A geological data quality evaluation method and system for geological division

Abstract

The application discloses a geological data quality evaluation method and system for geological division, and relates to the technical field of data quality evaluation. The method comprises the following steps: S1: acquiring multi-source geological data of a region to be evaluated and performing pretreatment; S2: identifying a key evaluation region in the multi-source geological data; S3: extracting multi-dimensional quality indexes of the multi-source geological data; S4: performing backfeeding optimization on the multi-dimensional quality indexes in combination with an AI intelligent interpretation result; S5: calibrating the multi-dimensional quality indexes to generate a comprehensive quality score; and S6: outputting a geological data quality evaluation report. According to the method, a scene priority-data coupling degree double-dimensional dynamic threshold model is constructed, the evaluation threshold can be accurately adjusted according to the core needs of different scenes such as coal seam thickness division and fault boundary identification, the problem that a traditional fixed threshold cannot adapt to the core needs of scenes is solved, and the evaluation standard is highly matched with the core demands of different geological division scenes.

Description

Technical Field

[0001] This invention relates to the field of data quality assessment technology, specifically to a method and system for assessing the quality of geological data used in geological classification. Background Technology

[0002] Geological data quality assessment is a core prerequisite for coal mine geological delineation (such as accurate coal seam thickness delineation and clear fault boundary identification), and the rationality of its evaluation standards directly determines the reliability of geological interpretation results. In coal mine exploration and development practice, the core data quality requirements differ significantly across different geological delineation scenarios. For example, coal seam thickness delineation requires ensuring the depth accuracy of borehole data and the stratigraphic resolution of SGY data volumes to avoid missing thin coal seams or inaccurate calculations of thick coal seam thickness. Fault boundary identification, on the other hand, relies more heavily on the phase continuity of SGY data (SGY data refers to seismic data format conforming to the SEG-Y (Society of Exploration Geophysicists-Y) standard, one of the most commonly used seismic data storage and exchange formats in petroleum exploration and geological research. Its core function is to store underground reflected wave signals acquired during seismic exploration in a standardized structure for subsequent processing, analysis, and interpretation) and the spatial accuracy of roadway line data to accurately define fault strike and displacement.

[0003] However, existing geological data quality assessment methods generally adopt a fixed threshold judgment mode. Because the threshold settings lack scenario adaptability—for example, uniformly setting the signal-to-noise ratio threshold for SGY data to 2.5 dB—fails to consider the higher signal-to-noise ratio requirement (≥3.0 dB) for thin coal seams (thickness <1 m) in coal seam thickness classification scenarios, nor does it take into account the special requirements of phase continuity (≤0.5 ms phase difference) in fault boundary identification scenarios. This leads to a disconnect between the evaluation results and the actual scenario requirements. Furthermore, threshold adjustments lack dynamism; once traditional methods set thresholds for depth error, phase deviation, etc., they are applied to all scenarios for an extended period. In geological regions and classification scenarios, the threshold cannot be dynamically optimized based on changes in geological conditions (such as complex structural areas versus simple coal seam areas) or adjustments to classification objectives (such as shifting from macroscopic coal seam distribution to microscopic fault branch identification). Due to the lack of correlation among thresholds from multiple data sources, for example, the borehole data depth error threshold (fixed at 0.1m) is not linked to the SGY data stratigraphic interpretation deviation threshold. In coal seam thickness classification scenarios, even if the borehole depth error meets the fixed threshold, the insufficient stratigraphic resolution of the SGY data may lead to deviations in the determination of the coal seam roof and floor. The fixed threshold system cannot identify such quality coordination defects across data types.

[0004] The aforementioned fixed threshold defects directly lead to a low degree of matching between the evaluation results and actual application scenarios: In thin coal seam areas, the low signal-to-noise ratio threshold of SGY data causes fuzziness of the stratigraphic phase axis, resulting in excessive errors in coal seam thickness calculation; in areas with dense faults, the excessively wide fixed threshold for phase deviation makes it impossible to identify minute phase anomalies, leading to excessive offset in fault boundary determination; in sparsely distributed edge areas, the fixed borehole depth error threshold cannot adapt to the scarcity of data in these areas, resulting in a significant decrease in the reliability of geological classification results. Summary of the Invention

