack time migration profiles and inverse Q-value migration profiles, an analysis window for the strip-shaped bending distribution of CMP points with suitable bending width is selected. Considering the coupling relationship between the equivalent Q-value and the high cutoff frequency F3, the acceptable noise level of the work area is taken as the core constraint. Combining the spectral width of the inverse Q-value migration profile with the actual imaging quality, the equivalent Q-value and the high cutoff frequency F3 are determined simultaneously. While ensuring the signal-to-noise ratio of seismic imaging, the profile resolution is improved, providing core basic parameters for subsequent equivalent Q-field inversion and high cutoff frequency field interpolation.

[0061] Specifically, perform the following operations on any preset analysis window on the inverse Q-value offset profile: (1) Select the curve width line adaptation analysis window: Based on conventional pre-stack time migration profiles and inverse Q-value migration profiles, three to six approximately equidistant common midpoints are selected, and additional pick-up points are added near the common midpoints with large curvature, which together serve as the selected common midpoints. At each selected common midpoint, four to six two-dimensional analysis windows are arranged along the time depth direction. Each analysis window covers at least fifteen consecutive curved line common midpoints laterally and contains several consecutive in-phase axes longitudinally, without crossing faults or large tectonic zones. The in-phase axes within the window have identifiable amplitude and morphological changes.

[0062] (2) Based on the in-phase axis tilt angle within the analysis window, the local fitting algorithm is used to perform signal-to-noise separation on the inverse Q-value offset profile to obtain the noise level distribution of the inverse Q-value offset profile within the analysis window.

[0063] In one specific implementation, a uniform acceptable noise level threshold is set for the work area. The tilt angle of the in-phase axis within each analysis window is picked up, and a local second-order polynomial approximation is applied to perform signal-to-noise separation on the inverse Q-value offset profiles for different preset Q-values. This yields the noise level distribution of the inverse Q-value offset profiles for each preset Q-value within the analysis window. The noise level distribution can be expressed as: ; in, The noise level distribution of a certain analysis window in the inverse Q-value offset profile of the k-th preset Q-value is represented by a dimensionless parameter. This indicates the dominant frequency of the superimposed spectrum within the analysis window. This indicates the maximum frequency corresponding to -20 dB in the superimposed spectrum within the analysis window. This represents the noise spectrum obtained after local polynomial filtering. This represents the spectrum of the effective signal remaining after filtering.

[0064] (3) If the noise level distribution of the inverse Q-value offset profile within the analysis window is less than or equal to the acceptable noise level threshold of the bend width line work area, the preset Q value corresponding to the inverse Q-value offset profile is used as a candidate equivalent Q value. In this embodiment of the invention, based on the obtained noise level distribution, preset Q values ​​with noise levels not exceeding the acceptable noise level threshold are selected as candidate equivalent Q values, and other preset Q values ​​are excluded to ensure that the selected candidate equivalent Q values ​​meet the signal-to-noise ratio requirements of the work area.

[0065] (4) With the goal of maximizing the effective spectral width of the inverse Q-value offset profile, the target equivalent Q-value corresponding to the analysis window is determined from the candidate equivalent Q-values, and the high cutoff frequency is determined based on the frequency points in the inverse Q-value offset profile corresponding to the target equivalent Q-value.

[0066] In one implementation, among the candidate equivalent Q values, the Q value that maximizes the effective spectral width of the corresponding inverse Q-offset profile is selected as the target equivalent Q value; or, among the candidate equivalent Q values, the target equivalent Q value is selected by comprehensively considering the effective spectral width of the corresponding inverse Q-offset profile, the continuity of the phase axis, the wavelet sidelobe separation effect, the resolvability of thin reservoirs, and the fidelity of structural morphology; the frequency point in the inverse Q-offset profile corresponding to the target equivalent Q value where the noise level is equal to a preset acceptable noise level threshold is determined as the high cutoff frequency F3 corresponding to the analysis window, thereby achieving the coupling determination of the target equivalent Q value and the high cutoff frequency F3.

[0067] It should be noted that, under the technical premise of determining the equivalent Q value of the target and the high cutoff frequency F3 coupling, the evaluation index on which the parameter selection is based can be simplified. As long as the simplification still aims to achieve a balance between the signal-to-noise ratio and imaging resolution of the wavy line seismic data, it falls within the protection scope of this invention.

