Method for dynamically correcting seismic data and picking up and positioning seismic phase of micro-seismic event

Abstract

The invention relates to the field of microseism signal identification and pickup, in particular to a seismic phase pickup and positioning method for dynamic correction of seismic data and microseism events, which comprises the following steps of: picking up a preliminary seismic phase for seismic channel data of each station in scatter station combined seismic data; calculating a travel time curved surface according to the perforation position and the speed model; and carrying out dynamic correction on the initial seismic phase by adopting the travel time curved surface. And obtaining a dynamic correction result of the initial seismic phase, and screening the dynamic correction result to obtain the seismic phase of the micro-seismic event. The seismic facies of the micro-seismic event obtained according to the seismic facies pickup method of the seismic event is positioned. According to the method, higher-quality travel time input is provided for microseism positioning, and the method is simple, efficient and high in practicability.

Description

Technical Field

[0001] The present invention relates to the field of microseismic signal recognition and picking, and more specifically, to a method for dynamic correction of seismic data and phase picking and positioning of microseismic events. Background Art

[0002] In hydraulic fracturing monitoring, the layout and maintenance costs of large lines are high (especially in areas with complex terrain such as mountains), which is not suitable for long-term observation needs. Scattered station observations are economical, flexible, and practical, and can replace large-line observations as a long-term monitoring solution. However, compared with seismic survey lines, the number of scattered stations is small, and the arrangement of the observation system is often not regular enough, which increases the difficulty of identifying the P and S seismic phases corresponding to the fracturing signal.

[0003] With the application of artificial intelligence and neural network technology in the field of microseismicity, the recognition and picking of microseismic signals have become more efficient and automated. Existing intelligent picking technologies are mostly based on the training of neural networks based on the data and labels of earthquakes at a single station, and the training set lacks earthquake correlation information from different stations. Therefore, on the seismic profiles of multiple stations, in addition to hydraulic fracturing events, there may be microseismic events or prediction errors that are not caused by hydraulic fracturing on the seismic traces of some stations.

[0004] High-precision microseismic positioning requires the combination of arrival time information of the same event at multiple stations. The input of non-fracturing events or prediction errors will lead to false positioning results. Therefore, a microseismic signal correlation and screening technology is urgently needed to eliminate isolated non-fracturing signals and provide a corresponding set of travel time data for the positioning of each microseismic event. Summary of the invention

[0005] The purpose of the present invention is to solve the above-mentioned problems existing in the prior art and provide a method for dynamic correction of seismic data.

[0006] The present invention is realized by the following technical scheme: a method for dynamic correction of seismic data, comprising the following steps:

[0007] Pick up the preliminary seismic phase from the seismic trace data of each station in the joint seismic data of scattered stations;

[0008] Calculate travel-time surfaces based on perforation locations and velocity models;

[0009] The travel time surface is used to perform dynamic correction on the preliminary seismic phase.

[0010] Furthermore, a neural network is used to pick up the preliminary seismic phases of the seismic trace data of each station one by one.

[0011] Furthermore, the travel time from the perforation to each station is calculated through 3D travel time forward modeling. Thus, the travel time surface is obtained.

[0012] Furthermore, the dynamic correction calculation formula is as follows:

[0013]

[0014] In the formula, i is the station number, j is the number of the preliminary seismic phase on the seismic trace data of the station; is the arrival time of the seismic phase after the dynamic correction, is the time when the initial seismic phase arrives, is the travel time of the perforation signal corresponding to the station.

[0015] The present invention also provides a method for picking up the seismic phase of a microseismic event, which uses the dynamic correction method of seismic data described in the present invention to obtain the dynamic correction result of the preliminary seismic phase, and screens the dynamic correction result to obtain the seismic phase of the microseismic event.

[0016] Furthermore, an event association window is used to screen the dynamic correction results to obtain the seismic phases of the microseismic event: the number of seismic phases picked in each time window is counted by a time window sliding method, and the time windows greater than the number threshold are saved as event association windows.