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for evaluating the quality of geological data for geological delineation, comprising the following steps: S1: Obtain multi-source geological data of the area to be evaluated and perform preprocessing. The multi-source geological data includes at least SGY seismic processing result data body, borehole result data and coal mine roadway line data. During the preprocessing process, a dynamic adaptation deviation threshold based on scene priority and data coupling degree is used to achieve preliminary screening and compensation of data quality. S2: Identify key evaluation areas in the multi-source geological data, wherein the key evaluation areas include at least the data loading edge connection area, the borehole data missing section, and the boundary area of ​​the seismic data coverage number; S3: For the global data and the key evaluation area, extract multi-dimensional quality indicators of the multi-source geological data. The multi-dimensional quality indicators include at least the index parameters used to evaluate the dynamic adaptability of the data, the small phase difference in stratigraphic interpretation, and the accuracy of stratigraphic interpretation. S4: Combine the AI ​​intelligent interpretation results to optimize the multi-dimensional quality indicators, establish the mapping relationship between AI interpretation confidence and data quality defects, and realize the retrospective location of data quality defects; S5: Construct a true value standard based on actual drilling data, calibrate the multi-dimensional quality indicators, and generate a comprehensive quality score; S6: Output a geological data quality assessment report that includes the quality status of key evaluation areas, the location of data quality defects, and optimization suggestions.

[0006] Preferably, step S1, which involves acquiring and preprocessing multi-source geological data of the area to be evaluated, specifically includes: S11: Load the SGY seismic processing results data, borehole data and coal mine roadway line data, and extract the basic characteristic parameters of each data type. The basic characteristic parameters of SGY data include signal-to-noise ratio, sampling interval and phase continuity; the basic characteristic parameters of borehole data include depth error and lithological labeling integrity; and the basic characteristic parameters of roadway line data include spatial location accuracy and extension continuity. S12: Establish a dynamic weight library for geological classification scenarios and data feature parameters. The geological classification scenarios include at least coal seam thickness classification scenarios and fault boundary classification scenarios. Dynamic weights are assigned to each data feature parameter for different scenarios. In the coal seam thickness classification scenario, the weight value of borehole data depth error is higher than that in the fault boundary classification scenario, and the weight value of SGY data phase continuity is higher than that in the fault boundary classification scenario. S13: Based on the priority of geological scene division and the coupling relationship between multi-source data features, a matching deviation threshold is dynamically generated. By comparing the actual matching deviation with the threshold, corresponding data fit judgment and compensation processing are performed. Specifically: when the actual matching deviation meets the threshold requirement, it is judged as a qualified fit; when the actual matching deviation exceeds the threshold but is within the preset range, it is compensated and corrected using associated data; when the actual matching deviation exceeds the preset range, it is judged as a misfit and the corresponding data optimization processing mechanism is triggered to obtain usable data.

[0007] Preferably, the identification of key evaluation areas in the multi-source geological data in step S2 specifically includes: S21: Identify the data loading edge connection area, which includes at least the splicing seam of multiple SGY data volumes, the edge blank zone of borehole distribution, the end area of ​​the roadway line extension, and the boundary area of ​​seismic data coverage number. The boundary range of the edge connection area is determined by data coordinate overlap detection, data density gradient analysis, and seismic coverage number distribution calculation. S22: Identify missing sections of borehole data, locate the depth interval with data gaps by detecting the continuity of the borehole depth sequence, and record the valid borehole data above and below the missing section. The valid borehole data includes at least the stratigraphic depth and lithological information. S23: Mark the edge connection area and the missing section of borehole data to generate a key evaluation area distribution map.

[0008] Preferably, step S3, which involves extracting multi-dimensional quality indicators from the multi-source geological data for the global data and the key evaluation area, specifically includes: S311: Extract dynamic data adaptation indicators, and calculate the adaptation score for each data type based on the dynamic weight library and coupling coefficient. Adaptation score = Σ (actual value of data feature parameter / standard value of feature parameter × dynamic weight × K). When the adaptation score is lower than the preset adaptation score threshold, it is determined that the dynamic adaptation indicator is not up to standard. S312: Extract the stratigraphic interpretation of small phase difference indicators, use wavelet transform algorithm to extract the phase features of the stratigraphic phase axis in the SGY data volume, capture the phase deviation within the predetermined small phase difference range, calculate the distribution density and continuous length of the small phase difference, and mark it as a phase abnormal region when the continuous length exceeds the preset sampling point threshold. S313: Extract the stratigraphic interpretation accuracy index, obtain the stratigraphic interpretation depth data on the SGY profile by AI or manual means, calculate the dispersion of the interpretation depth data, dispersion = (maximum interpretation depth - minimum interpretation depth) / average interpretation depth, when the dispersion exceeds the preset dispersion threshold, it is determined that the stratigraphic interpretation accuracy index does not meet the standard.

