Seismic horizon generation method based on peak search and relocation first and then breakpoint spatial tracking

Abstract

The present invention provides a method for generating seismic horizons that first searches for peak values ​​and then spatially tracks breakpoints. The method comprises: S101, determining the polarity of the target horizon based on well-seismic calibration, and clarifying the peak characteristics of the target horizon; S102, reading reference horizon data containing line number, channel number, and horizon time, and setting the search window size; S103, using the reference horizon as a constraint on the three-dimensional data volume to identify the peak or trough target horizon; S104, searching for breakpoints within the stratum one by one based on the continuity characteristics of the stratum and the characteristics of the faults, leaving the data for the left and right breakpoints blank, and saving the remaining search results, thereby obtaining the target horizon that achieves peak relocation and breakpoint characterization. The horizons generated by the present invention have the characteristics of high accuracy, fast search speed, and strong adaptability to geological conditions, and can provide basic data for subsequent seismic attribute extraction, reservoir description, and prediction.

Description

Technical Field

[0001] The present invention belongs to the field of seismic data target processing and reservoir description, and in particular relates to a seismic horizon generation method that first searches for and moves peak values ​​and then tracks breakpoint space. Background Art

[0002] Reservoir stratigraphic information is essential data for reservoir description and prediction. During seismic data interpretation, reservoir seismic horizons are first calibrated using synthetic records to obtain seismic horizon time information corresponding to wellpoint geological stratification. Then, through well-linked profile interpretation, the interpreted horizons for the backbone profile are obtained. Finally, through the interpretation and closure of all survey lines in the work area, stratigraphic information for the entire work area is obtained. For thin interbeds, calibration can lead to phase errors. For example, a peak reflection may correspond to a geological stratification, but it is actually interpreted as a trough, and stratigraphic information has already been traced for the entire area. In such cases, developing a technology that can automatically move the entire stratigraphic layer at the trough to the underlying peak is a valuable task. Alternatively, even if the seismic calibration is correct and full-area tracing has been completed for one interface, the desire to use the existing horizon as a reference to search for and trace another interface in the reservoir is essential. This can significantly reduce manual interpretation workload, improve interpretation accuracy, and quickly and efficiently obtain new stratigraphic information.

[0003] Strata and faults are inseparable. Automatically identifying breakpoints in newly moved stratigraphic layers based on their drop characteristics is extremely valuable for detailed reservoir description and fault characterization. Extracted stratigraphic layers can subsequently be subjected to detailed analysis of the dynamic properties of seismic waves within them, as well as the study of interlayer properties. This invention aims to develop such a practical technology, providing a novel method and technique for automated interpretation of thin interbedded stratigraphic layers and reservoir prediction. Summary of the Invention

[0004] The embodiment of the present application provides a seismic layer generation method that first searches for and moves the peak value and then tracks the breakpoint space, which provides basic data for subsequent seismic attribute extraction and reservoir description and prediction, and improves the exploration and development capabilities of thin interbeds.

[0005] The present application provides a method for generating seismic horizons by first searching for and moving peak values ​​and then spatially tracking breakpoints, including:

[0006] S101, determining the polarity of the target layer based on well-seismic calibration, and clarifying the peak characteristics of the search target layer;

[0007] S102, read the reference layer data including line number, track number, and layer time, and set the search window range size;

[0008] S103, using the reference layer as a constraint on the three-dimensional data volume, and selecting a peak or trough target layer;

[0009] S104, based on the continuity characteristics of the stratum and the characteristics of the fault, search for the breakpoints contained in the stratum one by one, leave the data of the left breakpoint and the right breakpoint blank, and save the other search results, so as to obtain the target layer for peak relocation and breakpoint characterization.

[0010] Wherein, step S102 includes:

[0011] If the downward search is for a strong peak reflection, a time window containing one peak is set. The strong peak reflection means that the amplitude value is positive and the maximum amplitude is greater than the maximum value of the adjacent peaks.

