Predictive data volume and crack identification method based on weak energy of seismic data

Abstract

This invention provides a prediction data volume based on weak energy seismic data, along with a method for picking and identifying fractures, belonging to the field of fracture prediction technology in seismic interpretation for oil and gas field exploration and development. The picking method includes: Step 1, preparing post-stack seismic data and determining the data range to be calculated; Step 2, determining the wavelength n and the calculation window size; Step 3, for any seismic sample point p0 to be calculated, executing sub-steps 3a-3c; Step 4, traversing all sample points to be calculated, repeating Step 3 to obtain the final weak energy picked fracture prediction data volume. This invention can consider tectonic variation factors, ensuring that weak energy attributes reflect differences within the same stratum; it can mark fractures by finding weak energy locations, to some extent compensating for the deficiency of conventional seismic fracture attributes that only consider differences between adjacent traces; it can consider stratigraphic anisotropy, adding geometric and energy difference scaling factors, making the reflection of fault fractures more reliable.

Description

Technical Field

[0001] This invention relates to the field of fracture prediction technology in seismic interpretation of oil and gas field exploration and development. Specifically, it relates to a fracture prediction data volume that can utilize the weak energy of seismic data and its picking method, as well as a fracture identification method using the aforementioned fracture prediction data volume. Background Technology

[0002] In oil and gas exploration, fracture prediction is a necessary step. Its purpose is to identify fracture development zones (sweet spots) both vertically and horizontally, or to avoid high-risk faults and fractures, thereby mitigating engineering risks. In recent years, horizontal well operations in shale gas exploration and development in the Weiyuan area of ​​Sichuan Province have experienced severe casing deformation (referred to as "casing deformation"). In the past two years, approximately 50% of wells have experienced casing deformation, resulting in significant economic losses. The causes of casing deformation are engineering factors, the root cause being the shear stress on the casing, while the direct causes are shear slippage caused by geological factors such as fractures, micro-folds, and abrupt lithological changes, with the presence of fractures being the primary cause. Therefore, accurate and effective prediction of fracture locations is crucial for engineering early warning. Regional fracture prediction mainly relies on seismic data. This invention is an algorithm for extracting and evaluating fault fracture development from seismic data, representing an improvement over existing similar technologies.

[0003] There are many methods for predicting fractures in seismic data, the most important of which is seismic attribute extraction technology, which marks the location of fault fractures by detecting geometric or energy differences. These methods include amplitude coherence, variance, chaos, curvature, fault likelihood, and ant tracking, among others. These methods and commercial software are the most popular and effective, with high acceptance, and can reflect the condition of fractures to varying degrees.

[0004] Existing methods also have drawbacks. For example, the techniques used in current commercial seismic interpretation software are generally based on methods such as amplitude coherence, variance, chaos, and curvature. They typically represent the displacement of in-phase axes by detecting the similarity between adjacent traces, or represent the changes in seismic profiles by detecting differences in energy or geometric information between adjacent traces, thereby highlighting the location of fault fractures. This constraint is generally not detected on the same stratum, but only calculated in the horizontal direction; furthermore, there are no attribute techniques based on the weak energy response at the fracture site. Another example is the ant tracking method, a fracture attribute enhancement algorithm widely used in the industry and achieving good results in many blocks, but it is greatly affected by the input data and calculation parameters, making it difficult to guarantee stability and reliability.

[0005] Chinese invention patent application CN111175816A, published on May 19, 2020, discloses a method and apparatus for real-time construction of a microseismic fracture network in reservoir stimulation. The method includes: Step 1, acquiring microseismic events in reservoir stimulation in real time; Step 2, performing a Hough transform and gridding the parameter space, calculating the frequency of microseismic events in each grid cell; Step 3, weighting the microseismic events based on their moment magnitude; Step 4, finding the set of events with the highest frequency in the parameter space; Step 5, performing least-squares plane fitting on the events within the set; and Step 6, drawing a discrete microseismic fracture network based on the obtained fracture propagation plane. This method and apparatus, through the construction of a fracture network from microseismic event points, can accurately depict the distribution of artificial fractures during reservoir stimulation. Based on the microseismic monitoring results, it can evaluate the effectiveness of reservoir stimulation, optimize processes and parameters, and deploy wells, thereby effectively improving oil and gas recovery. However, it does not address the location of weak energy points and their related aspects in fracture characterization.