[0009] Preferably, step S3 further includes a step for extracting quality indicators for missing segments of borehole data, specifically: S321: Based on the effective borehole data above and below the missing section, a theoretical geological model of the missing section is constructed using a cubic polynomial interpolation algorithm. The theoretical geological model includes at least the theoretical stratigraphic depth curve and the theoretical lithological distribution. S322: Extract SGY data for the area corresponding to the missing section of the borehole, perform stratigraphic interpretation on the area, and obtain the actual stratigraphic interpretation results; S323: Calculate the degree of agreement between the actual stratigraphic interpretation results and the theoretical geological model. The degree of agreement = (1 - |actual stratigraphic depth - theoretical stratigraphic depth| / theoretical stratigraphic depth) × 100%. When the degree of agreement is lower than the preset degree of agreement threshold, it is determined that the SGY data quality of the area corresponding to the missing section of the borehole is substandard.

[0010] Preferably, step S4, which combines the AI ​​intelligent interpretation results to optimize the multi-dimensional quality indicators, specifically includes: S41: Obtain AI intelligent interpretation results, which include at least automatic layer tracking results and fault intelligent identification results. Simultaneously record the confidence value of each interpretation result. The confidence value is calculated based on the continuity of seismic phase axes, borehole data consistency, and coupling coefficient. S42: Establish a mapping model between AI interpretation confidence and data quality defects. When the confidence value is lower than the preset confidence threshold, it is determined that there are data quality defects in the corresponding area. When the confidence value is in the first preset confidence interval, it corresponds to data noise interference defects. When the confidence value is in the second preset confidence interval, it corresponds to data sampling accuracy defects. The upper limit of the second preset confidence interval is lower than the lower limit of the first preset confidence interval. S43: Based on the mapping model, backtrack to locate the specific location of data quality defects, trigger the corresponding data preprocessing optimization process according to the defect type, and optimize multi-dimensional quality indicators.

[0011] Preferably, step S5, which involves constructing a truth standard based on actual drilling data and calibrating the multi-dimensional quality indicators, specifically includes: S51: Select the stratigraphic depth data verified in the field from the actual drilling data as the true value standard. The stratigraphic depth data shall include at least the depth of the coal seam roof and floor and the depth of the marker layer. S52: Calculate the depth deviation between the interpreted depth data and the true standard. Depth deviation = |interpreted depth - true depth|. S53: Construct a data quality calibration coefficient based on the depth deviation value. The calibration coefficient = 1 - (depth deviation value / true layer depth). When the calibration coefficient is lower than the preset calibration coefficient threshold, the layer interpretation accuracy index is corrected. The corrected layer interpretation accuracy index = original index × calibration coefficient. S54: Substitute the revised multi-dimensional quality indicators into the comprehensive quality scoring formula. The comprehensive quality score is calculated based on the quality indicators of each dimension and their preset weights. The quality indicators of each dimension include at least the data dynamic adaptation indicator, the layer interpretation micro phase difference indicator, the layer interpretation accuracy indicator, and the borehole missing segment matching indicator.

