A method and device for picking up residual time difference based on Markov decision process

Abstract

The present invention discloses a residual time difference picking method and device based on Markov decision process, including inputting offset imaging gathers, obtaining imaging profiles by flattening and optimal stacking; designing picking parameters of reflection layers, calculating attribute information of imaging profiles, using Markov decision process to track the event axis, and generating characteristic imaging profiles; determining the seed point state for a single CDP gather, and using energy spectrum for initial picking to obtain state space; establishing a Markov decision process model, and using Markov decision process to pick reflection layer information. The present invention regards the picking problem of RMO as a problem of finding the optimal curve, and establishes it in the Markov optimal decision process based on the structural constraints of the imaging profile, and dynamically changes the action of the model by establishing an interactive process between the picking target and the environmental state, obtains the corresponding value, and continuously adjusts the state of the system, thereby achieving the best picking.

Description

Technical Field

[0001] The invention relates to the technical field of geophysical exploration, in particular to a residual time difference picking method and device based on Markov decision process. Background Art

[0002] In exploration seismic, the common imaging point gather plays a role of connecting the past and the future. Not only can the common imaging point gather be used for migration velocity analysis and updating velocity models, but also for amplitude variation with angle / offset (AVA / AVO) inversion to guide reservoir prediction and reservoir description. In the process of background velocity modeling, when the velocity is inaccurate, the imaging gather will appear as an upward or downward bend; only when the velocity is accurate, the imaging results of different offsets / angles focus on one point, and the imaging gather will appear flat. The velocity information is implicit in the imaging gather, and the residual move-out (RMO) on the imaging gather can be used for migration velocity analysis and background velocity update. In addition, picking up accurate RMO information can achieve in-phase stacking in the gather stacking process, significantly improving the imaging quality. Therefore, it is very important to automatically and accurately pick up RMO information from complex and changeable imaging gathers.

[0003] For the picking of RMO on imaging gathers, the most widely used methods are the gamma spectrum scanning method based on a single parameter (Al-Yahya, 1989) or the scanning method based on polynomial parabolic equations (Adler and Brandwood, 1999). However, these methods are only suitable for simple small offset situations. In the case of long offset or anisotropy, the residual time difference becomes extremely complex, and it is difficult to pick the correct curve with a single parameter. Siliqi et al. (2003) used a dual-parameter scanning method to consider the influence of anisotropy and long offset. Siliqi et al. (2007) used a high-order polynomial expansion method to perform multi-parameter scanning to adapt to arbitrarily complex RMO curves. Liu and Han (2010) used the plane wave destruction method to calculate the slope of the residual time difference on the angle gather using local linear relationships. Hilburn et al. (2014) proposed to reduce RMO picking to a path search problem and use a dynamic programming algorithm to solve it. Zhang et al. (2014) used the energy spectrum method and the high-order cumulant method to pick up the RMO required at different stages of traveltime tomography inversion to cope with imaging gathers of different complex situations. Jin et al. (2017) used the similarity coefficient spectrum estimation method and the optical flow method to pick up the residual time difference in reflection wave tomography.

[0004] Due to factors such as inaccurate migration velocity, especially the residual depth difference caused by inaccurate near-surface velocity, wavelet space variation caused by the influence of complex near-surface factors on excitation and reception, and noise caused by data quality and migration algorithm, the reflection waveform information on the imaging gather changes abnormally complexly. It is difficult to correctly pick up the residual time difference information of the reflection axis using conventional methods, which affects the subsequent migration velocity analysis and optimal stacking. Summary of the invention

[0005] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0006] In view of the above and / or existing problems of automatically picking up residual moveout (RMO) on imaging gathers in exploration seismic, the present invention is proposed.

[0007] Therefore, the problem to be solved by the present invention is how to automatically and accurately pick up the residual time difference information from the complex and changeable imaging gathers.