[0006] Chinese invention patent application CN113655542A, published on November 16, 2021, discloses a geophysical-based method for acquiring reservoir information during the development stage of hot dry rock. Through indoor hot dry rock rock physics experiments, it establishes the correlation mapping relationship between the geophysical field and the hot dry rock reservoir, and the variation law of the relationship with reservoir parameters. Reservoir parameters are obtained through microseismic moment tensor inversion of hot dry rock hydraulic fracturing, thereby constructing a continuous fracture network model containing energy and morphological parameters. Information about hot dry rock reservoirs entering the development stage is then obtained from this continuous fracture network model. This invention patent application combines laboratory rock physics experiments with in-situ hot dry rock hydraulic fracturing, providing a basis for acquiring parameters of hot dry rock reservoirs entering the development stage and filling a gap in the prior art. However, it also does not address the location of weak energy and its related content on fracture characterization. Summary of the Invention

[0007] The purpose of this invention is to address at least one of the aforementioned shortcomings of the prior art. For example, one objective of this invention is to address the problems in conventional techniques that do not consider the determination of fracture properties within the same formation and do not consider the weak energy response at the fracture site, and to propose a method for characterizing fractures by detecting the location of weak energy within the same formation.

[0008] In general, the basic idea of ​​this invention—the prediction data volume based on weak energy seismic data, its picking method, and the fracture identification method—includes the following: At the fracture scale, due to the fractured rock at the fracture location, the reflected energy absorption is strong, generally not forming strong reflections, and the reflections are weaker than those at other locations in the same stratum. Forward modeling experiments have also verified this. Therefore, this invention searches for and marks such relatively weak energy locations by traversing seismic data samples point by point, forming a weak energy fracture prediction data volume (hereinafter referred to as: weak energy picking volume), thereby obtaining fault and fracture information that is difficult to capture using conventional fracture prediction techniques.

[0009] The improvements of this invention may include: (1) General seismic attribute techniques do not determine the spatial morphology of the stratum to which each point belongs before obtaining the differences, while this invention improves upon this point, so that the attribute differences obtained subsequently belong to the same stratum; (2) The definition of weak energy in this technical solution is: if the energy value of the seismic waveform at a certain point in the seismic data is the minimum value within a certain range of the stratum where that point is located, then the energy at that point is weak energy. Marking fractures by finding the location of weak energy makes up for the deficiency of conventional seismic fracture attributes that only consider the differences between adjacent traces to a certain extent; (3) The weak energy marking value obtained by this invention is a comprehensive response in multiple directions, taking into account the anisotropy of the strata, and adding the assistance of geometric differences and energy differences, making the reflection of fault fractures more reliable.

[0010] To address the issues of existing technologies failing to consider fracture attribute determination within the same stratum and lacking a basis for weak energy responses at fracture sites, one aspect of the present invention provides a method for extracting fracture prediction data volumes based on weak energy responses in seismic data. The method for extracting fracture prediction data volumes includes:

[0011] Step 1: Prepare post-stack seismic data (e.g., post-stack migrated seismic data) and determine the range of data to be calculated;

[0012] Step 2: Determine the wavelength n and the size of the calculation window, which includes the longitudinal time window and the lateral range;

[0013] Step 3: For any seismic sample point p0 that needs to be calculated, execute sub-steps 3a to 3c, where...

[0014] Sub-step 3a: Based on the above calculation parameters, determine the cubic data volume c of the calculation window centered on p0. Based on the similarity of the seismic waveforms, determine the spatial morphology and energy information of the stratum S to which p0 belongs in the data volume c. That is to say, in sub-step 3a, the similarity of seismic waveforms of adjacent traces represents the consistent strength of the strata. Therefore, the center positions of the two waveforms are regarded as the location of the same stratum. Thus, waveform similarity is used to determine the spatial morphology of the stratum S.

[0015] Sub-step 3b: Determine whether the energy of the waveform at point p0 is a minimum point in a certain direction on the stratum S to which it belongs. If it is a minimum point, increase the possibility that the point is a weak energy point.

[0016] Step 4: Iterate through all the sample points that need to be calculated, repeat step 3, and obtain the final weak energy picking crack prediction data volume.

[0017] In an exemplary embodiment of the present invention, in step 2, the wavelength may refer to the length or number of sample points of the seismic wave used for comparison of similarity when determining the spatial morphology of the stratum S in sub-step 3a; the longitudinal time window may be the longitudinal search range for seismic waveform similarity comparison; and the lateral range may refer to the number of seismic traces around the center point p0 used.