[0012] Preferably, step S3 further includes a quality index extraction step for the edge connection area, which uses a region type-influence factor dual-dimensional dynamic threshold model to calculate the qualified threshold of the edge connection consistency coefficient, specifically: S331: Calculate the edge connection consistency coefficient of the edge connection area, where: For the splicing seam of SGY data volumes, the consistency coefficient = 1 - (mean phase difference of in-phase axis of adjacent data volumes / dynamic phase difference threshold). For the edge blank zone of borehole distribution, the consistency coefficient = 1 - (the deviation rate of edge borehole depth from seismic horizon / dynamic deviation rate threshold). For the terminal area of ​​the tunnel line extension, the consistency coefficient = 1 - (the offset between the tunnel line end and the geological trend / the dynamic offset threshold). For the boundary area of ​​seismic data coverage, the consistency coefficient = 1 - (the deviation between the measured and designed coverage values ​​ / the dynamic coverage deviation threshold). S332: Construct a dynamic threshold calculation model with two dimensions of region type and impact factor to generate dynamic thresholds for each marginal region, including: determining the core impact factor and weight according to the marginal region type, calculating the actual contribution value of each impact factor, and dynamically generating the threshold of the corresponding region based on the contribution value; setting the dynamic threshold as the qualified threshold of the marginal connection consistency coefficient, and judging the data quality as substandard when the consistency coefficient is lower than the qualified threshold, and analyzing the reasons for the substandard based on the contribution value of the impact factor.

[0013] Preferably, the geological data quality evaluation report output in step S6 specifically includes: S61: Generate a data quality heatmap, using color gradients to represent the distribution of comprehensive quality scores, where the first color indicates a score below the first preset score threshold, the second color indicates a score between the first and second preset score thresholds, and the third color indicates a score above the second preset score threshold. The quality status of key evaluation areas is also marked, and the second preset score threshold is higher than the first preset score threshold. S62: Overlay data quality defect location markers on the heat map, and provide targeted optimization suggestions based on defect causes, dynamic adaptation deviation thresholds, and dynamic thresholds of edge connection areas: If the SGY splice seam quality is substandard, it is recommended to recalibrate the data splicing parameters; if the quality of the blank zone at the borehole edge is substandard, it is recommended to supplement the borehole measurement in the edge area. S63: The optimization suggestions are associated with the preprocessed multi-source geological data to form an iterative quality evaluation closed loop. When the geological data of the region is acquired again, the historical dynamic thresholds and quality evaluation reports are automatically called for comparative analysis, and the dynamic weight library, coupling coefficient and marginal area influence factor weights are updated.

[0014] A geological data quality evaluation system for geological delineation, specifically comprising: The data preprocessing module is used to acquire and preprocess multi-source geological data of the area to be evaluated. The multi-source geological data includes at least SGY seismic processing result data, borehole result data and coal mine roadway line data. During the preprocessing process, a dynamic adaptation deviation threshold based on scene priority and data coupling degree is used to achieve preliminary screening and compensation of data quality. The key area identification module is used to identify key evaluation areas in the multi-source geological data. The key evaluation areas include at least the data loading edge connection area, the borehole data missing section, and the boundary area of ​​the seismic data coverage number. The quality index extraction module is used to extract multi-dimensional quality indicators from the multi-source geological data for the global data and the key evaluation area. The multi-dimensional quality indicators include at least the index parameters used to evaluate the dynamic adaptability of the data, the small phase difference in stratigraphic interpretation, and the accuracy of stratigraphic interpretation. The AI ​​feedback optimization module is used to combine the AI ​​intelligent interpretation results to provide feedback optimization for the multi-dimensional quality indicators, establish a mapping relationship between the confidence of AI interpretation and data quality defects, and realize the retrospective location of data quality defects. The quality calibration and scoring module is used to construct a true value standard based on actual drilling data, calibrate the multi-dimensional quality indicators, and generate a comprehensive quality score. The evaluation report output module is used to output a geological data quality evaluation report that includes the quality status of key evaluation areas, the location of data quality defects, and optimization suggestions.