[0008] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0009] In a first aspect, an embodiment of the present invention provides a residual time difference picking method based on a Markov decision process, which comprises:

[0010] Input the migration imaging gathers and obtain the imaging profile by flattening and optimal stacking;

[0011] Design the picking parameters of the reflection layer, calculate the attribute information of the imaging section, use the Markov decision process to track the phase axis, and generate the characteristic imaging section;

[0012] For a single CDP gather, determine the seed point state, and use the energy spectrum for initial picking to obtain the state space;

[0013] A Markov decision process model is established, and the Markov decision process is used to pick up the reflection layer information, and the residual time difference results of the picked imaging gathers are output.

[0014] As a preferred solution of the residual time difference picking method based on Markov decision process of the present invention, wherein: the input migration imaging gather, through flattening and optimal stacking, obtains the imaging section including:

[0015] Input migration imaging gathers;

[0016] The imaging gathers are flattened by using the dynamic time planning method, and the weighting coefficients are calculated by using the local similarity coefficient method, and the flattened gathers are weightedly superimposed.

[0017] Acquire a focused imaging section.

[0018] As a preferred solution of the residual time difference picking method based on Markov decision process of the present invention, wherein: for a single CDP gather, determining the seed point state includes:

[0019] The generated characteristic imaging section is used as the initial seed point of each CDP gather;

[0020] The energy spectrum is used to perform a single parameter parabolic scan to form an energy spectrum;

[0021] The initial residual moveout position is determined based on the energy spectrum maximum.

[0022] As a preferred solution of the residual time difference picking method based on Markov decision process of the present invention, the specific method for determining the seed point state for a single CDP gather is:

[0023] Calculating the local similarity coefficient between each trace and the stacked profile along the initial residual moveout position;

[0024] The final seed point position is determined according to the size of the similarity coefficient spectrum.

[0025] As a preferred solution of the residual time difference picking method based on Markov decision process described in the present invention, wherein: the establishment of Markov decision process model includes decision time, state, action, state transition probability and benefit function, wherein,

[0026] The decision moment is the extra dimension of the expansion of a single CDP imaging gather, which is an offset or angle in the 2D case;

[0027] The state is the longitudinal spatial position on a single imaging gather, and the set of states is expressed as Where dz1 and dz2 are the widths of the upper and lower interfaces respectively. The middle position of the state space at different decision moments is determined by the single-parameter parabolic scanning method. The position of the seed point Determined by characteristic reflection structure;

[0028] The action represents the distance (angle) from the offset x i to x j movement;

[0029] The state transition probability is the system changing from state Transfer to state probability;

[0030] The profit function is determined according to the similarity between adjacent tracks.

[0031] As a preferred solution of the residual time difference picking method based on Markov decision process described in the present invention, wherein: the Markov decision process is used to pick up the reflection layer information, and the residual time difference result of the picked imaging gather is output, including:

[0032] When the decision moment reaches the boundary or the cumulative profit falls below a given threshold, tracking stops.

[0033] As a preferred solution of the residual time difference picking method based on Markov decision process of the present invention, wherein: the use of Markov decision process to pick up reflection layer information and output the picked up residual time difference result of imaging gathers also include:

[0034] When all seed points of a single CDP gather are tracked, the remaining time difference information of the current CDP gather is recorded;

[0035] When all CDPs are cycled, the remaining time difference picking of the imaging gathers of the entire profile is completed and the results are finally output.

[0036] In a second aspect, an embodiment of the present invention provides a residual time difference picking system based on a Markov decision process, which includes:

[0037] A processing module, used for obtaining an imaging profile by flattening and optimally stacking the input migration imaging gathers;

[0038] The calculation module is used to design the picking parameters of the reflection layer and calculate the attribute information of the imaging section, and use the Markov decision process to track the phase axis and generate the characteristic imaging section;

[0039] The picking module is used to determine the seed point state for a single CDP gather and use the energy spectrum for initial picking to obtain the state space;

[0040] The analysis module is used to establish a Markov decision process model, pick up reflection layer information using the Markov decision process, and output the residual time difference results of the picked imaging gathers.

[0041] In a third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein: the processor implements any step of the above method when executing the computer program.

[0042] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: the computer program implements any step of the above method when executed by a processor.