[0018] In an exemplary embodiment of the present invention, in sub-step 3a, when determining the spatial morphology of stratum S in data volume c, it is necessary to compare the similarity of the waveforms at stratum S in the current trace and the previous trace, rather than comparing the similarity of the waveforms at stratum S in the current trace and the center trace. For example, sub-step 3a may specifically be as follows: starting from the center point p0, expanding outwards, comparing one by one all seismic waveforms {w1, w2, ..., w...} of wavelength n within the longitudinal time window of the current trace. m The similarity between this waveform and the previous seismic waveform w0 with wavelength n at location S in the stratum is recorded; the waveform w with the highest similarity is recorded. k ∈{w1,w2,…,w m The spatial position p of the midpoint of} k And the energy E of this waveform k ; Traverse all seismic traces in the calculation window to obtain p for each trace k The formation S is formed by points. The energy of the waveform can be equal to the average of the sum of the squares of the amplitudes at each point. Furthermore, to compare the similarity of waveforms, custom similarity rules can be defined, such as comparing correlation coefficients or vector angles.

[0019] In an exemplary embodiment of the present invention, in sub-step 3b, when determining whether the energy of the waveform at point p0 is a minimum point in its respective stratum S, the specific method may be as follows: set an initial value f for point p0, and determine the energy in multiple directions centered on that point. If the point is a minimum point in a certain direction, then amplify f; if it is not a minimum point, then ignore it; after determining the energy in all directions, obtain the final value f', and then replace the value of f at point p0 with f'. For example, when amplifying f, multiplying by a first coefficient, the first coefficient is in the range of greater than 1 and less than 2, and can further be 1.05 to 1.25.

[0020] To address the issues of existing technologies failing to consider fracture attributes within the same stratum and lacking a basis for weak energy responses at fracture sites, another aspect of the present invention provides a method for extracting fracture prediction data volumes based on weak energy responses in seismic data. The method for extracting fracture prediction data volumes includes:

[0021] Step 1': Prepare post-stack seismic data (e.g., post-stack migrated seismic data) and determine the range of data to be calculated;

[0022] Step 2': Determine the wavelength n and the size of the calculation window, which includes the longitudinal time window and the lateral range;

[0023] Step 3′: For any seismic sample point p0 that needs to be calculated, execute sub-steps 3a′~3c′, where,

[0024] In sub-step 3a', based on the above calculation parameters, a cubic data volume c of the calculation window centered on p0 is determined. Based on the similarity of the seismic waveforms, the spatial morphology and energy information of the stratum S to which point p0 belongs in the data volume c are determined. That is to say, in sub-step 3a, the similarity of seismic waveforms of adjacent traces represents the strong consistency of the strata. Therefore, the center positions of the two waveforms are regarded as the location of the same stratum. Thus, waveform similarity is used to determine the spatial morphology of stratum S.

[0025] Sub-step 3b': Determine whether the energy of the waveform at point p0 is a minimum point in a certain direction on the stratum S to which it belongs. If it is a minimum point, increase the probability that the point is a weak energy point.

[0026] Sub-step 3c′ can be used to correct the confidence of point p0 as a fault or fracture by other geometric difference factors or energy difference scaling factors.

[0027] Step 4′: Traverse all the sample points that need to be calculated, and repeat step 3′ to obtain the final weak energy picking crack prediction data volume. That is, repeat the weak energy point discrimination and assignment operations in steps (3a′) to (3c′) for all the sample points that need to be calculated to obtain the entire weak energy picking crack prediction data volume.

[0028] Here, according to another aspect of the method of the present invention, a sub-step 3c' is additionally provided compared to the method described above, while the other steps are essentially the same. The additional sub-step 3c' helps to highlight locations more relevant to the crack while filtering out weak energy locations that may be unrelated to the crack.

[0029] In an exemplary embodiment of the present invention, in sub-step 3b', when determining whether the energy of the waveform at point p0 is a minimum point in its respective stratum S, the specific method may be as follows: set an initial value f for point p0, and determine in multiple directions centered on point p0. If point p0 is a minimum point in a certain direction, then amplify f; if it is not a minimum point in the energy, then ignore it; after determining in all directions, obtain the final value f', and then replace the value f of point p0 with f'.