[0068] Step 5: Using the target equivalent Q value and high cutoff frequency F3 determined by each analysis window as discrete nodes, an interpolation algorithm (such as cumulative chord length interpolation algorithm) that fits the CMP strip curved path is used to construct a continuously distributed target equivalent Q value field and target high cutoff frequency field throughout the entire work area.

[0069] In one implementation, the target equivalent Q value and high cutoff frequency F3 determined by each analysis window are used as discrete parameter nodes. An interpolation method that fits the bend width line CMP strip bending path is adopted to perform time direction interpolation, lateral interpolation and spatial smoothing in sequence. Finally, a target equivalent Q value field and target high cutoff frequency field continuously distributed throughout the entire work area are formed, which provides continuous parameter support for the viscoelastic compensation processing of bend width line reflection seismic data.

[0070] The specific process is as follows: (1) Perform time linear interpolation on the target equivalent Q value and high cutoff frequency corresponding to the analysis window set at the common center point to obtain the equivalent Q value sequence and high cutoff frequency sequence continuously distributed along the time depth direction of the common center point.

[0071] Interpolation of the target equivalent Q value and high cutoff frequency F3 at the CMP point of the analysis window in the time direction is performed. The target equivalent Q value and high cutoff frequency F3 determined at the corresponding CMP point of each analysis window are used as discrete parameter nodes in the time depth dimension. For each CMP point with an analysis window, the target equivalent Q value and high cutoff frequency F3 are interpolated along the time depth direction of the seismic data to fill the parameter gaps between discrete time nodes, forming a sequence of equivalent Q value and high cutoff frequency F3 continuously distributed along the time depth at the CMP point of each analysis window, thus constructing a continuous basic parameter set in the time domain.

[0072] (2) Using the cumulative chord length difference algorithm, the equivalent Q value sequence and high cutoff frequency sequence corresponding to the common center point are interpolated laterally by the bending width line to obtain the initial equivalent Q value field and initial high cutoff frequency field of the common center point along the band bending distribution characteristics.

[0073] The equivalent Q-value sequence and the high-cutoff-frequency F3 sequence are based on transverse interpolation and smoothing of the cumulative chord length. The cumulative chord length interpolation method used includes the following core formulas (taking the equivalent Q-value interpolation as an example, the high-cutoff-frequency F3 interpolation process and formulas are completely consistent): Cumulative chord length interpolation parameter calculation: The chord length between adjacent CMP points is (in (where is the coordinate of the i-th CMP point), the cumulative chord length of the i-th CMP point relative to the starting point is . The cumulative chord length is normalized to the [0,1] interval to obtain the cumulative chord length interpolation parameters. , where N is the total number of CMPs in two adjacent equivalent Q-picking windows, and i is the local index within the interval.

[0074] Linear interpolation: Calculates the equivalent Q value at the i-th CMP point based on normalized parameters. (in and These are the target equivalent Q values ​​picked from two adjacent equivalent Q analysis windows, respectively.

[0075] (3) The initial equivalent Q value field and the initial high cutoff frequency field are smoothed to obtain the viscoelastic compensation parameter field corresponding to the bending width line work area.

[0076] In this embodiment of the invention, the planar coordinates of all CMP points along the bend width line are first extracted. The chord length, cumulative chord length, and normalized parameters are then calculated using the aforementioned formula. Next, the equivalent Q-value sequence and high-cutoff frequency F3 sequence after time-direction interpolation are used as lateral discrete nodes. Linear interpolation along the bend width line is then performed to achieve full coverage of all CMP points. Finally, the interpolated equivalent Q-value and high-cutoff frequency F3 are spatially smoothed across the entire work area to eliminate local parameter abrupt changes, ultimately forming an equivalent Q-field and high-cutoff frequency field that are continuous along both the time depth and the lateral CMP direction. The core advantage of the cumulative chord length interpolation method is that by accumulating the actual chord length along the bend path, it accurately reflects the geometric shape of the bend width line, making the interpolation results more consistent with the propagation characteristics of seismic waves in the bend observation system. Compared with traditional interpolation methods based on straight-line distance or CMP index, the method provided in this embodiment of the invention is particularly suitable for processing seismic data of bend width lines in complex surface work areas such as the Loess Plateau.

