A multi-curve sublayer marker automatic construction method for geosteering

Abstract

The application discloses a kind of multilog small layer mark automatic construction methods for horizontal well geosteering.The method comprises: standardization processing is carried out to the multiple well logging curves of adjacent well, and the dimensional difference between different curves is eliminated;Difference calculation is carried out to the well logging curves after standardization;Based on difference intensity sorting and trend consistency determination model, trend consistency function is constructed and combined with threshold determination;At the same time, the stability and rationality of mark distribution are improved through interval rejection mechanism and minimum spacing constraint;The mark points identified by different well logging curves are subjected to spatial neighborhood fusion, merging sorting and deduplication processing, to form a unified mark set;For each mark point, a structured data unit is constructed, and a mark sequence is formed in the order of vertical depth, to realize the computability of the mark, and provide basic data for subsequent stratigraphic correlation, horizon matching and geosteering automation.The application can significantly improve the automation degree and precision of geosteering calculation.

Description

Technical Field

[0001] This invention belongs to the field of petroleum exploration and development technology, specifically, it relates to an automatic construction method for multi-curve sublayer markers for geological guidance. Background Technology

[0002] Horizontal well geological steering technology plays a crucial role in oil and gas field development. Its core objective is to improve reservoir penetration rate and single-well productivity by real-time identification of the wellbore's position within the formation and controlling its trajectory to extend along the optimal reservoir location. During geological steering, sub-layer markers, as important characteristic points reflecting changes in formation structure, are key foundational data for stratigraphic correlation, stratigraphic location determination, and trajectory adjustment.

[0003] Currently, the acquisition of sub-layer markers mainly relies on manual interpretation or semi-automatic processing methods, and can be broadly categorized into the following two technical approaches:

[0004] The first category is manual interpretation methods based on single-curve characteristics. This method typically uses single logging curves such as gamma curves and resistivity curves as a basis, manually observing changes in curve morphology to identify peaks, troughs, or inflection points as markers for smaller formations. While simple to implement, this method has significant shortcomings: firstly, different interpreters may have varying understandings of the curve characteristics, leading to inconsistent marker identification results; secondly, in complex formations or situations with high curve noise, manual judgment is prone to misjudgment or omission, resulting in poor stability.

[0005] The second category comprises multi-curve-assisted interpretation methods and interactive interpretation systems. This type of technology has already seen some application both domestically and internationally. For example, geological guidance software systems from companies such as Schlumberger and Baker Hughes integrate multiple well logging curves and combine them with a visual interface, allowing interpreters to compare and analyze the multiple curves and manually select representative feature points as sub-layer markers. In related domestic research and engineering practice, multi-curve joint analysis methods are also gradually being introduced to improve the reliability of marker identification.

[0006] Although the multi-curve method has improved the accuracy of marker identification to some extent, the following problems still exist: First, there is a lack of a unified quantitative fusion mechanism among the multiple curves, and the weights and roles of different curves mainly rely on human experience to determine; second, the marker identification process is still mainly based on human participation, with low automation, making it difficult to meet the needs of real-time geological guidance; third, markers usually exist in the form of discrete points, lacking a unified data structure, making it difficult to use directly for subsequent automatic matching and stratigraphic correlation calculations; fourth, when the curves are similar in shape or have complex variations, manual identification is prone to producing unstable results, affecting the accuracy of guidance decisions.

[0007] In summary, existing technologies for identifying sub-layer markers generally suffer from problems such as heavy reliance on human experience, low automation, insufficient multi-curve fusion mechanisms, and inconsistent marker expression forms, making it difficult to meet the requirements of high precision, high consistency, and computability in geological guidance processes.

[0008] Based on the above problems, it is necessary to propose a method that can uniformly process multiple logging curves, automatically identify sub-layer markers, and construct a structured data model, so as to achieve stable extraction and standardized expression of sub-layer markers, and provide a reliable data foundation for marker matching and stratigraphic determination in subsequent geological steering. Summary of the Invention

[0009] The purpose of this invention is to provide an automatic construction method for multi-curve sub-layer markers for geological guidance, which mainly solves the problems of existing technologies in the sub-layer marker identification process, such as strong reliance on human experience, low degree of automation, insufficient multi-curve fusion mechanism, and inconsistent marker expression forms.