[0012] If the downward search is for a weak peak reflection under a strong reflection, it is planned to first set a start time and then set a time window containing the weak peak. The weak peak reflection means that the amplitude value is positive but the maximum amplitude is less than a preset threshold.

[0013] Wherein, step S103 includes:

[0014] For strong reflections, search using the maximum value search algorithm;

[0015] For weak peaks, search using the maximum value algorithm.

[0016] Among them, in step S103, according to the characteristics of the earthquake phase, the layer search and movement are performed in the following two ways:

[0017] Search and move of strong reflection layers. Strong reflection layers mean that the amplitude value is positive and the maximum amplitude is greater than the maximum value of adjacent peaks:

[0018] Under the constraint of the reference layer T0(i,j,k), the maximum value Xmax is set to 0 initially, and within the search window, it is determined whether X(i,j,k)>Xmax holds. X(i,j,k) is the three-dimensional data volume, Xmax is the maximum peak value to be searched, i, j, k are the line number, channel number, and time sequence number. After the search is completed, the layer time T1=T0(i,j,k)+(k-1)*dt is recorded, where dt is the sampling interval, thus achieving the move from the reference layer to the strong peak.

[0019] Search and move the weak reflection layer. A weak reflection layer means that the amplitude value is positive but the maximum amplitude is less than the preset threshold:

[0020] Constrained by the reference horizon T0(i,j,k), starting from the search point K0, within the search time window, it is determined that three conditions are simultaneously satisfied during the search: First, X(i,j,k)>0, which is defined as the main lobe of the wave crest; second, X(i,j,k)>X(i,j,k - 1), which means it is larger than the value of the adjacent point on the left; third, X(i,j,k)>=X(i,j,k + 1), which means it is not less than the value of the adjacent point on the right. During actual tests, it is found that there is a phenomenon of truncated peaks in seismic data, and it will be the same as the value of the adjacent point.

[0021] Among them, in step S104, the method for determining the fault boundary, that is, the left and right breakpoints tracking, includes:

[0022] Judgment of the left boundary: j = ISTR, NTR - 1, where NTR represents the number of traces in a line, and "j = ISTR, NTR - 1" means that the value of j ranges from ISTR to NTR - 1. If |T j+1 -T j | < dt_LV_max, save the trace number j and time T j , where dt_LV_max is set as the maximum undulation time of the adjacent traces on the left side of the fault, and the initial value of ISTR is set to 1; otherwise, record the left boundary of the fault as LF = j, and start the judgment of the right boundary;

[0023] Judgment of the right boundary: j = LF + N_HC, LF + 1, where "j = LF + N_HC, LF + 1" means that the value of j ranges from LF + N_HC to LF + 1), search in reverse. If |T j -T j-1 | > dt_RV_max, where dt_RV_max is set as the maximum undulation time of the adjacent traces on the right side of the fault, save the right boundary of the fault RF = j + 1, and record the trace number l and time T of this segment in the small loop l , l = RF, LF + N_HC, where "RF, LF + N_HC" represents the range of the trace number to be recorded, and N_HC is the number of points for lateral continuity control of the formation; otherwise, continue the search; the small loop is the loop for judging a single fault;

[0024] After processing one fault, if LF + N_HF < NTR - N_HC, update the starting trace number ISTR = LF + N_HC + 1 for the search, and start a new fault identification loop, where N_HF is the number of points for lateral control of the fault; otherwise, save the trace numbers and times from LF + N_HF + 1 to NTR, and end the profile loop.