[0077] It should be noted that, under the technical premise of conforming to the CMP strip bending path, the interpolation method based on the cumulative chord length can be replaced by other methods that can perform parameter interpolation along the bending path, as long as the method still aims to avoid the equivalent quality factor Q field space distortion caused by linear interpolation, it falls within the protection scope of this invention.

[0078] This invention further provides an application example of a viscoelastic compensation parameter field modeling method for flexural width line reflection seismic data. Taking two-dimensional flexural width line data acquired from a loess plateau as an example, the application effect of this method is illustrated. This data, after undergoing a migration preprocessing workflow for two-dimensional flexural width line pre-stack seismic data, has a total data volume of approximately 1.1 GB, a maximum migration distance of 3903 meters, a data sampling interval of 2 milliseconds, and 2501 data sampling points. Figure 3 This is a schematic diagram of the CMP point distribution for the curve width data. It can be seen that the CMP points are distributed along a curved path. Figure 4 This is a schematic diagram of a conventional pre-stack time-migrated local imaging profile provided in an embodiment of the present invention, based on... Figure 4 The conventional pre-stack time-migrating imaging profile shown and Figure 3 The diagram shows the CMP point distribution of the curve width data. The equivalent Q picking window is selected, and the values ​​of equivalent Q and high cutoff frequency F3 are picked. Figure 5 The diagram shown illustrates an equivalent Q selection window, equivalent Q value, and high-cutoff frequency F3 pickup result provided by an embodiment of the present invention. Figure 5 The picked equivalent Q-value and high-cutoff F3 value are used to interpolate the equivalent Q-field and high-cutoff field based on the cumulative chord length method. Figure 6 This is a schematic diagram of the equivalent Q-value field of a bend width line provided in an embodiment of the present invention. Figure 7The image shown is a schematic diagram of a two-dimensional bend width line viscoelastic pre-stack time-migrating local imaging profile based on an equivalent Q value, provided by an embodiment of the present invention. Figure 8 The comparison is shown below. Figure 4 The conventional pre-stack time offset and Figure 7 The diagram shows a comparison of the local imaging profile spectrum of the viscoelastic pre-stack time migration. It can be seen that the spectrum of the viscoelastic pre-stack time migration based on the equivalent Q value is improved by about 12 Hz compared with the conventional pre-stack time migration. Figure 9 Provided for embodiments of the present invention Figure 4 The conventional pre-stack time offset and Figure 7 The magnified comparison of the local imaging profile details of the viscoelastic pre-stack time migration shown can be seen that after viscoelastic compensation, the details of the in-phase axis become richer and the resolution is significantly improved.

[0079] In summary, the core improvement of this invention lies in its global adaptation optimization of the traditional straight-line equivalent Q-value field modeling method, taking into account the unique characteristics of curved-width line reflection seismic data (band-shaped curved distribution of CMP points, complex noise levels, and special structural adaptation requirements). This solves the problems of inaccurate Q-value picking and Q-field modeling distortion caused by the default straight-line topology in existing technologies. The specific core differences are as follows: (1) Adaptation and optimization of noise and Q value scanning of curved width line: In view of the characteristics of complex noise distribution and low signal-to-noise ratio of different offset distance groups of curved width line, the noise processing mode of the traditional straight measurement line is abandoned. A noise identification and elimination scheme adapted to the data characteristics of curved width line is designed. At the same time, the range and density setting of Q value scanning are optimized to ensure that the accuracy and efficiency of Q value scanning are adapted to the large-scale data processing needs of curved width line.

[0080] (2) Parameter selection design for CMP distribution of bend width line: break through the traditional linear CMP point layout and window design logic, combine the bend width line CMP strip bending distribution and structural trend, optimize the layout principle and parameter determination method of analysis window, avoid invalid signal interference caused by window crossing structure / fault, and improve the accuracy of Q value and high cutoff frequency selection.

[0081] (3) Construction of parameter field interpolation for topology adaptation of the bend width line: abandoning the traditional straight-line topology interpolation method, we innovatively adopt the interpolation logic that fits the bending shape of the center line of the bend width line CMP, so as to realize the continuous construction of the equivalent Q field and the high cutoff frequency field, ensuring that the parameter field distribution is consistent with the absorption characteristics of the underground strata and the structural trend of the bend width line, and avoiding the distortion of Q field characterization caused by straight-line interpolation.