[0015] This invention discloses a method and system for evaluating the quality of geological data for geological delineation, which has the following beneficial effects: (1) By constructing a dual-dimensional dynamic threshold model of scenario priority and data coupling degree, this invention can accurately adjust the evaluation threshold for the core needs of different scenarios such as coal seam thickness division and fault boundary identification: In the coal seam thickness division scenario, the SGY data signal-to-noise ratio threshold can be dynamically increased to ≥3.0dB according to the high requirements of data accuracy for thin coal seams (thickness <1m), avoiding the omission of thin coal seams or thickness calculation deviation caused by the blurring of the stratum phase axis; In the fault boundary identification scenario, the phase continuity threshold can be strictly controlled to ≤0.5ms to ensure that small phase anomalies are accurately captured, solving the problem that traditional fixed thresholds cannot adapt to the core needs of the scenario, and making the evaluation standard highly matched with the core requirements of different geological division scenarios; (2) This invention breaks through the limitation of using the traditional fixed threshold indefinitely. It can be adjusted according to the differences in geological conditions and the division target, and optimize the evaluation threshold in real time. In view of the strong interference of data in complex structural areas, the stringency of the SGY data phase continuity threshold can be appropriately increased to avoid misjudgment caused by structural interference. In view of the refined requirements of micro fault branch identification, the threshold range of the spatial position accuracy of roadway line data can be narrowed to ensure the accurate definition of fault branch boundaries, effectively solving the defect that the traditional fixed threshold cannot adapt to the dynamic changes of geological conditions and division targets. (3) This invention constructs a multi-source data threshold linkage system by calculating the coupling coefficient between SGY data, borehole data, and roadway line data. It can identify cross-data type quality defects that cannot be detected by the traditional fixed threshold system. For example, in the scenario of coal seam thickness division, the borehole data depth error threshold can be linked with the SGY data layer resolution threshold. If the borehole depth error meets the threshold but the SGY data layer resolution is insufficient, the system can identify the collaborative defect through the coupling coefficient and trigger the compensation mechanism to avoid the deviation in the judgment of the coal seam top and bottom plates caused by the traditional fixed threshold not being associated with SGY data, and improve the reliability of multi-source data collaborative support for geological division. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic flowchart of the method of the present invention; Figure 2 This is a schematic block diagram of the system of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are described clearly and completely. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0020] This invention provides, for example Figure 1-2The method and system for evaluating the quality of geological data for geological delineation, as shown, include the following steps: S1: Acquire multi-source geological data of the area to be evaluated and preprocess it. The multi-source geological data includes at least SGY seismic processing data, borehole data, and coal mine roadway line data. During the preprocessing process, a dynamic adaptation deviation threshold calculated based on scene priority and data coupling degree is used to achieve preliminary screening and compensation of data quality; S2: Identify key evaluation areas in the multi-source geological data. Key evaluation areas include at least the data loading edge connection area, the missing section of borehole data, and the boundary area of ​​seismic data coverage times; S3: Extract multi-dimensional quality indicators from the multi-source geological data for the global data and key evaluation areas. The multi-dimensional quality indicators include at least the indicator parameters used to evaluate the dynamic adaptability of the data, the small phase difference of the stratigraphic interpretation, and the accuracy of the stratigraphic interpretation; S4: Combine the AI ​​intelligent interpretation results to optimize the multi-dimensional quality indicators, establish the mapping relationship between the AI ​​interpretation confidence and the data quality defects, and realize the retrospective location of data quality defects; S5: Construct a true value standard based on the actual borehole drilling data, calibrate the multi-dimensional quality indicators, and generate a comprehensive quality score; S6 This system outputs a geological data quality assessment report that includes the quality status of key evaluation areas, the location of data quality defects, and optimization suggestions. By constructing a complete geological data quality assessment process, it effectively solves the multi-dimensional deficiencies in traditional assessments: Firstly, by acquiring multi-source geological data and conducting preliminary screening with dynamically adapted deviation thresholds, it addresses the problem that traditional fixed-threshold screening cannot adapt to different geological scenarios, improving the accuracy of preliminary data screening. Secondly, by identifying key evaluation areas, it solves the interpretation bias problem caused by traditional methods ignoring quality-sensitive areas, laying the foundation for accurate assessment. Thirdly, by extracting multi-dimensional quality indicators, it addresses the problem of traditional assessment indicators being singular and only focusing on the physical characteristics of data, achieving a comprehensive consideration of data quality. Fourthly, by using AI intelligent interpretation results to feed back optimization and locate quality defects, it addresses the problem of traditional assessments relying on human experience and being inefficient, improving the level of intelligence. Fifthly, by constructing true value standard calibration indicators and generating comprehensive scores using borehole drilling data, it addresses the problem of traditional assessments lacking objective references and insufficient accuracy, ensuring the reliability of assessment results. Finally, by outputting an assessment report containing key information, it addresses the problem of traditional reports having weak decision-making support capabilities, providing complete quality support for geological delineation.