[0025] The seismic horizon generation method of the embodiment of the present application, which first performs peak search relocation and then breakpoint spatial tracking, has the following beneficial effects:

[0026] The present invention discloses a method for generating seismic horizons by first searching for and moving peak values ​​and then spatially tracking breakpoints, comprising: S101, determining the polarity of the target horizon based on well seismic calibration, and clarifying the peak characteristics of the target horizon; S102, reading reference horizon data including line number, channel number, and horizon time, and setting the search window size; S103, using the reference horizon as a constraint to target the peak or trough layer of the three-dimensional data volume; S104, searching for breakpoints contained in the stratum one by one based on the continuity characteristics of the stratum and the characteristics of the fault, leaving the data of the left and right breakpoints blank, and saving the other search results, thereby obtaining the target horizon for peak relocation and breakpoint characterization. The horizons generated by the present invention have the characteristics of high accuracy, fast search speed, and strong adaptability to geological conditions, and can provide basic data for subsequent seismic attribute extraction, reservoir description, and prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a cross-section diagram based on the interpretation of trough and peak;

[0028] Figure 2 A seismic synthetic record calibration map for obtaining well point horizon information;

[0029] Figure 3 Flow chart for moving layers with strong reflections from trough to peak;

[0030] Figure 4 It is a practical application section for moving the layer with strong reflection from the trough to the peak;

[0031] Figure 5 Flow chart for moving from strong reflection peak to weak reflection layer;

[0032] Figure 6 Practical application section for moving the strong reflection peak to the weak reflection layer;

[0033] Figure 7 Automatically trace and identify flow charts for breakpoints;

[0034] Figure 8 This is the effect diagram of identifying small micro-faults with abnormal points;

[0035] Figure 9 This is the effect diagram of fault identification with phase change;

[0036] Figure 10 It is the time slice along the layer of T40 layer reflected by the original trough;

[0037] Figure 11 The new T40 layer time slice is obtained by automatic tracking;

[0038] Figure 12 For the new one, the slice with the maximum amplitude along the layer of T41 is obtained by automatic tracking;

[0039] Figure 13 The new T41 layer time slice is obtained by automatic tracking;

[0040] Figure 14 This is the slice with the maximum amplitude along the layer obtained by tracking the new T41 layer. DETAILED DESCRIPTION

[0041] The present application will be further described below with reference to the accompanying drawings and embodiments.

[0042] The following description provides multiple embodiments of the present invention. Different embodiments may be replaced or combined, and thus this application may be considered to include all possible combinations of the same and / or different embodiments described. Thus, if one embodiment includes features A, B, and C, and another embodiment includes features B and D, then this application should also be considered to include embodiments containing one or more of all other possible combinations of A, B, C, and D, even though such embodiments may not be explicitly described in the following text.

[0043] Example 1

[0044] The present application provides a seismic layer generation method that first searches for and moves the peak value and then tracks the breakpoint space, including: S101, determining the polarity of the target layer according to well seismic calibration, and clarifying the peak characteristics of the search target layer; S102, reading the reference layer data including the line number, channel number, and layer time, and setting the search time window range size; S103, forcing the three-dimensional data body to the peak or trough target layer with the reference layer as a constraint; S104, searching the breakpoints contained in the formation one by one according to the continuity characteristics of the formation and the characteristics of the fault, leaving the data of the left breakpoint and the right breakpoint blank, and saving the other search results, so as to obtain the target layer for achieving peak movement and breakpoint characterization.

[0045] The present invention can provide a new extraction method for the interpretation of thin interbedded layers, thereby improving the description capability of thin interbedded oil and gas reservoirs.

[0046] Example 2

[0047] like Figure 1 As shown, Figure 1 For cross-sections based on trough interpretation and crest interpretation, the stratum interpreted as trough is located at the top, which does not conform to the well-seismic relationship and needs to be moved to the crest position below. This is the original intention of this patent. Figure 2 This is a calibration diagram of a seismic synthetic record in a certain actual work area. It can be seen that the target layer corresponds to the peak, which indicates that Figure 1 The horizon shown does require moving target processing.