[0043] The beneficial effects of the present invention are:

[0044] 1. By flattening and stacking the imaging gathers, the problem of unfocused imaging caused by inaccurate migration velocity can be effectively alleviated. By picking up the structural information of the focused imaging section, an effective seed point can be provided for RMO picking. Since only the characteristic reflection information is retained, the non-reflection structural information can be effectively removed to avoid interference from multiple waves and other problems.

[0045] 2. Using the energy spectrum for preliminary picking can effectively reduce the size of the state space, avoid interference from other reflection axes, and improve computational efficiency. The results of the preliminary picking can also be used in the Markov decision process.

[0046] 3. The Markov decision process can effectively improve the accuracy of residual time difference picking. The complex and changing residual time difference on the imaging gather cannot obtain accurate information through the parameter scanning method. Through the interaction of state, action and benefit at different times, the influence of local single channel and bad channel can be avoided, and the correct residual time difference information can be tracked robustly. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0048] Figure 1 Flowchart of the residual time difference picking method based on Markov decision process.

[0049] Figure 2 Schematic diagram of updating seed points for conventional imaging gathers.

[0050] Figure 3 Schematic diagram for updating missing seed points for small-offset imaging gathers.

[0051] Figure 4 is the migration imaging gather.

[0052] Figure 5 The imaging results after flattening the gathers, optimal stacking and the extracted reflection structure.

[0053] Figure 6 Schematic diagram of the scanning range of the RMO picking process for a single CDP imaging gather.

[0054] Figure 7 Energy spectrum calculated for the RMO picking process for a single CDP imaging gather.

[0055] Figure 8 RMO results picked for a single CDP imaging gather.

[0056] Fig. 9 is the picked RMO and the imaging gather at the corresponding position.

[0057] Fig.10 is the imaging gather at a certain position.

[0058] Fig.11 Compare the picking results for a certain location. DETAILED DESCRIPTION

[0059] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0060] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0061] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0062] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0063] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0064] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0065] Example 1

[0066] Reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and provides a residual time difference picking method based on a Markov decision process, comprising:

[0067] S100: Input the migration imaging gathers and obtain the imaging profile by flattening and optimal stacking.

[0068] It should be noted that the imaging gathers are flattened by using the dynamic time planning method, and the weighting coefficients are calculated by using the local similarity coefficient method. The flattened gathers are weightedly superimposed to obtain the focused imaging profile.

[0069] This step process can be expressed as:

[0070]

[0071]

[0072]

[0073] Where I(x,i) is the input migration imaging gather, is the imaging gather corrected by the dynamic time planning method, ε is the threshold of the local correlation coefficient, η(x,i) is the calculated local correlation coefficient, and w(x,i) is the weighting coefficient calculated by the local similarity coefficient method. K(x)=∑ i w(x,i) is the normalization coefficient, and I(x) is the focused imaging result after processing.

[0074] S200: preprocessing is performed on the focused imaging section, picking parameters of the reflection layer are designed, and attribute information of the imaging section is calculated.

[0075] Furthermore, based on the attribute information, the Markov decision process is used to track the reflection event and obtain the characteristic imaging profile S(x):

[0076]

[0077] S300: The obtained characteristic imaging profile S(x) is used as the initial seed point of each CDP gather, and a single-parameter parabolic scan is performed using the energy spectrum to form an energy spectrum, and then the initial RMO position is determined based on the maximum value of the energy spectrum. All CDPs are looped to obtain the initial RMO positions of all CDPs. Finally, the RMO positions of all CDPs are filtered along the structure of the characteristic imaging profile to remove outliers.

[0078] It should be noted that in order to avoid small offset (angle) loss, the Figure 2 The strategy shown determines the lateral position of the seed point. The local similarity coefficients of each trace and the stacked section are calculated along the initial RMO position shown by the dotted line, and the final seed point position is determined according to the size of the similarity coefficient spectrum, as shown by the five-pointed star.

[0079] S400: For each CDP gather, a Markov decision process event picking model is established, including decision time, state, action, state transition probability and benefit function.