[0021] Specifically, step S1 involves acquiring and preprocessing multi-source geological data of the area to be evaluated, including: S11: Loading the SGY seismic processing data, borehole data, and coal mine roadway line data, and extracting the basic characteristic parameters for each data type. The basic characteristic parameters for SGY data include signal-to-noise ratio, sampling interval, and phase continuity; the basic characteristic parameters for borehole data include depth error and lithological labeling integrity; and the basic characteristic parameters for roadway line data include spatial location accuracy and extension continuity. S12: Establishing a dynamic weight library for geological classification scenarios and data characteristic parameters. The geological classification scenarios must include at least coal seams. For thickness classification scenarios and fault boundary classification scenarios, dynamic weights are assigned to each data feature parameter for different scenarios. Specifically, in the coal seam thickness classification scenario, the weight of borehole data depth error is higher than in the fault boundary classification scenario, while in the fault boundary classification scenario, the weight of SGY data phase continuity is higher than in the coal seam thickness classification scenario. S13: A two-dimensional model of scenario priority and data coupling degree is used to calculate the dynamic adaptation deviation threshold, and data adaptability is judged based on this threshold. The specific steps are: S131: Determine the priority weight of the current geological classification scenario, setting the priority weight of the coal seam thickness classification scenario to W1, and the fault boundary classification scenario... The priority weight for each scenario is W2, and W1 > W2 (W1 + W2 = 1); S132: Calculate the coupling coefficient between features of multi-source data. The coupling coefficient includes the coupling coefficient K1 between the signal-to-noise ratio of SGY data and the depth error of borehole data, and the coupling coefficient K2 between the phase continuity of SGY data and the spatial position accuracy of the roadway line. The coupling coefficient K = (Pearson correlation coefficient of the two data feature parameters × 0.6 + data co-interpretation contribution × 0.4), where the data co-interpretation contribution is obtained through statistical analysis of historical interpretation cases; S133: Dynamically generate the adaptation deviation threshold T, T = (Σ(standard data feature parameters)). S134: Calculate the actual adaptation deviation D between each data type and the current geological classification scenario, D = |(measured value of data feature parameter × dynamic weight) - (standard value of data feature parameter × dynamic weight)| / (standard value of data feature parameter × dynamic weight); S135: Compare the actual adaptation deviation D with the dynamic adaptation deviation threshold T: if D ≤ T, the data is deemed to be adapted; if T < D ≤ 1.2T, a mild compensation mechanism is triggered, and adjacent associated data is used for correction; if D > 1.2T, if data mismatch is determined, a heavy compensation mechanism is triggered, outputting a warning message indicating that the data needs to be re-collected or reprocessed. Preprocessing based on conservative parameters is then performed on the current data to obtain usable multi-source geological data. By refining the multi-source data preprocessing process, the core shortcomings of traditional preprocessing are further addressed: Firstly, by comprehensively extracting the basic feature parameters of multi-source data, the problem of traditional preprocessing extracting only a small number of features and failing to reflect the complete quality of the data is solved, providing a sufficient data foundation for subsequent evaluation. Secondly, by establishing a scenario-based dynamic weight library, the weights of data feature parameters are adjusted for different geological classification scenarios, solving the problem of traditional methods treating data importance indiscriminately across different scenarios, ensuring that weight allocation matches the core needs of the scenario. Thirdly, by calculating the dynamic adaptation deviation threshold through a two-dimensional model and implementing a graded compensation mechanism based on actual deviations, the problem of traditional fixed thresholds failing to adapt to the relationship between scenarios and data, and lacking effective processing methods when data is mismatched, is solved, ensuring that the preprocessed data meets the geological classification requirements and avoiding interruptions in the evaluation process.