[0048] Figure 1Such a migration from a trough to a strong peak can be achieved by searching for the maximum position where the peak is located. The technical flowchart used is as shown in Figure 3 Figure []. Another search window size Nw1 is a control parameter. For seismic data sampled at 1 ms, a 40-ms window is more appropriate. Figure 4 Figure [ ] is an actual application example of the horizon migration from a trough to a strong peak. After the above target processing, the horizon has been accurately migrated to the position of the underlying peak.

[0049] From the Figure 2 calibration, after obtaining the peak horizon T40 of the strong reflection, the slightly weaker peak reflection below corresponds to the bottom of the sand group, which is the T41 standard horizon. It is not as easy to trace as T40, and the energy and phase of the reflection axis change greatly. Using a fixed window to search for the maximum value as in Figure 3 is prone to errors. Therefore, the search algorithm shown in Figure 5 is adopted. First, a travel time is extended downward from the peak, and then the maximum value of the weak peak is searched. Tests on actual data show that there will be a peak truncation phenomenon in actual data, that is, there is no maximum value at the peak, but two or more equal values. At this time, if judged by X(i,j,k)>X(i,j,k+1), the searched extreme value cannot be obtained, and the judgment condition X(i,j,k)>=X(i,j,k+1) must be used. Figure 6 Figure [ ] is an application example of migrating the strong reflection peak T40 to the weaker reflection horizon T41. It can be seen that most of the horizon points have been well migrated, but there are also individual lithology change points. Since there is no characteristic of the amplitude changing from the previous peak to the trough and then to the peak, these local abnormal points need to be interactively corrected manually.

[0050] The above horizon migration does not consider the problem of fault breaks. In actual seismic areas, faults often exist. Therefore, it is necessary to identify fault breaks. The core idea of fault break identification is: to judge whether it is a fault according to the throw of the horizon at the fault. A very small time jitter may be caused by lithology changes or data anomalies. Therefore, a fault throw threshold needs to be set to avoid it. Faults have a certain strike. The main survey line is generally perpendicular to the strike of the structure. In addition, through the analysis of actual data, it is also found that there are differences in the throws of the left and right fault breaks. These are all technical problems to be considered. Figure 7 Figure [ ] is the flowchart for automatic tracking and identification of fault breaks. The left and right fault boundaries adopt the following search method. Note that the left and right fault throw thresholds dt_LV_max and dt_RV_max are different.

[0051] Judgment of the left boundary: j = ISTR, NTR - 1, where NTR represents the number of traces in a line. "j = ISTR, NTR - 1" means that the value of j ranges from ISTR to NTR - 1. If |T j+1 - T j | < dt_LV_max, save the trace number j and time Tj If dt_LV_max is set as the maximum undulating time of adjacent traces on the left side of the fault, set the initial value of ISTR as 1; otherwise, record the left boundary of the fault as LF = j and start the judgment of the right boundary.

[0052] Judgment of the right boundary: j = LF + N_HC, LF + 1 (where "j = LF + N_HC, LF + 1" means j takes values from LF + N_HC to LF + 1), search in reverse. If |T j -T j-1 | > dt_RV_max, set dt_RV_max as the maximum undulating time of adjacent traces on the right side of the fault, save the right boundary of the fault RF = j + 1, and record the trace number l and time T of this segment in the small loop. l where l = RF, LF + N_HC (where "RF, LF + N_HC" represents the range of trace numbers to be recorded), and N_HC is the number of points for lateral continuity control of the formation; otherwise, continue the search; the small loop is the loop for judging a single fault.

[0053] After processing one fault, if LF + N_HF < NTR - N_HC, update the starting trace number ISTR = LF + N_HC + 1 for the search and start a new fault identification loop, where N_HF is the number of points for lateral control of the fault; otherwise, save the trace numbers and times from LF + N_HF + 1 to NTR and end the section loop. <​​​​​​​​​​​​​For the new automatic tracking to obtain the maximum amplitude along the layer slice, from the structurally favorable area, the most favorable area in this area - the Bao 1 well area has a very strong cap rock amplitude, and the cap rock in the northwest of the work area is also very good; at the same time, we can see that the T40 strong shielding is not constant in amplitude, but changes, which provides a basis for the next step of strong shielding removal. Figure 13 The new T41 layer time slice is obtained by automatic tracking. From the perspective of structural favorable areas, the most favorable area in this area is the Bao 1 well area.