[0082] Based on this, the embodiments of the present invention have at least the following characteristics: (1) The modeling method of equivalent quality factor Q value field of the bend width line covers the complete processing flow: from the migration preprocessing of pre-stack reflection seismic data of the bend width line, high-frequency bad channel removal and equivalent quality factor Q value scanning parameter setting, to the generation of inverse Q migration gather, the selection of analysis window along the bend width line CMP strip bending path and the coupling determination of equivalent quality factor Q and high cutoff frequency F3, and then to the interpolation based on the bend width line CMP strip bending path to construct the continuous equivalent quality factor Q field and high cutoff frequency F3 field of the whole work area, forming a closed-loop modeling process. While retaining the advantages of equivalent quality factor Q modeling without well data, stable picking and efficient calculation, it achieves accurate adaptation to the observation characteristics of the bend width line. (2) This method includes three key modules: a signal preprocessing and Q-value scanning parameter setting module adapted to the noise characteristics of the bend width line; a module for selecting the analysis window along the bend width line CMP strip bending path and coupling to determine the equivalent quality factor Q and the high cutoff frequency F3; and a module for constructing the equivalent quality factor Q field and the high cutoff frequency F3 field based on the bend width line CMP strip bending path interpolation. The three modules together constitute the equivalent quality factor Q field modeling technology system for bend width lines. (3) This method combines the technical advantages of modeling the equivalent quality factor Q of the straight test line (no need for downhole data, stable Q value picking, and high modeling efficiency) with the common center point band bending distribution characteristics of the reflection data of the bend width line, forming an equivalent quality factor Q value field modeling scheme suitable for complex surface bend width line work areas, which can support viscoelastic pre-stack time migration processing.

[0083] Based on the foregoing embodiments, this invention provides a viscoelastic compensation parameter field modeling device for flexural width line reflection seismic data, see [link to relevant documentation]. Figure 10 The diagram shows a structural schematic of a viscoelastic compensation parameter field modeling device for curved width line reflection seismic data. The device mainly includes the following parts: The gather construction module 1002 is used to determine the inverse Q-value migration gather based on the seismic data corresponding to the curve width line work area and the preset Q value. The inverse Q-value migration gather includes the inverse Q-value migration profile corresponding to the preset Q value. The Q-value and high cutoff frequency determination module 1004 is used to determine the target equivalent Q-value and high cutoff frequency corresponding to the analysis window based on the preset analysis window on the inverse Q-value offset profile, with the acceptable noise level threshold of the bend width line work area as a constraint. The analysis window is determined based on the strip-shaped bend distribution characteristics of the common center point. The parameter field construction module 1006 is used to perform interpolation processing on the target equivalent Q value and high cutoff frequency based on the characteristics of the strip-shaped bending distribution, so as to construct the viscoelastic compensation parameter field corresponding to the bending width line work area. The viscoelastic compensation parameter field includes the equivalent Q value field and the high cutoff frequency field.

[0084] The viscoelastic compensation parameter field modeling device for curved-width line reflection seismic data provided in this invention determines the analysis window based on the characteristics of the zonal curved distribution, determines the target equivalent Q value and high cutoff frequency by coupling and constraining the noise level, and constructs the target equivalent Q value field and target high cutoff frequency field by using an interpolation method that fits the characteristics of the zonal curved distribution. This solves the core problem of existing equivalent Q value modeling methods based on surface reflection data, which are adapted to straight survey lines and cannot fit the zonal distribution and structural characteristics of curved-width lines (CMP), resulting in inaccurate Q value picking and distorted Q field modeling. This invention retains the core advantages of equivalent Q value modeling and achieves accurate modeling through targeted optimization of curved-width lines.

[0085] In one implementation, the Daogathering construction module 1002 is specifically used for: Identify high-frequency bad paths in the seismic data corresponding to the bend width line work area, so as to remove high-frequency bad paths from the seismic data; Based on the preset Q value, the effective seismic traces after removing high-frequency bad traces in the seismic data are subjected to inverse Q-value filtering and pre-stack time migration processing to obtain the inverse Q-value migration trace set.

[0086] In one implementation, the Daogathering construction module 1002 is specifically used for: Fourier transform processing is performed on all original seismic traces contained in the seismic data corresponding to the bend width line work area to obtain the dominant frequency of the seismic data. Based on the dominant frequency, the highest frequency to be compensated, and the lowest frequency of the seismic data, the high frequency band and the low frequency band are divided. For any original seismic trace, determine the ratio between the total energy of the original seismic trace in the high-frequency band and the total energy in the low-frequency band to obtain the energy ratio corresponding to the original seismic trace; The bad track identification threshold is determined based on the energy ratio corresponding to all original seismic traces. For any original seismic trace, if the energy ratio corresponding to the original seismic trace is higher than the bad trace identification threshold, the original seismic trace is determined to be a high-frequency bad trace.