[0010] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0011] An automatic construction method for multi-curve sublayer markers for geological steering, which uses a set of logging curves from adjacent wells as input data, includes the following steps:

[0012] S1, standardize multiple logging curves within the target analysis layer to eliminate dimensional differences between different logging curves;

[0013] S2, perform differential calculations on the standardized logging curves to obtain differential sequences that characterize the local variation features of the curves, and calculate the differential intensity of each sampling point;

[0014] S3, candidate marker points are identified based on differential intensity ranking and trend consistency determination;

[0015] S4, the marker points identified from different logging curves are fused, including marker set construction, marker set merging, vertical depth sorting, and marker deduplication based on depth threshold, so as to form a unified marker sequence;

[0016] S5. Construct a structured marker data model based on the unified marker sequence, so that each marker includes marker number, top of the stratum, bottom of the stratum, marker vertical depth, marker shape, distance from top of the stratum, distance from bottom of the stratum, corresponding curve value, and marker deviation parameter.

[0017] S6. The structured markers are sorted according to their vertical depth to form a sequence of small-layer markers with stratigraphic sequence constraints, which is used for stratigraphic identification and marker matching calculation in geological guidance.

[0018] Furthermore, in step S1, the standardization process involves normalizing the mean and standard deviation of the logging curves to ensure that each logging curve has a uniform data scale within the analysis interval.

[0019] Furthermore, in step S2, a difference sequence is obtained by calculating the change between the logging curve values ​​of adjacent sampling points, which is used to characterize the change trend and intensity of the logging curve in the vertical direction.

[0020] Furthermore, the specific process of step S3 is as follows:

[0021] S31, sort the sampling points according to the differential intensity of each sampling point, and prioritize the processing of sampling points whose change amplitude exceeds the preset value;

[0022] S32, calculate the consistency of the difference sign within the neighborhood of the candidate sampling point to determine whether the curve change trend near the sampling point is stable;

[0023] S33, when a candidate sampling point satisfies the trend consistency parameter being greater than a preset threshold, the sampling point is determined as a candidate marker point;

[0024] S34. Starting from the candidate marker point, search along the direction of the curve change trend to determine the position where the trend reverses, thereby determining the corresponding marker interval;

[0025] S35, perform interval elimination processing on the marker interval, and delete other candidate marker points within the interval from the candidate set to avoid duplicate identification of markers in the same change area;

[0026] S36, apply a minimum spacing constraint to the identified marker points. When the vertical distance between the newly identified marker point and the existing marker point is less than the preset minimum spacing threshold, discard the current marker point.

[0027] Further, in step S32, by statistically analyzing the consistency ratio of the difference signs within the neighborhood of the candidate sampling points, a trend consistency parameter is calculated using the trend consistency function to determine whether the curve's trend is stable; wherein, the expression for the trend consistency function is:

[0028]

[0029] in: This is an indicator function; it is 1 if the condition is met, and 0 otherwise. The local window length is the neighborhood window length. Trend sign function:

[0030]

[0031] In the formula, This represents the curve difference value.

[0032] Furthermore, the specific steps of step S4 are as follows:

[0033] S41, Construct a set of markers. Suppose there are K logging curves in total, and the set of marker points identified for each curve is as follows:

[0034]

[0035] The set of markers for each curve is sorted in ascending order of vertical depth. Indicates the first The first curve obtained by curve recognition One flag bit; Indicates the first A set of symbols for a curve;

[0036] S42, Merge the marker set: Merge the marker points of all logging curves to obtain a unified candidate marker set:

[0037]

[0038] S43, Unified sorting: Sort the merged flag set in ascending order according to vertical depth to obtain an ordered flag sequence:

[0039]

[0040] S44, Multi-Curve Marker Deduplication, addresses the issue of duplicate markers arising from different logging curves at the same geological location by introducing a depth resolution threshold. The sorted flag sequences are then fused.

[0041] When any two flags satisfy:

[0042]

[0043] Therefore, it is assumed that both correspond to the same geological marker, and only one marker point is retained; where: The threshold for marking is used to control the spatial consistency of markings between different logging curves;

[0044] S45, the fusion result, after merging, sorting, and deduplication, yields the final unified identifier sequence:

[0045]

[0046] This marker sequence serves as input data for subsequent geological guidance calculations and marker matching.