[0022] Specifically, step S2, identifying key evaluation areas in multi-source geological data, includes: S21: Identifying data loading edge connection zones, which include at least the splicing seams of multiple SGY data volumes, the edge blank zones of borehole distribution, the end areas of roadway extensions, and the boundary areas of seismic data coverage. The boundary range of the edge connection zones is determined through data coordinate overlap detection, data density gradient analysis, and seismic coverage distribution calculation; S22: Identifying missing borehole data segments, locating depth intervals with data gaps through borehole depth sequence continuity detection, and recording valid borehole data above and below the missing segments. Valid borehole data includes at least stratigraphic depth and lithological information; S23: [Further details about edge connection zones and borehole data are needed for accurate translation.] Missing sections of borehole data are marked to generate a key evaluation area distribution map. By refining the key evaluation area identification method, the accuracy and practicality issues of traditional identification processes are addressed: professional methods are used to accurately locate the boundary ranges of various edge connection areas, solving the problems of low positioning accuracy and easy omissions caused by traditional reliance on manual experience, ensuring that all quality-sensitive areas are included in the evaluation; by detecting missing sections of borehole data and recording associated valid data, the problem of neglecting missing sections and resulting in blank evaluations in these areas is solved, providing data support for subsequent quality evaluation of missing sections; by marking key areas and generating distribution maps, the problem of the lack of visual markers and the difficulty in conducting targeted subsequent evaluations is solved, providing clear area guidance for staff.

[0023] Specifically, step S3 involves extracting multi-dimensional quality indicators from multi-source geological data for global data and key evaluation areas, including: S311: Extracting dynamic adaptation indicators. Based on a dynamic weight library and coupling coefficients, the adaptation score for each data type is calculated. The adaptation score is calculated as Σ(measured value of data feature parameter / standard value of feature parameter × dynamic weight × K). When the adaptation score is lower than a preset adaptation score threshold, the dynamic adaptation indicator is deemed unqualified. S312: Extracting stratigraphic interpretation micro-phase difference indicators. Wavelet transform algorithms are used to extract phase features from stratigraphic phase axes in the SGY data volume, capturing phase deviations within a predetermined micro-phase difference range. The distribution density and continuous length of the micro-phase difference are calculated. When the continuous length exceeds a preset sampling point threshold, it is marked as a phase anomaly area. S313: Extracting stratigraphic interpretation accuracy indicators. The stratigraphic position on the SGY profile is obtained by AI or manual intervention. The layer interpretation depth data is interpreted by calculating its dispersion, calculated as (maximum interpretation depth - minimum interpretation depth) / average interpretation depth. When the dispersion exceeds a preset threshold, the layer interpretation accuracy is deemed substandard. This is achieved by refining multi-dimensional quality indicator extraction methods to address the limitations and inaccuracies of traditional indicator extraction: By combining scene weights and data correlation to calculate the adaptation score, the traditional adaptation indicators, which only consider data characteristics and are disconnected from practical applications, are addressed, making the adaptation evaluation more aligned with the collaborative needs of multi-source data; by using high-precision algorithms to capture minute phase deviations in SGY data and analyze their distribution characteristics, the traditional detection methods are found to lack precision and fail to identify minor anomalies, thus avoiding layer interpretation bias; and by quantifying the dispersion of layer interpretation depth, the traditional methods rely on subjective judgment and lack quantitative standards, enabling objective assessment of layer interpretation accuracy.

[0024] Step S3 further includes a quality index extraction step for the missing borehole data segment, specifically: S321: Based on the valid borehole data above and below the missing segment, a cubic polynomial interpolation algorithm is used to construct a theoretical geological model for the missing segment. The theoretical geological model includes at least the theoretical stratigraphic depth curve and the theoretical lithology distribution; S322: Extract SGY data for the corresponding area of ​​the missing borehole segment, perform stratigraphic interpretation on the area, and obtain the actual stratigraphic interpretation result; S323: Calculate the consistency between the actual stratigraphic interpretation result and the theoretical geological model. Consistency = (1 - |actual stratigraphic depth - theoretical stratigraphic depth| / theoretical stratigraphic depth) × 100%. When the consistency is lower than the preset consistency threshold... When the SGY data in the area corresponding to the missing borehole segment is determined to be of substandard quality, the problem of traditional methods being unable to evaluate the data quality of this area is solved by refining the process of extracting quality indicators for the missing borehole segment: A theoretical geological model is constructed based on the effective data above and below the missing segment, addressing the lack of geological truth references and the inability to measure the accuracy of interpretation, thus providing a benchmark for quality evaluation; SGY data corresponding to the missing segment is extracted and stratigraphic interpretation is conducted to obtain actual geological information, filling the interpretation gaps in this area; and data quality is determined by comparing the actual interpretation results with the theoretical model, solving the problem of traditional methods being unable to evaluate the quality of SGY data in the missing segment, ensuring that the geological delineation of this area has reliable quality support.