[0056] Using the layer information obtained by the present invention, the maximum amplitude information along the layer of T41 is extracted, and the results are as follows: Figure 14 It can be seen that the amplitude of the Bao 1 well area is relatively strong, which is a favorable area; the amplitude of T41 in the west is generally strong, but this is caused by the presence of ash, and the 1st sand group is not favorable; judging from the amplitude information, the Bao 5 side well is also not favorable.

[0057] In summary, the present invention can provide more accurate basic horizon information for the exploration and development of thin interbedded oil and gas reservoirs, and can be directly used for structural mapping, strong shielding removal, attribute extraction and reservoir prediction.

[0058] The present invention relates to an automatic generation method of seismic horizons by first searching for and moving the peak value and then tracking the breakpoints in space. The method comprises the following steps: determining the polarity of the target horizon according to well seismic calibration, clarifying the peak characteristics of the target horizon; reading reference horizon data including line number, channel number, and horizon time, and setting the search window size; using the reference horizon as a constraint to target the peak or trough layer of the three-dimensional data volume; searching for the breakpoints contained in the stratum one by one according to the continuity characteristics of the stratum and the characteristics of the fault, leaving the data of the left and right breakpoints blank, and saving the other search results, thereby obtaining the target horizon for achieving peak relocation and breakpoint characterization; finally, manually browsing, checking, and correcting the horizon automatically extracted by the computer, and obtaining detailed horizon information. The horizons developed and generated by the present invention have the characteristics of high accuracy, fast search speed, and strong adaptability to geological conditions, and can provide basic data for subsequent seismic attribute extraction, reservoir description, and prediction.

[0059] On the second aspect, the present application provides a seismic layer generation device that first searches for and moves the peak value and then tracks the breakpoint space. The device is used to: determine the polarity of the target layer according to the well seismic calibration, and clearly search for the peak characteristics of the target layer; read the reference layer data containing the line number, channel number, and layer time, and set the search time window range size; for the three-dimensional data body, use the reference layer as a constraint to search for the peak or trough target layer; according to the continuity characteristics of the formation and the characteristics of the fault, search the breakpoints contained in the formation one by one, leave the data of the left breakpoint and the right breakpoint blank, and save other search results, so as to obtain the target layer that realizes peak movement and breakpoint characterization.

[0060] On the third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of a seismic layer generation method of first searching for and moving the peak value and then spatially tracking the breakpoint.

[0061] In a fourth aspect, the present application provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps of a method for generating seismic layers are implemented, which includes first searching for and moving peak values, and then spatially tracking breakpoints.

[0062] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that the present invention is susceptible to various modifications and variations. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A method for generating seismic horizons by first searching for and moving peak values ​​and then spatially tracking breakpoints, characterized in that: Including: S101, determining the polarity of the target horizon according to well-seismic calibration, and clarifying the peak characteristics of the target horizon for search; S102, reading reference horizon data including line number, trace number, and horizon time, and setting the size of the search time window range; S103, constraining the 3D data volume by the reference horizon to search for the target horizon by peak or trough; S104, according to the continuous characteristics of the formation and the characteristics of the fault, searching for the breakpoints contained in the formation one by one, filling the data gaps of the left breakpoint and the right breakpoint, and saving other search results, that is, obtaining the target horizon for realizing peak relocation and breakpoint characterization.