[0080] Specifically, the decision moment is the extra dimension to which a single CDP imaging gather is extended, which is the offset or angle in the 2D case.

[0081] The state represents the longitudinal spatial position on a single imaging gather. The set of states can be expressed as Where dz1 and dz2 are the widths of the upper and lower interfaces respectively. is the middle position of the state space at different decision moments, which is determined by a single-parameter parabolic sweep method. The position of the seed point Determined by characteristic reflection structure.

[0082] Action represents the offset (angle) x i to x j movement.

[0083] The state transition probability is the system transition from state Transfer to state Probability:

[0084]

[0085]

[0086]

[0087] The profit function is determined based on the similarity of adjacent channels.

[0088] Furthermore, a Markov decision process is used for tracking, and tracking is stopped when the decision moment reaches the boundary or the cumulative benefit is lower than a given threshold.

[0089] Furthermore, when the tracking of all seed points of a single CDP gather is completed, the RMO information of the current CDP gather is recorded.

[0090] Furthermore, all CDPs are looped, and when the remaining time difference picking of the imaging gathers of the entire profile is completed, the results are finally output.

[0091] Furthermore, this embodiment also provides a residual time difference picking system based on a Markov decision process, comprising:

[0092] A processing module, used for obtaining an imaging profile by flattening and optimally stacking the input migration imaging gathers;

[0093] The calculation module is used to design the picking parameters of the reflection layer and calculate the attribute information of the imaging section, and use the Markov decision process to track the phase axis and generate the characteristic imaging section;

[0094] The picking module is used to determine the seed point state for a single CDP gather and use the energy spectrum for initial picking to obtain the state space;

[0095] The analysis module is used to establish a Markov decision process model, pick up reflection layer information using the Markov decision process, and output the residual time difference results of the picked imaging gathers.

[0096] This embodiment further provides a computer device, which is applicable to the case of the residual time difference picking method based on the Markov decision process, including:

[0097] A memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the residual time difference picking method based on the Markov decision process as proposed in the above embodiment.

[0098] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0099] This embodiment further provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for picking up residual time differences based on a Markov decision process as proposed in the above embodiment is implemented.

[0100] The storage medium proposed in this embodiment and the data storage method proposed in the above embodiment belong to the same inventive concept. The technical details not fully described in this embodiment can be found in the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0101] Example 2

[0102] Reference Figures 3 to 11 , which is the second embodiment of the present invention, provides a residual time difference picking method based on a Markov decision process. In order to verify the beneficial effects of the present invention, a scientific demonstration is carried out through comparative experiments.

[0103] First, taking the whale model as an example, the process of picking up the residual time difference is explained.

[0104] It should be noted that the imaging gathers obtained by prestack depth migration using background velocity are as follows: Figure 4 As shown in Figure 2. Since the background velocity is lower than the accurate velocity, the imaging gathers appear to be warped, and there is an obvious residual time difference at different offsets. Therefore, the dynamic time planning method is used to flatten the imaging gathers and perform weighted stacking. The results are shown in Figure 2. Figure 5 Then the Markov decision method is used to track the characteristic reflection axis on the imaging section, and the result is as follows Figure 5 The position marked in is shown.

[0105] For each CDP imaging gather, Figure 5 The position marked in , calculates its energy spectrum, selects the coefficient with the largest energy spectrum and performs filtering. Figures 6 to 8 The results for the CDP201 position are shown. Figure 6 The energy spectrum scan range is shown. Figure 7 As shown. Figure 5 It can be seen that only three valid reflection axes are selected. Therefore, the corresponding three position energy spectrum values ​​are extracted to obtain the initial scanning results. Based on the initial results, the Markov decision method is used to update the remaining time difference, and the result is as follows: Figure 8 As shown in the middle circle, it is overlaid on the original imaging gather. It can be seen that the final picked residual time difference result is consistent with the real result. After looping through all CDP positions, the final residual time difference result is as follows Fig. 9 shown.

[0106] Next, some actual data is used to further test the effectiveness of the residual time difference picking.