[0047] Further, in step S5, the flag structure in the structured flag data model is defined:

[0048]

[0049] In the formula, Number the mark; The name of the curve; The name of the top of the target segment interval; The name of the bottom of the target segment interval; To indicate vertical depth; Indicates the shape of the curve; Indicates the distance of the marker from the top of the floor it is located on; To indicate the distance from the bottom of the current floor; This represents the amplitude characteristics of a layer segment, specifically the difference between the maximum and minimum values ​​of the curve for that layer segment. This represents the maximum value of the curve in that segment. This is the minimum value of the curve in this segment; For the corresponding curve value; This is the indicator bias.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] (1) This invention uses a differential intensity ranking and trend consistency judgment model to automatically identify the peak and trough positions in well logging curves, completing the screening and verification of candidate marker points without manual intervention. Compared with the existing technology that relies on the experience of interpreters for manual interpretation, this invention avoids the problem of inconsistent identification results caused by different interpreters' different understandings of curve characteristics, eliminates the risk of human misjudgment or omission in complex strata or when the curve is noisy, and makes the small-layer marker identification process objective and repeatable, providing a stable and reliable source of marker data for geological guidance operations.

[0052] (2) This invention constructs a process for merging multi-curve marker sets, sorting by vertical depth, and deduplicating markers based on depth resolution thresholds. This process enables spatial neighborhood fusion of marker points identified from different logging curves, effectively solving the problems of marker duplication and inconsistent spatial positions among multiple curves. Compared with the existing technology that relies on manual experience to determine the weights of multiple curves and lacks a quantitative fusion mechanism, this invention achieves automatic alignment and unified output of multi-curve markers, avoiding subjective bias in manual comparison and analysis, and significantly improving the consistency and reliability of marker recognition results across multiple curve dimensions.

[0053] (3) This invention constructs a structured data unit for each marker, including the marker number, the name of the top and bottom of the stratum to which it belongs, the vertical depth location, the distance from the top and bottom of the stratum, the amplitude characteristics of the stratum segment, the curve response value, and the marker deviation parameters, and forms a marker sequence in order of vertical depth. Compared with the shortcomings of the prior art in which markers exist in the form of discrete points and lack a unified data structure, this invention elevates the small-layer markers from simple location points to computable data units with multi-attribute information, which can directly participate in subsequent curve matching, stratum identification, and guidance decision calculation, providing standardized basic data support for the automation of geological guidance.

[0054] (4) This invention automatically eliminates duplicate candidate markers within the same changing area during the identification process through an interval elimination mechanism and minimum spacing constraints, and imposes constraints on the vertical depth distance between markers, effectively avoiding excessive marker density and duplicate identification problems. Compared with the existing technology where manual identification is prone to unstable results when curves are similar or complex, this invention ensures a reasonable distribution of markers in the vertical depth direction, giving the identification results better geological significance and spatial consistency, thereby improving the accuracy and stability of marker matching and stratigraphic determination during geological guidance. Attached Figure Description

[0055] Figure 1 This is a schematic diagram of the process structure of the present invention.

[0056] Figure 2 The original logging curves (GR, Mg, Ga) of Shale 2 well in the Long-13-T to Baota Formation-T layer interval in this embodiment of the invention are shown.

[0057] Figure 3 This is a schematic diagram of the standardized processing result of the original well logging curves in an embodiment of the present invention.

[0058] Figure 4 This is a schematic diagram of the difference calculation results (dashed line) of the standardized curve in an embodiment of the present invention.

[0059] Figure 5 This is a schematic diagram of the candidate marker point results (dashed line) obtained by the standardized curve based on differential intensity sorting in an embodiment of the present invention.

[0060] Figure 6 This is a schematic diagram of the marker recognition result determined by the standardized curve based on trend consistency judgment in an embodiment of the present invention.

[0061] Figure 7 This is a schematic diagram of the final small-layer marker recognition result obtained by multi-curve fusion in an embodiment of the present invention. Detailed Implementation

[0062] The present invention will be further described below with reference to the accompanying drawings and embodiments. The embodiments of the present invention include, but are not limited to, the following embodiments.