[0087] In one implementation, the Daogathering construction module 1002 is specifically used for: The energy ratios corresponding to all original seismic traces are compiled into an energy ratio sequence to determine the median of the energy ratio sequence; Determine the absolute deviation between the energy ratios contained in the energy ratio sequence and the median of the energy ratio sequence, and take the median of the absolute deviation sequence composed of all absolute deviations as the median absolute deviation; The bad sector identification threshold is determined based on the median and median absolute deviation of the energy ratio sequence.

[0088] In one implementation, the Daogathering construction module 1002 is specifically used for: Perform the following operation on any valid seismic trace after removing high-frequency bad traces from the seismic data: Based on the preset Q value, the effective seismic trace is subjected to inverse Q-value filtering to obtain the inverse Q-value filtered data corresponding to the effective seismic trace. For all inverse Q-value filtered data of the effective seismic trace, pre-stack time migration processing is performed in batches to obtain the imaging results corresponding to the effective seismic trace. Based on the preset Q value, the imaging results corresponding to all valid seismic traces are classified into the corresponding data buffers. The imaging results stored in the data buffers with the preset Q value are sorted, repositioned and processed into gathers to obtain inverse Q migration gathers.

[0089] In one implementation, the Q-value and high cutoff frequency determination module 1004 is specifically used for: Perform the following operation on any preset analysis window on the inverse Q-value offset profile: Based on the in-phase axis tilt angle within the analysis window, a local fitting algorithm is used to perform signal-to-noise separation on the inverse Q-value offset profile, and the noise level distribution of the inverse Q-value offset profile within the analysis window is obtained. If the noise level distribution of the inverse Q-value offset profile within the analysis window is less than or equal to the acceptable noise level threshold of the bend width line work area, the preset Q value corresponding to the inverse Q-value offset profile is used as the candidate equivalent Q value. With the goal of maximizing the effective spectral width of the inverse Q-value offset profile, the target equivalent Q-value corresponding to the analysis window is determined from the candidate equivalent Q-values, and the high cutoff frequency is determined based on the frequency points in the inverse Q-value offset profile corresponding to the target equivalent Q-value.

[0090] In one implementation, the parameter field construction module 1006 is specifically used for: Perform the following operation on any common center point where analysis windows are set up: By performing time-linear interpolation on the target equivalent Q value and high cutoff frequency corresponding to the analysis window set at the common center point, the equivalent Q value sequence and high cutoff frequency sequence continuously distributed along the time depth direction of the common center point are obtained; The cumulative chord length difference algorithm is used to perform lateral interpolation of the equivalent Q value sequence and high cutoff frequency sequence corresponding to the common center point, so as to obtain the initial equivalent Q value field and initial high cutoff frequency field of the common center point along the strip-shaped bending distribution characteristics. The initial equivalent Q-value field and the initial high-frequency cutoff field are smoothed to obtain the viscoelastic compensation parameter field corresponding to the bending width line work area.

[0091] The device provided in this embodiment of the invention has the same implementation principle and technical effect as the aforementioned method embodiment. For the sake of brevity, any parts not mentioned in the device embodiment can be referred to the corresponding content in the aforementioned method embodiment.

[0092] This invention provides an electronic device, specifically, the electronic device includes a processor and a memory; the memory stores a computer program, which, when run by the processor, executes the method described in any of the above embodiments.

[0093] Figure 11 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. The electronic device 100 includes: a processor 110, a memory 111, a bus 112, and a communication interface 113. The processor 110, the communication interface 113, and the memory 111 are connected through the bus 112. The processor 110 is used to execute executable modules, such as computer programs, stored in the memory 111.

[0094] The memory 111 may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 113 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc.

[0095] Bus 112 can be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 11 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0096] The memory 111 is used to store programs. After receiving an execution instruction, the processor 110 executes the program. The method executed by the device for defining the flow process disclosed in any of the foregoing embodiments of the present invention can be applied to the processor 110 or implemented by the processor 110.