[0017] Furthermore, each event correlation window T valid :

[0018]

[0019] In the formula, i is the station number, j is the number of the preliminary seismic phase on the seismic trace data of the station; is the arrival time of the seismic phase after the dynamic correction, Yes The result after applying the event correlation window. After the reaction correction, the microseismic signal after screening arrives at Can be used as input to the positioning program.

[0020] Furthermore, the sliding step size of the time window is smaller than the time window size.

[0021] Furthermore, the seismic trace data of each station after dynamic correction are correlated through the event association window, and the seismic phases that do not belong to the event association window are removed, and then the seismic trace data are subjected to dynamic correction processing to obtain a group of seismic phases corresponding to each microseismic event.

[0022] The present invention also provides a method for locating microseismic events, which locates the seismic phases of microseismic events obtained by the seismic phase picking method of seismic events described in the present invention.

[0023] Compared with the prior art, the beneficial effects of the present invention include:

[0024] 1. The present invention uses the perforation position as the source point, and the travel time surface obtained by calculation is very close to the travel time surface of the fracturing microseismic event. The travel time surface is an excellent choice for seismic correlation and screening. The fracturing signal is correlated, screened and extracted by the surface, and only one or more groups of picking results corresponding to the microseismic event are retained on the seismic profile, and the picking corresponding to noise, isolated events, etc. is successfully eliminated.

[0025] 2. The method of using a unique travel time surface (fracture well perforation position) for earthquake correlation is the first application in the field of "microseismic". The present invention not only provides higher quality travel time input for microseismic positioning, but is also simple, efficient and highly practical. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a flow chart;

[0027] Figure 2 The results of using the PhaseNet neural network to predict and pick up seismic phases at each station;

[0028] FIG3 is a schematic diagram of a travel time surface (travel time from the perforation position to each station) obtained by performing travel time forward modeling based on the perforation position and the three-dimensional velocity model;

[0029] Figure 4 The schematic diagram of the effect of dynamic correction of the picking results of multiple stations using the perforation travel time surface. At this time, the seismic phase picking is close to a straight line;

[0030] Figure 5 This is a schematic diagram of the time window sliding search and phase picking statistics based on the NMO post-picking results;

[0031] Figure 6 is a schematic diagram of the effect of applying to actual data. Figure 1 In comparison, the event association window in this paper successfully eliminates non-fracturing signals and noise signals. DETAILED DESCRIPTION

[0032] Most of the microseismic events generated by hydraulic fracturing are located near the perforations, and the extension distance of the hydraulic fractures in the horizontal direction (hundreds of meters) is much smaller than the buried depth of the well section (several thousand meters). Therefore, if the perforation location is taken as the source point, the travel time surface obtained by forward calculation is very close to the travel time surface of the fracturing microseismic event, and the travel time surface is an excellent choice for seismic correlation and screening.

[0033] Since fracturing events mostly occur near perforations, the travel time surface calculated by perforation and detector position is close to that of fracturing events; however, non-fracturing events do not have such travel time characteristics.

[0034] The method of using a unique travel time surface (fracture well perforation position) for seismic correlation is the first application in the field of "microseismic". The present invention not only provides higher quality travel time input for microseismic positioning, but is also simple, efficient and highly practical.

[0035] The present invention aims to use the travel-time forward modeling surface corresponding to the perforation position to correlate and screen the microseismic events at multiple stations, eliminate the influence of non-fracturing events, prediction errors, etc., and provide a set of reliable travel-time data for high-precision positioning of each microseismic event.

[0036] First, the present invention provides a method for dynamic correction of seismic data, comprising the following steps:

[0037] Pick up the preliminary seismic phase from the seismic trace data of each station in the joint seismic data of scattered stations;

[0038] Calculate travel-time surfaces based on perforation locations and velocity models;

[0039] The travel time surface is used to perform dynamic correction on the preliminary seismic phase.

[0040] Preferably, a neural network is used to pick up preliminary seismic phases of the seismic trace data of each station one by one.

[0041] Preferably, the travel time from the perforation to each station is calculated by 3D travel time forward modeling. Thus, the travel time surface is obtained.