[0107] The imaging gathers at a certain location are as follows: Fig.10 As shown in Figure 2. Due to inaccurate migration velocity, different types of residual time differences appear on the imaging gathers. The energy spectrum scanning method is used to obtain the initial picking results, and the Markov decision method is used to update the picking results on this basis. The final picking results are shown in Figure 2. Fig.11 It can be seen that the residual time difference obtained by the method proposed in the present invention conforms to the variation trend of the imaging gather.

[0108] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A residual time difference picking method based on Markov decision process, characterized in that: include, Input the migration imaging gathers and obtain the imaging profile by flattening and weighted stacking; Design the picking parameters of the reflection layer, calculate the attribute information of the imaging section, use the Markov decision process to track the phase axis, and generate the characteristic imaging section; For a single CDP gather, determine the seed point state, and use the energy spectrum for initial picking to obtain the state space; Establish a Markov decision process model, use the Markov decision process to pick up the reflection layer information, and output the residual time difference results of the picked imaging gathers; The Markov decision process model is established including decision time, state, action, state transition probability and benefit function, wherein: The decision moment is the extra dimension of the expansion of a single CDP imaging gather, which is an offset or angle in the 2D case; The state is the longitudinal spatial position on a single imaging gather, and the set of states is expressed as Where dz1 and dz2 are the widths of the upper and lower interfaces respectively. The middle position of the state space at different decision moments is determined by the single-parameter parabolic scanning method. The position of the seed point Determined by characteristic reflection structure; The action represents the distance or angle x from the offset i to x j movement; The state transition probability is the system changing from state Transfer to state probability; The profit function is determined according to the similarity between adjacent tracks.

2. The residual time difference picking method based on Markov decision process as claimed in claim 1, characterized in that: The input migration imaging gathers are flattened and weighted stacked to obtain imaging profiles including: Input migration imaging gathers; The imaging gathers are flattened by using the dynamic time planning method, and the weighting coefficients are calculated by using the local similarity coefficient method, and the flattened gathers are weightedly superimposed. Acquire a focused imaging section.

3. The residual time difference picking method based on Markov decision process as claimed in claim 2, characterized in that: For a single CDP gather, determining the seed point state includes: The generated characteristic imaging section is used as the initial seed point of each CDP gather; The energy spectrum is used to perform a single parameter parabolic scan to form an energy spectrum; The initial residual moveout position is determined based on the energy spectrum maximum.

4. The residual time difference picking method based on Markov decision process as claimed in claim 3, characterized in that: The specific method for determining the seed point state for a single CDP gather is: Calculating the local similarity coefficient between each trace and the stacked profile along the initial residual moveout position; The final seed point position is determined according to the size of the similarity coefficient spectrum.

5. The residual time difference picking method based on Markov decision process as claimed in claim 4, characterized in that: The Markov decision process is used to pick up the reflection layer information, and the residual time difference result of the picked imaging gather is output, including: When the decision moment reaches the boundary or the cumulative profit falls below a given threshold, tracking stops.

6. The method for extracting residual time difference based on Markov decision process as claimed in claim 5, characterized in that: The method of picking up the reflection layer information by using the Markov decision process and outputting the residual time difference result of the picked imaging gather also includes: When all seed points of a single CDP gather are tracked, the remaining time difference information of the current CDP gather is recorded; All CDPs are looped, and when the remaining time difference picking of the imaging gathers of the entire profile is completed, the results are finally output.

7. A residual time difference picking system based on a Markov decision process, based on the residual time difference picking method based on a Markov decision process according to any one of claims 1 to 6, characterized in that: include, A processing module, used for obtaining an imaging profile by flattening and weighted stacking the input migration imaging gathers; The calculation module is used to design the picking parameters of the reflection layer and calculate the attribute information of the imaging section, and use the Markov decision process to track the phase axis and generate the characteristic imaging section; The picking module is used to determine the seed point state for a single CDP gather and use the energy spectrum for initial picking to obtain the state space; The analysis module is used to establish a Markov decision process model, adopt the Markov decision process to pick up the reflection layer information, and output the residual time difference result of the picked imaging gather.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images