[0063] like Figure 1 As shown, this invention discloses an automatic construction method for multi-curve sublayer markers for geological steering, the original data of which is a set of logging curves from adjacent wells:

[0064]

[0065] in, For the first well logging curves, It is vertical depth.

[0066] Specifically, the following steps are included:

[0067] S1, within the target analysis interval, multiple logging curves are standardized to eliminate dimensional differences between different logging curves; the standardization process involves normalizing the mean and standard deviation of the logging curves to ensure a uniform data scale for each logging curve within the analysis interval. Specifically, within the target analysis interval:

[0068]

[0069] In the formula, Indicates the minimum vertical depth. Indicates the maximum vertical depth.

[0070] Each logging curve is standardized:

[0071]

[0072] in:

[0073]

[0074]

[0075] This represents the mean of the curve over the analysis interval. Standard deviation This represents the number of sampling points within the interval.

[0076] S2, perform differential calculations on the standardized logging curves to obtain a differential sequence characterizing the local variations of the curves, and calculate the differential intensity at each sampling point; specifically, obtain the differential sequence by calculating the variation between logging curve values ​​at adjacent sampling points, which is used to characterize the variation trend and intensity of the logging curves in the vertical direction. The differential calculation expression is as follows:

[0077]

[0078] in: The difference value, For extremely small positive numbers (to prevent the denominator from being 0, the range of values ​​is specified). ).

[0079] Define the differential strength:

[0080] .

[0081] S3 identifies candidate marker points based on differential intensity ranking and trend consistency judgment. First, locations with large fluctuations in the curve are selected as initial candidate points (not peaks or troughs). Then, the trend consistency within their neighborhood is used for verification to determine whether the change has a continuous and stable direction, thereby eliminating local fluctuation points caused by noise. Finally, geologically significant stable marker candidate points (peaks or troughs) are obtained. Specifically, this includes:

[0082] Sort by difference intensity, prioritizing the most significant change point (maximum difference first).

[0083] Within interval Z, sort all points according to their differential intensity:

[0084] .

[0085] The trend consistency determination function is used to determine whether the trend of a candidate point is stable in its neighborhood, thereby filtering out noise points and retaining the real formation response points (marker points).

[0086] Define the trend sign function:

[0087]

[0088] Define a local window of length L:

[0089]

[0090] Define the trend consistency function:

[0091]

[0092] in: It is an indicator function (1 if satisfied, 0 otherwise).

[0093] Peak and trough determination: Identify candidate markers that conform to trend changes, combine the trend consistency (T value) of their preceding and following neighborhood intervals, and determine whether the point is a valid marker by evaluating the trend stability of the local interval in which the point is located.

[0094] Set threshold:

[0095]

[0096] in, This represents the threshold range for determining trend consistency, used to control whether candidate points are considered stable markers. A larger value indicates stricter screening, while a smaller value indicates higher tolerance. However, at least 60% of the local points must have a trend consistent with the current point for the trend change of that point to meet the conditions for a candidate marker (i.e., sufficiently high trend consistency). Furthermore, if... Then it is determined to be an upward trend, if If so, it is determined to be a downward trend.

[0097] Marker interval positioning: After confirming that the current point is in a stable trend, use this point as the starting point and search forward and backward along the trend direction to find the position where the trend reverses, thereby determining the complete "marker interval" or marker boundary.

[0098] Define trend reversal points:

[0099]

[0100] in, This represents the continuous trend interval for searching forward or backward from the current point. The point where the change is most stable (the extreme point) within the trend range is used as a marker.

[0101] Interval elimination mechanism: Delete other candidate flags in the flag interval to avoid duplicate identification.

[0102] Define the flag range:

[0103]

[0104] Then remove it from the candidate set of flags:

[0105]

[0106] Minimum Spacing Constraint: By setting a minimum spacing threshold, the distance between newly identified marker points and existing marker points is constrained. When the distance between the two is less than the threshold, the current marker point is discarded, thereby ensuring a reasonable spatial distribution of marker points and improving the stability and geological significance of marker identification results.

[0107]

[0108] in, This is the minimum spacing parameter. Its value is determined based on the sampling interval.