[0030] In an exemplary embodiment of the present invention, in sub-step 3c′, in the direction where the energy of the waveform at point p0 is at its minimum, any factor representing the energy difference or geometric difference between point p0 and other points in that direction can be used, such as variance, curvature, etc., which can be one or more. Based on the magnitude of this factor, a certain scaling range is limited, and f' is scaled, thereby correcting the fault or fracture confidence at point p0. For example, when scaling f', it can be done by multiplying by a second coefficient, which can be greater than 0 and less than 1, and further can be 5% to 25%. For example, wavelength refers to the length or number of seismic waves used for comparing similarity when determining the spatial morphology of strata S in sub-step 3a′; the longitudinal time window is the longitudinal search range for seismic waveform similarity comparison; the lateral range refers to the number of seismic traces around the center point p0 used.

[0031] In an exemplary embodiment of the present invention, in sub-step 3a′, when determining the spatial morphology of stratum S in data volume c, it is necessary to compare the similarity of the waveforms at stratum S in the current trace and the previous trace, rather than comparing the similarity of the waveforms at stratum S in the current trace and the center trace. For example, sub-step 3a′ can specifically be: starting from the center point p0, expanding outwards, comparing one by one all seismic waveforms {w1, w2, ..., w...} of wavelength n within the longitudinal time window of the current trace. m The similarity between this waveform and the previous seismic waveform w0 with wavelength n at location S in the stratum is recorded; the waveform w with the highest similarity is recorded. k ∈{w1,w2,…,w m The spatial position p of the midpoint of} k And the energy E of this waveform k ; Traverse all seismic traces in the calculation window to obtain p for each trace k The formation S of the points. For example, the energy of a waveform can be equal to the average of the sum of the squares of the amplitudes at each point. For example, to compare the similarity of waveforms, custom similarity rules can be defined, such as comparing correlation coefficients or vector angles.

[0032] To address the issues in existing technologies that do not consider the determination of fracture attributes within the same stratum and lack a basis for weak energy response at fracture locations, another aspect of the present invention provides a fracture identification method based on weak energy picking of seismic data. This fracture identification method uses the final weak energy picked fracture prediction data volume obtained by any of the fracture prediction data volume picking methods described above as the data volume for fracture identification.

[0033] Compared with the prior art, the beneficial effects of the present invention include at least one of the following:

[0034] 1. It can obtain the geometric information of the strata to which a single seismic sample point belongs. The weak energy search and scaling factor calculation in the process are based on this level, taking into account the factors of tectonic changes, so that the weak energy attributes reflect the differences on the same stratum.

[0035] 2. It can mark cracks by finding weak energy locations, which is in line with geological and physical laws and to some extent makes up for the shortcomings of conventional earthquake crack attributes that only consider the differences between adjacent traces.

[0036] 3. The weak energy label value that can be obtained is a comprehensive response in multiple directions, taking into account the anisotropy of the formation, and adding geometric difference and energy difference scaling factors, making the reflection of fault fractures more reliable. Attached Figure Description

[0037] The above and other objects and / or features of the present invention will become clearer from the following description taken in conjunction with the accompanying drawings, in which:

[0038] Figure 1 A flowchart illustrating an exemplary embodiment of the method for extracting fracture prediction data volumes based on weak energy picking of seismic data according to the present invention is shown.

[0039] Figure 2 A schematic diagram of the weak energy picking profile effect is shown in an exemplary embodiment of the method for weak energy picking of fracture prediction data volume based on seismic data according to the present invention.

[0040] Figure 3 A schematic diagram of the weak energy pickup planar effect is shown, illustrating another exemplary embodiment of the method for weak energy pickup of crack prediction data volume based on seismic data according to the present invention. Detailed Implementation

[0041] In the following sections, the present invention will be described in detail with reference to exemplary embodiments, including the prediction data volume based on weak energy of seismic data, the method for picking the data, and the crack identification method.

[0042] It should be noted that "first," "second," etc., are merely for the convenience of description and distinction, and should not be interpreted as indicating or implying relative importance.

[0043] Example 1

[0044] Figure 1 A flowchart illustrating an exemplary embodiment of the method for extracting crack prediction data volumes based on weak energy picking of seismic data according to the present invention is shown. Figure 2 A schematic diagram of the weak energy pickup profile effect is shown in an exemplary embodiment of the method for weak energy pickup of fracture prediction data volume based on seismic data according to the present invention.

[0045] This invention was applied to seismic data from a region in Sichuan Province to predict the fracture development of a specific stratum in the seismic data. This stratum is mainly composed of carbonate rocks and shale, and fractures are well-developed; therefore, weak energy pick-ups can easily reveal the location of fractures.