2. The seismic horizon generation method of claim 1, wherein the peak value search and relocation is performed first and the breakpoint space tracking is performed later. Step S102 includes: If the strong peak reflection is searched downward, set a time window containing one peak. The meaning of the strong peak reflection is that the amplitude value is positive and the maximum amplitude value is greater than the maximum value of the adjacent peak; If the weak peak reflection under the strong reflection is searched downward, first set a starting time, and then set a time window containing the weak peak. The meaning of the weak peak reflection is that the amplitude value is positive but the maximum amplitude value is less than the preset threshold.

3. The seismic horizon generation method of claim 2, wherein the peak value search and relocation are performed first and the breakpoint space tracking is performed later, Step S103 includes: For strong reflections, search according to the maximum value search algorithm; For weak peaks, search according to the maximum value algorithm.

4. The seismic horizon generation method of first searching for and moving the peak value and then tracking the breakpoint space according to any one of claims 1 to 3, characterized in that: In step S103, according to the seismic facies characteristics, the horizon search relocation is carried out in the following two ways: Search relocation of the strong reflection layer. The meaning of the strong reflection layer is that the amplitude value is positive and the maximum amplitude value is greater than the maximum value of the adjacent peak: Under the constraint of the reference horizon T0(i,j,k), set the initial value of the maximum value Xmax to 0, and judge whether X(i,j,k)>Xmax holds within the search time window; X(i,j,k) is the 3D data volume, Xmax is the maximum peak value to be searched, and i, j, k are the line number, trace number, and time sequence number; after the search is completed, record the horizon time T1=T0(i,j,k)+(k-1)*dt, where dt is the sampling interval, that is, the relocation from the reference horizon to the strong peak is realized; Search relocation of the weak reflection layer. The meaning of the weak reflection layer is that the amplitude value is positive but the maximum amplitude value is less than the preset threshold: Under the constraint of the reference horizon T0(i,j,k), starting from the search point K0, judge that three conditions are simultaneously satisfied within the search time window: one is X(i,j,k)>0, which is agreed to be the main lobe of the peak; the second is X(i,j,k)>X(i,j,k-1), which is larger than the value of the adjacent point on the left; the third is X(i,j,k)>=X(i,j,k+1), which is not less than the value of the adjacent point on the right. In actual tests, it is found that there is a peak truncation phenomenon in the seismic data, and it will be the same as the value of the adjacent point.

5. The seismic horizon generation method of first searching for and moving the peak value and then tracking the breakpoint space according to any one of claims 1 to 3, characterized in that: In step S104, the method for determining the fault boundary, that is, the left and right breakpoint tracking, includes: Judgment of the left boundary: j = ISTR, NTR - 1, where NTR represents the number of traces of a line, and "j = ISTR, NTR - 1" means that the value of j ranges from ISTR to NTR - 1. If |T j+1 - T j |< dt_LV_max, save the trace number j and time T j , where dt_LV_max is set as the maximum undulation time of the adjacent traces on the left side of the fault, and the initial value of ISTR is set to 1; otherwise, record the left boundary of the fault as LF = j and start the judgment of the right boundary; Right boundary judgment: j = LF + N_HC, LF + 1, "j = LF + N_HC, LF + 1" means that j takes values ​​from LF + N_HC to LF + 1, and searches in reverse. If | T j -T j-1 |>dt_RV_max, dt_RV_max is set to the maximum fluctuation time of the adjacent trace on the right side of the fault, and the right boundary of the fault RF=j+1 is saved. The small loop records the trace number l and time T of this segment l , l=RF,LF+N_HC, "RF,LF+N_HC" represents the range of the recording channel number, N_HC is the number of lateral continuous control points of the formation; otherwise, the search continues; the small cycle is the cycle for determining a single fault; After a fault is processed, if LF+N_HF<NTR-N_HC, update the starting trace number ISTR=LF+N_HC+1 for the new fault identification loop, where N_HF is the number of lateral control points of the fault; otherwise, save the trace numbers and times from LF+N_HF+1 to NTR and end the section loop.

Images