[0042] Preferably, the dynamic correction calculation formula is as follows:

[0043]

[0044] In the formula, i is the station number, j is the number of the preliminary seismic phase on the seismic trace data of the station; is the arrival time of the seismic phase after the dynamic correction, is the time when the initial seismic phase arrives, is the travel time of the perforation signal corresponding to the station.

[0045] The present invention also provides a method for picking up the seismic phase of a microseismic event, which uses the dynamic correction method of seismic data described in the present invention to obtain the dynamic correction result of the preliminary seismic phase, and screens the dynamic correction result to obtain the seismic phase of the microseismic event.

[0046] Preferably, an event association window is used to screen the dynamic correction results to obtain the seismic phases of the microseismic event: the number of seismic phases picked up in each time window is counted by a time window sliding method, and the time windows greater than the number threshold are saved as event association windows.

[0047] Preferably, each event association window T valid :

[0048]

[0049] In the formula, i is the station number, j is the number of the preliminary seismic phase on the seismic trace data of the station; is the arrival time of the seismic phase after the dynamic correction, Yes The result after applying the event correlation window. After the reaction correction, the microseismic signal after screening arrives at Can be used as input to the positioning program.

[0050] Preferably, the sliding step size of the time window is smaller than the time window size.

[0051] Preferably, the seismic trace data of each station after dynamic correction are correlated through the event association window, and the seismic phases that do not belong to the event association window are removed, and then the seismic trace data are subjected to dynamic correction processing to obtain a group of seismic phases corresponding to each microseismic event.

[0052] The present invention also provides a method for locating microseismic events, which locates the seismic phases of microseismic events obtained by the seismic phase picking method of seismic events described in the present invention.

[0053] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0054] Example 1

[0055] This embodiment provides a method for dynamic correction of seismic data, comprising the following steps:

[0056] Pick up the preliminary seismic phase from the seismic trace data of each station in the joint seismic data of scattered stations;

[0057] Calculate travel-time surfaces based on perforation locations and velocity models;

[0058] The travel time surface is used to perform dynamic correction on the preliminary seismic phase.

[0059] The specific steps include:

[0060] PhaseNet neural network is used to predict and pick up seismic phases at each station, and preliminary seismic phase picking results are obtained, such as Figure 2 As shown in the figure, the short red horizontal line represents the result of seismic phase picking, that is, each vertical line represents the seismic trace data of a station.

[0061] Example 2

[0062] The travel time from the perforation to each station is calculated through 3D travel time forward modeling. Thus, the travel time surface is obtained.

[0063] The calculated travel time surface is as follows: Figure 3a :Travel time calculation based on perforation, station location and velocity model. There are many ways to implement it. In this example, Tomo3D software is used for 3D travel time forward modeling (shortest path method and bending method) to calculate the travel time from the perforation to each station. like Figure 3b shown.

[0064] Example 3

[0065] The travel time surface is calculated based on the perforation position and the velocity model; the normal moveout correction (NMO) is performed on the initial picking result of the neural network using the travel time surface. NMO is also called dynamic correction.

[0066] Use the travel time surface to perform dynamic correction on the picking results, such as Figure 4 : Subtract the perforation signal travel time corresponding to each station from the picking result of each station.

[0067] The dynamic correction calculation formula is as follows:

[0068]

[0069] In the formula, i is the station number, j is the number of the preliminary seismic phase on the seismic trace data of the station; is the arrival time of the seismic phase after the dynamic correction, is the time when the initial seismic phase arrives, is the travel time of the perforation signal corresponding to the station.

[0070] Example 4

[0071] refer to Figure 1 As shown, a method for picking up the seismic phase of a microseismic event adopts a dynamic correction method of seismic data to obtain a dynamic correction result of a preliminary seismic phase, and screens the dynamic correction result to obtain the seismic phase of the microseismic event.

[0072] The method for dynamic correction of seismic data comprises the following steps:

[0073] Pick up the preliminary seismic phase from the seismic trace data of each station in the joint seismic data of scattered stations;

[0074] Calculate travel-time surfaces based on perforation locations and velocity models;

[0075] The travel time surface is used to perform dynamic correction on the preliminary seismic phase.