[0046] In this exemplary embodiment, the method for extracting fracture prediction data volumes based on weak energy picking of seismic data is implemented according to the following steps:

[0047] Step (1), prepare post-stack seismic data, for example, Figure 1 Using high-fidelity, amplitude-preserving post-stack seismic data, the range of data to be calculated is determined.

[0048] Step (2) determines the wavelength n and calculates the window size, including the longitudinal time window and the lateral range. For example, Figure 1 The wavelength is determined, and parameters such as the window size are calculated.

[0049] In step (2), wavelength refers to the length of the seismic wave used for comparison when determining the spatial morphology of stratum S in step (3a), and is set to 30 points. The longitudinal time window refers to the longitudinal search range for seismic waveform similarity comparison, and is set to 37 points. The lateral range refers to the number of seismic traces around the center point p0 used, and is set to 10 traces around point p0.

[0050] Step (3) involves performing the following sub-steps 3a to 3c for each or any seismic sample point p0 that needs to be calculated.

[0051] Sub-step (3a): Based on the above calculation parameters, determine the cubic data volume c of the calculation window centered at p0. Based on the similarity of the seismic waveforms, determine the spatial morphology and energy information of the stratum S to which point p0 belongs within data volume c. For example, Figure 1 The spatial morphology of the stratum S to which the seismic sample point p0 belongs is determined.

[0052] In sub-step (3a), when determining the spatial morphology of strata S in data volume c, starting from the center point p0, the process expands outwards, comparing each of the current trace's seismic waveforms {w1, w2, ..., w...} with a wavelength of 30 nm within the longitudinal time window. mThe similarity between this waveform and the previous seismic waveform w0 with a wavelength of 30 nm at location S in the stratum is recorded; the waveform w with the highest similarity is recorded. k ∈{w1,w2,…,w m The spatial position p of the midpoint of} k And the energy E of this waveform k Traverse all seismic traces in the calculation window to obtain p for each trace. k The stratum S formed by the point.

[0053] In substep (3a), the energy of the waveform is equal to the average of the sum of the squares of the amplitudes at each point.

[0054] In sub-step (3a), the similarity of waveforms is compared by comparing the magnitude of the correlation coefficient.

[0055] Sub-step (3b) determines whether the energy of the waveform at point p0 is a minimum point in a certain direction within the corresponding stratum S. If it is a minimum point, the probability that this point is a weak energy point is increased. For example, Figure 1 In the process, we determine whether p0 is an energy minimum point in multiple directions of S.

[0056] In sub-step (3b), when determining whether the energy of the waveform at point p0 is a minimum point in its respective stratum S, the specific method is as follows: An initial value f = 1 is set for point p0, and the energy is evaluated in four directions centered on this point: the line direction, the trace direction, and two oblique directions. If the point is a minimum point in any direction, f is multiplied by a coefficient of 1.10; otherwise, it is ignored. After evaluating all directions, the final value f' is obtained, and then f' is used to replace the f value at point p0.

[0057] Sub-step (3c) modifies the confidence level of point p0 as a fault or fracture using other geometric difference factors or energy difference scaling factors. For example, Figure 1 The scaling factor is calculated to correct the crack confidence at point p0. It should be noted that this sub-step (3c) in this embodiment is an optional step, which highlights some locations that are more relevant to the crack while filtering out some weak energy locations that may be irrelevant to the crack.

[0058] In sub-step (3c), in the direction where the energy of the waveform at point p0 is at its minimum, any factor representing the energy difference or geometric difference between point p0 and other points in that direction can be used, such as variance, curvature, etc., which can be one or multiple. Then, a certain scaling range is limited to further scale f', thereby correcting the confidence of point p0 as a fault or fracture. Here, two scaling factors are selected: (1) energy variance, which is the variance of the energy of all points in that direction, representing the instability of energy change in that direction; (2) curvature offset, which is the difference between the maximum and minimum absolute values ​​of the curvature of stratum S in that direction, representing the instability of the geometric structure of the stratum in that direction. Based on the overall range of the values ​​of these two factors, the value of f' is scaled by a small ratio, with the scaling range controlled within 10%.

[0059] Step (4): Iterate through all the sample points that need to be calculated, repeating sub-steps (3a) to (3c) to obtain the final weak energy picking crack prediction data volume. For example, Figure 1 All samples have been traversed, as well as the weak energy picker.