[0097] Processor 110 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of processor 110 or by instructions in software form. Processor 110 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this invention can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software module can reside in a mature storage medium in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory 111, and processor 110 reads the information in memory 111 and, in conjunction with its hardware, completes the steps of the above method.

[0098] The computer program product of the readable storage medium provided in the embodiments of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods described in the foregoing method embodiments. For specific implementation, please refer to the foregoing method embodiments, which will not be repeated here.

[0099] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0100] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A viscoelastic compensation parameter field modeling method for seismic data with curved width lines of reflection, characterized in that, include: Based on the seismic data corresponding to the bend width line work area and the preset Q value, the inverse Q value migration gather is determined. The inverse Q value migration gather includes the inverse Q value migration profile corresponding to the preset Q value. Based on the preset analysis window on the inverse Q-value offset profile, and constrained by the acceptable noise level threshold of the bend width line work area, the target equivalent Q-value and high cutoff frequency corresponding to the analysis window are coupled and determined. The analysis window is determined based on the strip-shaped bending distribution characteristics of the common center point. The strip-shaped bending distribution characteristics refer to the composite spatial distribution pattern in which the common center point of the bend width line presents a horizontally spreading strip structure in the planar space, and also presents a continuous longitudinal bending along the surface obstacle avoidance path. The target equivalent Q-value and the high cutoff frequency are interpolated based on the strip-shaped bending distribution characteristics to construct the viscoelastic compensation parameter field corresponding to the bending width line work area. The viscoelastic compensation parameter field includes a target equivalent Q-value field and a target high cutoff frequency field, including: performing the following operations on any common center point with an analysis window: performing time-linear interpolation on the target equivalent Q-value and the high cutoff frequency corresponding to the analysis window at the common center point to obtain an equivalent Q-value sequence and a high cutoff frequency sequence continuously distributed along the time depth direction at the common center point; using a cumulative chord length interpolation algorithm, performing bending width line lateral interpolation on the equivalent Q-value sequence and the high cutoff frequency sequence corresponding to the common center point to obtain an initial equivalent Q-value field and an initial high cutoff frequency field along the strip-shaped bending distribution characteristics at the common center point; and smoothing the initial equivalent Q-value field and the initial high cutoff frequency field to obtain the viscoelastic compensation parameter field corresponding to the bending width line work area.

2. The viscoelastic compensation parameter field modeling method for reflection seismic data with curved width lines according to claim 1, characterized in that, Based on the seismic data corresponding to the curve width line work area and the preset Q value, determine the inverse Q value migration gather: Identify high-frequency bad paths in the seismic data corresponding to the curved-width line work area, and remove the high-frequency bad paths from the seismic data; Based on a preset Q value, the effective seismic traces in the seismic data after removing the high-frequency bad traces are subjected to inverse Q-value filtering and pre-stack time migration processing to obtain an inverse Q-value migration trace set.

3. The viscoelastic compensation parameter field modeling method for reflection seismic data with curved width lines according to claim 2, characterized in that, Identify high-frequency bad paths in the seismic data corresponding to the bend width line work area, including: Fourier transform processing is performed on all the original seismic traces contained in the seismic data corresponding to the bend width line work area to obtain the dominant frequency corresponding to the seismic data. Based on the dominant frequency, the highest frequency to be compensated and the lowest frequency of the seismic data, the high frequency band and the low frequency band are divided. For any of the original seismic traces, determine the ratio between the total energy of the original seismic trace in the high-frequency band and the total energy in the low-frequency band to obtain the energy ratio corresponding to the original seismic trace; The bad track identification threshold is determined based on the energy ratio corresponding to all the original seismic traces. For any of the original seismic traces, if the energy ratio corresponding to the original seismic trace is higher than the bad trace identification threshold, the original seismic trace is determined to be a high-frequency bad trace.

4. The viscoelastic compensation parameter field modeling method for reflection seismic data with curved width lines according to claim 3, characterized in that, Determine the bad track identification threshold based on the energy ratio corresponding to all the original seismic traces, including: The energy ratios corresponding to all the original seismic traces are combined into an energy ratio sequence to determine the median of the energy ratio sequence; Determine the absolute deviation between the energy ratio contained in the energy ratio sequence and the median of the energy ratio sequence, and take the median of the absolute deviation sequence composed of all the absolute deviations as the median absolute deviation; The bad sector identification threshold is determined based on the median and the absolute deviation of the median in the energy ratio sequence.