[0076] The specific steps include:

[0077] PhaseNet neural network is used to predict and pick up seismic phases at each station, and preliminary seismic phase picking results are obtained, such as Figure 2As shown in the figure, the short red horizontal line represents the result of seismic phase picking, that is, each vertical line represents the seismic trace data of a station.

[0078] The travel time from the perforation to each station is calculated through 3D travel time forward modeling. Thus, the travel time surface is obtained.

[0079] The calculated travel time surface is as follows: Figure 3a :Travel time calculation based on perforation, station location and velocity model. There are many ways to implement it. In this example, Tomo3D software is used for 3D travel time forward modeling (shortest path method and bending method) to calculate the travel time from the perforation to each station. like Figure 3b shown.

[0080] The travel time surface is calculated based on the perforation position and the velocity model; the normal moveout correction (NMO) is performed on the initial picking result of the neural network using the travel time surface. NMO is also called dynamic correction.

[0081] Use the travel time surface to perform dynamic correction on the picking results, such as Figure 4 : Subtract the perforation signal travel time corresponding to each station from the picking result of each station.

[0082] The dynamic correction calculation formula is as follows:

[0083]

[0084] In the formula, i is the station number, j is the number of the preliminary seismic phase on the seismic trace data of the station; is the arrival time of the seismic phase after the dynamic correction, is the time when the initial seismic phase arrives, is the travel time of the perforation signal corresponding to the station.

[0085] The event association window is used to screen the dynamic correction results to obtain the seismic phases of the microseismic event: the number of seismic phases picked in each time window is counted by a time window sliding method, and the time windows greater than the number threshold are saved as event association windows.

[0086] Based on the NMO post-picking results, the number of seismic phases picked in each time window is counted by sliding the time window, and the time windows that meet the quantity threshold are saved as event association windows; the time window with the number of seismic phases greater than the threshold is the event association window.

[0087] According to the event association window and its time range, the time range refers to the duration of the time window, and the association and screening of the NMO post-seismic phase picking results are achieved.

[0088] Use the time window sliding method to search and count the number of events in each time window, and determine the event correlation window based on the threshold, such as Figure 5: For the picking results of each data file, take a certain size of time window and sliding step (the sliding step should be slightly smaller than the time window size, that is, there should be a certain overlap between two adjacent time windows to avoid missing events).

[0089] The number of seismic phases in each time window is counted. If the number of seismic phases is greater than the preset value (4 in this example), the time window is considered to be successfully associated with a group of seismic phases (time window n) and the time window is saved; if it is less than the preset value, the time window is considered not to be a group of microseismic phases (time window m) and the time window is not retained.

[0090] In the above event correlation window application process, for each event correlation window, the picking results within each time window are retained, and the picking results outside the time window are discarded.

[0091] Each event correlation window T valid :

[0092]

[0093] In the formula, i is the station number, j is the number of the preliminary seismic phase on the seismic trace data of the station; is the arrival time of the seismic phase after the dynamic correction, Yes The result after applying the event correlation window. After the reaction correction, the microseismic signal after screening arrives at Can be used as input to the positioning program.

[0094] If there are two phases on a line, j is 1 or 2, and the size represents the order of arrival. For example, if there is a group of phases at 15000ms (there is none in this example; if there is, the red line is also picked up), then this group of phases arrives earlier than the group near 20000ms.

[0095] The seismic trace data of each station after dynamic correction are correlated through the event association window, and the seismic phases that do not belong to the event association window are removed. Then, the seismic trace data are processed for dynamic correction to obtain a set of seismic phases corresponding to each microseismic event. The final travel time data reference Figure 6 shown.

[0096] Example 5

[0097] A method for locating a microseismic event, locating the microseismic event phase obtained by the method for picking the seismic event phase of the seismic event in Example 4. After the reaction correction, the microseismic signal after screening arrives at Can be used as input to the positioning program.

[0098] Use the time window sliding method to search and count the number of events in each time window, and determine the event association window based on the threshold. For the picking result of each data file, take a certain size of time window and sliding step (the sliding step should be slightly smaller than the time window size, that is, there should be a certain overlap between two adjacent time windows to avoid missing events).