[0060] In step (4), the weak energy point discrimination and assignment operations of sub-steps (3a) to (3c) are repeated for all the sample points that need to be calculated to obtain the entire data volume. From Figure 2 The cross-section shows that weak energy pick-up bodies exhibit high and continuous response values ​​(greater than 1.13) at major faults, extending linearly along the fault plane of the seismic phase axis. They also appear (less than 1.13) at locations where there is no significant fault displacement on the seismic phase axis but only a slight weakening of energy (difficult to see with the naked eye), with relatively short extension lengths. These locations are usually strongly correlated with fractures. From a geological perspective, the response of weak energy properties to fault fractures is generally quite accurate and natural.

[0061] Example 2

[0062] Figure 3 A schematic diagram of the weak energy pickup planar effect is shown, illustrating another exemplary embodiment of the method for weak energy pickup of crack prediction data volume based on seismic data according to the present invention.

[0063] This invention was applied to seismic data from a region in Sichuan Province to predict fracture development in the Longmaxi-Wufeng Formation shale strata. While horizontal well development is predominant, casing deformation accidents are frequent, and the presence of fractures is a prerequisite for these accidents. The shale strata in this area are relatively stable in terms of deposition, with few faults exhibiting significant displacement and a high prevalence of fractures, making it difficult to identify anomalies caused by the displacement of seismic axes. Therefore, a fracture prediction method based on weak energy picking from seismic data was used to explore and supplement fracture locations.

[0064] In this exemplary embodiment, the method for extracting fracture prediction data volumes based on weak energy picking of seismic data is implemented according to the following steps:

[0065] Step (1): Prepare post-stack seismic data and determine the range of data to be calculated.

[0066] Step (2): Determine the wavelength n and calculate the window size, including the longitudinal time window and the lateral range.

[0067] In step (2), the wavelength refers to the length of the seismic wave used for comparison when determining the spatial morphology of strata S in sub-step (3a), and is set to 32 points. The longitudinal time window is the longitudinal search range for seismic waveform similarity comparison, and is set to 41 points. The lateral range refers to the number of seismic traces around the center point p0 used, and is set to 10 traces around point p0.

[0068] Step (3) involves performing the following sub-steps 3a to 3c for each seismic sample point p0 that needs to be calculated.

[0069] Sub-step (3a): Based on the above calculation parameters, determine the cubic data volume c of the calculation window centered on p0. Based on the similarity of the seismic waveforms, determine the spatial morphology and energy information of the stratum S to which point p0 belongs in the data volume c.

[0070] In sub-step (3a), when determining the spatial morphology of strata S in data volume c, starting from the center point p0, the process expands outwards, comparing each of the current trace's seismic waveforms {w1, w2, ..., w...} with a wavelength of 30 nm within the longitudinal time window. m The similarity between this waveform and the previous seismic waveform w0 with a wavelength of 30 nm at location S in the stratum is recorded; the waveform w with the highest similarity is recorded. k ∈{w1,w2,…,w m The spatial position p of the midpoint of} k And the energy E of this waveform k Traverse all seismic traces in the calculation window to obtain p for each trace. k The formation S is formed by points. The similarity of the waveforms is compared by comparing the magnitude of the correlation coefficient.

[0071] Sub-step (3b) determines whether the energy of the waveform at point p0 is a minimum point in a certain direction on the stratum S to which it belongs. If it is a minimum point, the probability that the point is a weak energy point is increased.

[0072] In sub-step (3b), when determining whether the energy of the waveform at point p0 is a minimum point in its respective stratum S, the specific method is as follows: An initial value f = 1 is set for point p0, and the energy is evaluated in four directions centered on this point: the line direction, the trace direction, and two oblique directions. If the point is a minimum point in any direction, f is multiplied by a coefficient of 1.3; otherwise, it is ignored. After evaluating all directions, the final value f' is obtained, and then f' replaces the f value at point p0.

[0073] Sub-step (3c) corrects the confidence of point p0 as a fault or fracture by using other geometric difference factors or energy difference scaling factors.