[0109] .

[0110] S4 involves fusing marker points identified from different well logging curves, including marker set construction, marker set merging, vertical depth sorting, and marker deduplication based on depth thresholds, thereby forming a unified marker sequence; specifically including:

[0111] Construct a set of markers. Suppose there are K logging curves in total, and the set of marker points identified for each curve is as follows:

[0112]

[0113] The set of markers for each curve is sorted in ascending order of vertical depth. Indicates the first The first curve obtained by curve recognition One flag bit; Indicates the first A set of symbols for a curve;

[0114] Merge the marker set by combining the marker points from all well logging curves to obtain a unified candidate marker set:

[0115]

[0116] The merged set of flags is sorted in ascending order according to its vertical depth to obtain an ordered sequence of flags.

[0117]

[0118] To address the issue of duplicate markers arising from different logging curves at the same geological location, a depth resolution threshold is introduced. The sorted flag sequences are then fused.

[0119] When any two flags satisfy:

[0120]

[0121] Therefore, it is assumed that both correspond to the same geological marker, and only one marker point is retained; where: The threshold for marking is used to control the spatial consistency of markings between different logging curves;

[0122] After merging, sorting, and deduplication, the fusion results yield the final unified identifier sequence:

[0123]

[0124] This marker sequence serves as input data for subsequent geological guidance calculations and marker matching.

[0125] S5. Based on the unified marker sequence, a structured marker data model is constructed, whereby each marker includes a marker number, its corresponding stratum, marker vertical depth, marker shape, distance from the top of the stratum, distance from the bottom of the stratum, corresponding curve value, and marker deviation parameters. The structured marker data model elevates markers from simple location points to computational units with multiple attributes. This structured marker model can directly participate in subsequent curve matching, stratigraphic identification, and guidance decision calculations, forming a crucial foundation for achieving automated geological guidance. First, the marker structure in the structured marker data model is defined:

[0126]

[0127] In the formula, Number the mark; The name of the curve; The name of the top of the target segment interval; The name of the bottom of the target segment interval; To indicate vertical depth; This indicates the shape of the curve (peak is 1, trough is -1). Indicates the distance of the marker from the top of the floor it is located on; To indicate the distance from the bottom of the current floor; This represents the amplitude characteristics of a layer segment, specifically the difference between the maximum and minimum values ​​of the curve for that layer segment. This represents the maximum value of the curve in that segment. This is the minimum value of the curve in this segment; For the corresponding curve value; This is the indicator bias.

[0128] To statistically analyze the difference between the maximum and minimum values ​​of logging curves within the target layer of the reference well, and to describe the range of logging response variation in that layer, characteristic constraint information is provided for subsequent geological steering calculations. It is not used as a fixed threshold, but as a correction basis for subsequent marker matching. Its main function is to record the deviation value of the matched markers during the marker matching process in geological steering, and to dynamically correct the marker values ​​of subsequent actual drilling curves in order to improve the accuracy of the next marker matching.

[0129] S6. The structured markers are sorted according to their vertical depth to form a sequence of small-layer markers with stratigraphic sequence constraints, which is used for stratigraphic identification and marker matching calculation in geological steering. By sorting all markers according to their vertical depth from smallest to largest, a marker sequence with a chronological order is constructed, so that each marker not only has its own attributes, but also has spatial relationships with adjacent markers.

[0130]

[0131]

[0132] This marker sequence can be used for subsequent stratigraphic correlation, stratigraphic tracing, and steering path adjustment, enabling continuous geological interpretation based on sequence constraints and improving the stability and consistency of the steering process.

[0133] This embodiment also automatically identifies and constructs layer markers for adjacent well logging data, and analyzes and verifies the identification results.

[0134] Figures 2-7 This paper presents the automatic construction effect of sub-layer markers in Shale 2 well within the Long-13-T to Baota Formation-T stratigraphic interval. The data used is logging curve data from a reference well. The selected stratigraphic interval is Long-13-T to Baota Formation-T. This interval exhibits significant lithological variations, complex sub-layer structures, and frequent curve changes, making it difficult to reliably identify sub-layer markers using traditional manual interpretation methods. To verify the effectiveness of the method, GR, Mg, and Ga curves were selected as input curves for automatic marker identification.