[0099] The number of seismic phases in each time window is counted. If the number of seismic phases is greater than the preset value (6 in this example), the time window is considered to be successfully associated with a group of seismic phases (time window n) and the time window is saved; if it is less than the preset value, the time window is considered not to be a group of microseismic phases (time window m) and the time window is not retained.

[0100] The event association window is used for association and screening, and then the reaction correction processing is performed to obtain a set of seismic phases corresponding to each microseismic event.

[0101] The outdated window sliding and threshold screening method (step 5 of the flowchart) only finds the time windows that meet the conditions (event association windows), but this step does not eliminate invalid picks (non-fracturing events, prediction errors); step 6 uses the time window obtained in step 5 to eliminate invalid picks, that is, "using event association windows to associate and screen".

[0102] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "connected" and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0103] In the description of the present invention, unless otherwise specified, the terms "upper", "lower", "left", "right", "inside", "outside", etc. indicate directions or positional relationships based on the directions or positional relationships shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction. Therefore, they cannot be understood as limitations on the present invention.

[0104] The above technical scheme is only a specific implementation method of the present invention. For those skilled in the art, it is easy to make various types of improvements or modifications based on the principles disclosed in the present invention, and it is not limited to the technical scheme described in the above specific embodiments of the present invention. Therefore, the above description is only preferred and does not have a restrictive meaning.

Claims

1. A method for dynamic correction of seismic data, characterized in that: The following steps are involved: Pick up the preliminary seismic phase from the seismic trace data of each station in the joint seismic data of scattered stations; Calculate travel-time surfaces based on perforation locations and velocity models; The travel time surface is used to perform dynamic correction on the preliminary seismic phase.

2. The method for dynamic correction of seismic data according to claim 1, characterized in that: A neural network is used to pick up the preliminary seismic phases of the seismic trace data of each station one by one.

3. The method for dynamic correction of seismic data according to claim 1, characterized in that: The travel time from the perforation to each station is calculated through 3D travel time forward modeling. Thus, the travel time surface is obtained.

4. The method for dynamic correction of seismic data according to claim 1, characterized in that: The dynamic correction calculation formula is as follows: In the formula, i is the station number, j is the number of the preliminary seismic phase on the seismic trace data of the station; is the arrival time of the seismic phase after the dynamic correction, is the time when the initial seismic phase arrives, is the travel time of the perforation signal corresponding to the station.

5. A method for picking up seismic phases of microseismic events, characterized in that: The method for dynamic correction of seismic data as claimed in any one of claims 1 to 4 is used to obtain the dynamic correction result of the preliminary seismic phase, and the dynamic correction result is screened to obtain the seismic phase of the microseismic event.

6. The method for picking up microseismic phases according to claim 5, characterized in that: The event association window is used to screen the dynamic correction results to obtain the seismic phases of the microseismic event: the number of seismic phases picked in each time window is counted by a time window sliding method, and the time windows greater than the number threshold are saved as event association windows.

7. The method for picking up microseismic phases according to claim 6, characterized in that: For each event correlation window T valid : In the formula, i is the station number, j is the number of the preliminary seismic phase on the seismic trace data of the station; is the arrival time of the seismic phase after the dynamic correction, Yes The result after applying the event correlation window. After the reaction correction, the microseismic signal after screening arrives at Can be used as input to the positioning program.

8. The method for picking up microseismic phases according to claim 6, characterized in that: The sliding step size of the time window is smaller than the time window size.

9. The method for picking up microseismic phases according to claim 6, characterized in that: The seismic trace data of each station after dynamic correction are correlated through the event correlation window, and the seismic phases not belonging to the event correlation window are removed. Then, the seismic trace data are subjected to dynamic correction processing to obtain a group of seismic phases corresponding to each microseismic event.

10. A method for locating microseismic events, characterized in that: The seismic phase of the microseismic event obtained by the seismic phase picking method of seismic events according to claim 6 is used for positioning.

Images