[0074] In sub-step (3c), in the direction where the energy of the waveform at point p0 is at its minimum, any factor representing the energy difference or geometric difference between point p0 and other points in that direction can be used, such as variance, curvature, etc., which can be one or multiple. Then, a certain scaling range is limited, and f' is further scaled to correct the confidence of point p0 as a fault or fracture. Two scaling factors are selected here: (1) energy variance, which is the variance of the energy of all points in that direction, representing the instability of energy change in that direction; (2) curvature offset, which is the difference between the maximum and minimum absolute values ​​of the curvature of stratum S in that direction, representing the instability of the geometric structure of the stratum in that direction. Based on the overall range of the values ​​of these two factors, the value of f' is scaled by a small ratio, with the scaling range controlled within 20%.

[0075] Step (4): Traverse all the sample points that need to be calculated, and repeat sub-steps (3a) to (3c) to obtain the final weak energy picking crack prediction data volume.

[0076] In sub-step (4), repeat the weak energy point discrimination and assignment operations of sub-steps (3a) to (3c) for all sample points that need to be calculated to obtain the entire data volume. Figure 3 Based on the seismic horizon of the bottom boundary of the Longmaxi Formation, a planar map was generated by extracting weak energy picking attributes within a time window of approximately 15ms. This map was then overlaid with the horizontal well trajectory and the actual location of the nesting deformation indicated by the triangular graphic. Figure 3 In this map, with the ruler directly in front as the viewing angle, north is at the top, south at the bottom, west on the left, and east on the right. That is, in this attached map, the right side of the page represents north, and the bottom side represents east. From... Figure 3 As can be seen above, the weak energy picking attribute plan map reflects the cracks well and has a good correspondence with the location of the overlay transformation: (1) On the west side of the work area, the crack direction is mainly NE (i.e., northeast) and NEE (i.e., northeast-east), while on the east side of the work area, cracks in multiple directions are comprehensively developed, which is consistent with the geological research conclusions of this block; (2) At or near the overlay transformation point, continuous lines of weak energy attributes can be seen in most areas, and some low-value continuous responses can also be considered as crack responses. The overlay transformation prediction accuracy rate is nearly 70% in this area.

[0077] In summary, the advantages of this invention over existing technologies include: effectively reflecting fracture information ignored by conventional methods, overcoming the shortcomings of conventional fracture attribute assessments which do not consider differences in the same stratum and do not take into account the relatively weak energy response at the fracture site. The main reasons for this are at least twofold: First, conventional fracture attribute techniques represent the displacement of the same phase axis by detecting the similarity of adjacent traces, or represent changes in the seismic profile by detecting differences in energy or geometric information between adjacent traces, thereby highlighting the location of fault fractures. However, this usually does not consider the constraint of detection in the same stratum, and is only obtained in the horizontal direction. Second, conventional fracture attribute techniques generally consider the similarity of adjacent seismic traces, but do not base them on the weak energy response at the fracture site.

[0078] Although the present invention has been described above in conjunction with exemplary embodiments and accompanying drawings, those skilled in the art should understand that various modifications can be made to the above embodiments without departing from the spirit and scope of the claims.

Claims

1. A method for weak energy picking fracture prediction data volume based on seismic data, characterized in that, The method for picking up the fracture prediction data volume comprises the following steps: Step 1, preparing post-stack seismic data, and determining the data range to be calculated; Step 2, determining the wavelength n and the calculation window size including the longitudinal time window and the lateral range; Step 3, for any seismic sample point p0 to be calculated, performing sub-steps 3a-3b, wherein, Sub-step 3a, according to the wavelength and the window size, determining the calculation window cubic data volume c centered on p0, and according to the similarity of the seismic waveforms, determining the spatial form and energy information of the stratum S to which p0 belongs in the data volume c; Sub-step 3b, judging whether the energy of the p0 point waveform is a minimum point in a certain direction on the stratum S, if it is a minimum point, then increase the possibility of the point being a weak energy point; in the sub-step 3b, when judging whether the energy of the p0 point waveform is a minimum point on the stratum S, the specific method is: setting an initial value for the p0 point f , and judging in multiple directions centered on the point, if the point is an energy minimum point in a certain direction, then amplifying the value of the point f ; if it is not an energy minimum point, then ignoring; after judging in all directions, obtaining the final value f’ , and then replacing the value of the p0 point with the value of the point f’ ; the value of the p0 point f ; Step 4, traversing all the sample points to be calculated, repeating step 3, and obtaining the final weak-energy picked-up fracture prediction data volume.

2. The method for picking up weak energy based on seismic data to predict the data volume of cracks according to claim 1, characterized in that, For f When amplifying, a multiplication by a first coefficient is used, the first coefficient ranging from greater than 1 and less than 2.