5. The viscoelastic compensation parameter field modeling method for reflection seismic data with curved width lines according to claim 2, characterized in that, Based on a preset Q-value, the effective seismic traces in the seismic data after removing the high-frequency bad traces are subjected to inverse Q-value filtering and pre-stack time migration processing to obtain an inverse Q-value migration gather, including: For any valid seismic trace in the seismic data after removing the high-frequency bad traces, perform the following operation: Based on the preset Q value, the effective seismic trace is subjected to inverse Q-value filtering to obtain the inverse Q-value filtered data corresponding to the effective seismic trace. For all the inverse Q-value filtered data of the effective seismic trace, pre-stack time migration processing is performed in batches to obtain the imaging results corresponding to the effective seismic trace; Based on the preset Q value, the imaging results corresponding to all the valid seismic traces are classified into the corresponding data buffers. The imaging results stored in the data buffers of the preset Q value are sorted, repositioned and processed into gathers to obtain inverse Q-migrated gathers.

6. The viscoelastic compensation parameter field modeling method for reflection seismic data with curved width lines according to claim 1, characterized in that, Based on the preset analysis window on the inverse Q-value offset profile, and constrained by the acceptable noise level threshold of the bend width line work area, the target equivalent Q-value and high cutoff frequency corresponding to the analysis window are coupled and determined, including: Perform the following operation on any of the preset analysis windows on the inverse Q-value offset profile: Based on the in-phase axis tilt angle within the analysis window, a local fitting algorithm is used to perform signal-to-noise separation on the inverse Q-value offset profile to obtain the noise level distribution of the inverse Q-value offset profile within the analysis window. If the noise level distribution of the inverse Q-value offset profile within the analysis window is less than or equal to the acceptable noise level threshold of the bend width line work area, the preset Q value corresponding to the inverse Q-value offset profile is taken as the candidate equivalent Q value. With the goal of maximizing the effective spectral width of the inverse Q-value offset profile, the target equivalent Q-value corresponding to the analysis window is determined from the candidate equivalent Q-values, and the high cutoff frequency is determined based on the frequency points in the inverse Q-value offset profile corresponding to the target equivalent Q-value.

7. A viscoelastic compensation parameter field modeling device for seismic data with curved width line reflection, characterized in that, include: The gather construction module is used to determine the inverse Q-value migration gather based on the seismic data corresponding to the curve width line work area and the preset Q value. The inverse Q-value migration gather includes the inverse Q-value migration profile corresponding to the preset Q value. The Q-value and high cutoff frequency determination module is used to determine the target equivalent Q-value and high cutoff frequency corresponding to the analysis window based on the preset analysis window on the inverse Q-value offset profile, with the acceptable noise level threshold of the bend width line work area as a constraint. The analysis window is determined based on the strip-shaped bending distribution characteristics of the common center point. The strip-shaped bending distribution characteristics refer to the composite spatial distribution pattern in which the common center point of the bend width line presents a horizontally spreading strip structure in the planar space, and also presents a continuous longitudinal bending along the surface obstacle avoidance path. The parameter field construction module is used to perform interpolation processing on the target equivalent Q value and the high cutoff frequency based on the strip-shaped bending distribution characteristics to construct the viscoelastic compensation parameter field corresponding to the bending width line work area. The viscoelastic compensation parameter field includes an equivalent Q value field and a high cutoff frequency field, including: performing the following operations on any common center point with an analysis window: performing time linear interpolation on the target equivalent Q value and the high cutoff frequency corresponding to the analysis window at the common center point to obtain an equivalent Q value sequence and a high cutoff frequency sequence continuously distributed along the time depth direction at the common center point; using a cumulative chord length interpolation algorithm, performing bending width line lateral interpolation on the equivalent Q value sequence and the high cutoff frequency sequence corresponding to the common center point to obtain an initial equivalent Q value field and an initial high cutoff frequency field along the strip-shaped bending distribution characteristics at the common center point; and smoothing the initial equivalent Q value field and the initial high cutoff frequency field to obtain the viscoelastic compensation parameter field corresponding to the bending width line work area.

8. An electronic device, characterized in that, The method includes a processor and a memory, the memory storing computer-executable instructions executable by the processor, the processor executing the computer-executable instructions to implement the method of any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions that, when invoked and executed by a processor, cause the processor to perform the method according to any one of claims 1 to 6.

Images