[0135] Figure 2 The original logging curves of Shale 2 in the Long-13-T to Baota Formation-T layer interval are shown, including three curves: GR, Mg and Ga. Each curve shows obvious fluctuation characteristics in different layers, providing basic data for the identification of small layer markers.

[0136] Figure 3 To standardize the results of the original logging curves, the different curves were normalized to eliminate the dimensional differences between them, making the multi-curve data comparable and providing a unified data foundation for subsequent differential calculations and marker identification.

[0137] Figure 4 The difference calculation results for the standardized curve are obtained by performing difference operations on the curve to obtain a difference sequence reflecting the intensity of curve changes. The larger the difference value, the more significant the curve change, providing a basis for the subsequent identification of candidate marker points. The dashed lines in the figure represent the difference value of each point on the curve.

[0138] Figure 5 The image shows the candidate marker points (dashed line) obtained based on differential intensity ranking. By globally ranking the differential intensities, locations with larger changes in intensity are prioritized as candidate marker points, and the corresponding peaks or troughs are marked on the original curve.

[0139] Figure 6 The results of marker identification are based on trend consistency judgment. Centered on candidate marker points, the reversal points of the curve's trend are searched along the depth direction to determine the extreme points before and after the marker interval, thus forming geologically significant small-layer marker intervals. After interval elimination and minimum spacing constraint processing, duplicate marker identification is effectively avoided, improving the stability of marker distribution.

[0140] Figure 7 This is the final sub-layer marker recognition result obtained by multi-curve fusion. By performing spatial neighborhood fusion on the markers identified by the three curves GR, Ga, and Mg, a unified sub-layer marker position is obtained, and a structured marker sequence is formed according to depth order.

[0141] Figure 7 This invention utilizes the method of the present invention to identify markers on three logging curves (GR, Ga, and Mg) within the reference well-Long-13-T to Baota Formation-T layer interval, and obtains the final marker identification result after multi-curve fusion. As can be seen from the figure, the marker points identified by each logging curve have a good correspondence in the vertical depth direction, and after fusion processing, they form a unified sub-layer marker location.

[0142] To facilitate subsequent geological guidance calculations, this invention further constructs the identified sublayer markers into structured data units. Table 1 provides examples of some of the calculated sublayer marker data structures.

[0143] Table 1. Examples of partial layer flag data structures obtained by the present invention.

[0144]

[0145] The above embodiments powerfully demonstrate that this embodiment achieves automatic identification of sub-layer markers through differential feature analysis, differential intensity ranking, and trend consistency judgment model, avoiding the instability problems caused by the reliance on experience in traditional manual interpretation methods; by interval elimination and minimum spacing constraints, the stability and reliability of marker identification are improved; by multi-curve fusion and structured marker model construction, sub-layer markers are transformed from discrete points into computable data units, providing a reliable data foundation for marker matching and stratigraphic determination in subsequent stratigraphic correlation and geological steering, thereby significantly improving the automation and accuracy of geological steering calculation.

[0146] The above embodiments are merely one of the preferred embodiments of the present invention and should not be used to limit the scope of protection of the present invention. Any modifications or refinements made to the main design concept and spirit of the present invention that are not of substantial significance, but solve the same technical problem as the present invention, should be included within the scope of protection of the present invention.

Claims

1. A method for automatically constructing multi-curve sub-layer markers for geological guidance, characterized in that, This method uses a set of logging curves from adjacent wells as input data and includes the following steps: S1, standardize multiple logging curves within the target analysis layer to eliminate dimensional differences between different logging curves; S2, perform differential calculations on the standardized logging curves to obtain a differential sequence characterizing the local variation features of the curves, and calculate the differential intensity of each sampling point; S3, candidate marker points are identified based on differential intensity ranking and trend consistency determination; S4, the marker points identified from different logging curves are fused, including marker set construction, marker set merging, vertical depth sorting, and marker deduplication based on depth threshold, so as to form a unified marker sequence; S5. Construct a structured marker data model based on the unified marker sequence, so that each marker includes marker number, top of the stratum, bottom of the stratum, marker vertical depth, marker shape, distance from top of the stratum, distance from bottom of the stratum, corresponding curve value, and marker deviation parameter. S6. The structured markers are sorted according to their vertical depth to form a sequence of small-layer markers with stratigraphic sequence constraints, which is used for stratigraphic identification and marker matching calculation in geological guidance.