3. A method for weak energy picking fracture prediction data volume based on seismic data, characterized in that, The method for picking up the fracture prediction data volume comprises the following steps: Step 1, preparing post-stack seismic data, and determining the data range to be calculated; Step 2, determining the wavelength n and the calculation window size including the longitudinal time window and the lateral range; Step 3, for any seismic sample point p0 to be calculated, performing sub-steps 3a-3c, wherein, Sub-step 3a, according to the wavelength and the window size, determining the calculation window cubic data volume c centered on p0, and according to the similarity of the seismic waveforms, determining the spatial form and energy information of the stratum S to which p0 belongs in the data volume c; Sub-step 3b involves determining whether the energy of the waveform at point p0 is a minimum point in a certain direction within the formation S. If it is a minimum point, the probability that this point is a weak energy point is increased. Specifically, in sub-step 3b, determining whether the energy of the waveform at point p0 is a minimum point within the formation S involves setting an initial value for point p0. f The criteria are determined in multiple directions centered on that point. If the point is an energy minimum point in a certain direction, then... f Magnification is applied; if it is not an energy minimum, it is ignored; after determining the values ​​in all directions, the final value is obtained. f’ Then use f’ Replace point p0 f value; Sub-step 3c, correcting the fault or fracture credibility of p0 through other geometric difference factors or energy difference scaling factors; Step 4, traversing all the sample points to be calculated, repeating step 3, and obtaining the final weak-energy picked-up fracture prediction data volume.

4. The method for picking up weak energy based on seismic data to predict crack data volume according to claim 1 or 3, characterized in that, In step 2, the wavelength refers to the length or sample number of the seismic wave used for comparing the similarity when the spatial form of the stratum S is determined in sub-step 3a; the longitudinal time window is the longitudinal search range for the seismic waveform similarity comparison; and the lateral range refers to the number of seismic traces around the center point p0.

5. The method for picking up weak energy based on seismic data to predict crack data volume according to claim 1 or 3, characterized in that, In sub-step 3a, when the spatial form of the stratum S in the data volume c is determined, the similarity of the waveforms at the stratum S of the current trace and the previous trace needs to be compared.

6. The method for picking up weak energy based on seismic data to predict crack data volume according to claim 5, characterized in that, The sub-step 3a specifically involves: starting from the center point p0, expanding outwards, and comparing each of the current trace's seismic waveforms {w1, w2, ..., w...} within the longitudinal time window with wavelength n. m The similarity between this waveform and the previous seismic waveform w0 with wavelength n at location S in the stratum is recorded; the waveform w with the highest similarity is recorded. k ∈{w1,w2,…,w m The spatial position p of the midpoint of} k And the energy E of this waveform k ; Traverse all seismic traces in the calculation window to obtain p for each trace k The stratum S formed by the point.

7. The method for picking fracture prediction data volume based on weak energy of seismic data according to claim 1 or 3, characterized in that, In sub-step 3a, the energy of the waveform is equal to the average value of the square sum of the amplitudes of each point.

8. The method for picking up weak energy based on seismic data to predict crack data volume according to claim 1 or 3, characterized in that, In sub-step 3a, the similarity of the waveforms is compared by using the correlation coefficient or the vector angle.

9. The method for picking up weak energy based on seismic data to predict crack data volume according to claim 3, characterized in that, In sub-step 3c, the energy of the p0point waveform is scaled in the direction of the minimum point, using any factor representative of the energy or geometric difference between the p0point and other points in that direction; depending on the magnitude of the factor, a certain scaling amplitude is limited, and the p0point is scaled by that amplitude, thereby modifying the fault or fracture confidence of the p0point. f’ In sub-step 3c, the energy of the p0point waveform is scaled in the direction of the minimum point, using any factor representative of the energy or geometric difference between the p0point and other points in that direction; depending on the magnitude of the factor, a certain scaling amplitude is limited, and the p0point is scaled by that amplitude, thereby modifying the fault or fracture confidence of the p0point.

10. The method for picking up weak energy based on seismic data to predict the data volume of cracks according to claim 9, characterized in that, In said substep 3c, when scaling f’ a second coefficient is used, said second coefficient being greater than 0 and less than 1.

11. A method for fracture identification based on weak energy picking of seismic data, characterized in that, The fracture identification method uses the final weak-energy picked-up fracture prediction data volume obtained by the method for picking up the fracture prediction data volume as claimed in any one of claims 1-10 as the data volume for fracture identification.

Images