2. The method for automatically constructing multi-curve sub-layer markers for geological guidance according to claim 1, characterized in that, In step S1, the standardization process involves normalizing the mean and standard deviation of the logging curves to ensure that each logging curve has a uniform data scale within the analysis interval.

3. The method for automatically constructing multi-curve sub-layer markers for geological guidance according to claim 2, characterized in that, In step S2, a difference sequence is obtained by calculating the change between the logging curve values ​​of adjacent sampling points, which is used to characterize the change trend and intensity of the logging curve in the vertical direction.

4. The method for automatically constructing multi-curve sub-layer markers for geological guidance according to claim 3, characterized in that, The specific process of step S3 is as follows: S31, sort the sampling points according to the differential intensity of each sampling point, and prioritize the processing of sampling points whose change amplitude exceeds the preset value; S32, calculate the consistency of the difference sign within the neighborhood of the candidate sampling point to determine whether the curve change trend near the sampling point is stable; S33, when a candidate sampling point satisfies the trend consistency parameter being greater than a preset threshold, the sampling point is determined as a candidate marker point; S34. Starting from the candidate marker point, search along the direction of the curve change trend to determine the position where the trend reverses, thereby determining the corresponding marker interval; S35, perform interval elimination processing on the marker interval, and delete other candidate marker points within the interval from the candidate set to avoid duplicate identification of markers in the same change area; S36, apply a minimum spacing constraint to the identified marker points. When the vertical distance between the newly identified marker point and the existing marker point is less than the preset minimum spacing threshold, discard the current marker point.

5. The method for automatically constructing multi-curve sub-layer markers for geological guidance according to claim 4, characterized in that, In step S32, the trend consistency parameter is calculated using the trend consistency function by statistically analyzing the consistency ratio of the difference signs within the neighborhood of the candidate sampling points. This parameter is used to determine whether the curve's trend is stable. The expression for the trend consistency function is: in: This is an indicator function; it is 1 if the condition is met, and 0 otherwise. The local window length is the neighborhood window length. Trend sign function: In the formula, This represents the curve difference value.

6. The method for automatically constructing multi-curve sub-layer markers for geological guidance according to claim 5, characterized in that, The specific steps of step S4 are as follows: S41, Construct a set of markers. Suppose there are K logging curves in total, and the set of marker points identified for each curve is as follows: The set of markers for each curve is sorted in ascending order of vertical depth. Indicates the first The first curve obtained by curve recognition One flag bit; Indicates the first A set of symbols for a curve; S42, Merge the marker set: Merge the marker points of all logging curves to obtain a unified candidate marker set: S43, Unified sorting: Sort the merged flag set in ascending order according to vertical depth to obtain an ordered flag sequence: S44, Multi-Curve Marker Deduplication, addresses the issue of duplicate markers arising from different logging curves at the same geological location by introducing a depth resolution threshold. The sorted flag sequences are then fused. When any two flags satisfy: Therefore, it is assumed that both correspond to the same geological marker, and only one marker point is retained; where: The threshold for marking is used to control the spatial consistency of markings between different logging curves; S45, the fusion result, after merging, sorting, and deduplication, yields the final unified identifier sequence: This marker sequence serves as input data for subsequent geological guidance calculations and marker matching.

7. The method for automatically constructing multi-curve sub-layer markers for geological guidance according to claim 6, characterized in that, In step S5, the flag structure in the structured flag data model is defined: In the formula, Number the mark; The name of the curve; The name of the top of the target segment interval; The name of the bottom of the target segment interval; To indicate vertical depth; Indicates the shape of the curve; Indicates the distance of the marker from the top of the floor it is located on; To indicate the distance from the bottom of the current floor; This represents the amplitude characteristics of a layer segment, specifically the difference between the maximum and minimum values ​​of the curve for that layer segment. This represents the maximum value of the curve in that segment. This is the minimum value of the curve in this segment; For the corresponding curve value; This is the indicator of bias.

Images