[0025] Specifically, step S4, which combines AI intelligent interpretation results to optimize multi-dimensional quality indicators, includes: S41: Obtaining AI intelligent interpretation results, which at least include automatic layer tracking results and fault intelligent identification results. The confidence value of each interpretation result is recorded synchronously. The confidence value is calculated based on the continuity of seismic phase axes, borehole data consistency, and coupling coefficient. S42: Establishing a mapping model between AI interpretation confidence and data quality defects. When the confidence value is lower than a preset confidence threshold, a data quality defect is determined to exist in the corresponding area. Specifically, a confidence value within the first preset confidence interval corresponds to data noise interference defects, and a confidence value within the second preset confidence interval corresponds to data sampling accuracy defects. Furthermore, the upper limit of the second preset confidence interval is lower than the lower limit of the first preset confidence interval. S43: Based on the mapping model... The system traces back to pinpoint the exact location of data quality defects and triggers corresponding data preprocessing optimization processes based on defect type (for noise interference defects, SGY data denoising is performed; for sampling accuracy defects, the data sampling interval is adjusted). It optimizes multi-dimensional quality indicators and addresses the issues of insufficient integration between traditional evaluation and AI, and inefficient defect location and optimization by refining the AI-feedback optimization process. Specifically, it calculates AI interpretation confidence scores that include the correlation between multi-source data, addressing the weak correlation between traditional confidence scores and data quality, making confidence scores more reflective of data quality. Furthermore, it establishes a mapping model between confidence scores and quality defects to differentiate between different types of defects, resolving the traditional inability to quickly determine defect types through AI interpretation. Finally, it traces back to pinpoint defect locations and triggers targeted optimization processes, addressing the slow traditional defect location and generalized optimization measures, thus achieving precise optimization of quality indicators.

[0026] Specifically, step S5, which involves constructing a true standard based on actual borehole drilling data and calibrating multi-dimensional quality indicators, includes: S51: Selecting field-verified stratigraphic depth data from the actual borehole drilling data as the true standard. The stratigraphic depth data must include at least the depth of the coal seam roof and floor and the depth of the marker layer; S52: Calculating the depth deviation between the interpreted stratigraphic depth data and the true standard. The depth deviation is calculated as |interpreted stratigraphic depth - true stratigraphic depth|; S53: Constructing a data quality calibration coefficient based on the depth deviation. The calibration coefficient is calculated as 1 - (depth deviation / true stratigraphic depth). When the calibration coefficient is lower than a preset calibration coefficient threshold, the stratigraphic interpretation accuracy indicator is corrected. The corrected stratigraphic interpretation accuracy indicator is calculated as: original indicator × calibration coefficient; S54: Substituting the corrected multi-dimensional quality indicators into the comprehensive quality scoring formula. The comprehensive quality score is based on each dimension... The quality indicators are calculated by weighting them with their preset weights. Each dimension of the quality indicators includes at least the data dynamic adaptation indicator, the stratigraphic interpretation micro-phase difference indicator, the stratigraphic interpretation accuracy indicator, and the borehole missing section conformity indicator. By refining the true value calibration and comprehensive scoring process, the problems of insufficient accuracy and lack of systematic scoring methods in traditional evaluation are solved: by selecting actual borehole drilling data verified in the field as the true value standard, the problem of lack of objective benchmark in traditional calibration is solved, providing a reliable reference for quality indicator calibration; by calculating the depth deviation and constructing calibration coefficients to correct the stratigraphic interpretation accuracy indicator, the problem of uncalibrated traditional indicators and large deviations from actual accuracy is solved, improving the reliability of the indicators; by weighting the multi-dimensional correction indicators to obtain the comprehensive quality score, the problem of fragmented traditional scoring methods and inability to reflect the overall quality is solved, realizing a comprehensive quantitative evaluation of the overall data quality.

[0027] Specifically, step S3 also includes a quality index extraction step for the edge connection zone. A two-dimensional dynamic threshold model based on region type and influence factors is used to calculate the acceptable threshold for the edge connection consistency coefficient. Specifically: S331: Calculate the edge connection consistency coefficient for the edge connection zone, where: for SGY data volume splicing seams, consistency coefficient = 1 - (mean phase difference of adjacent data volume phase axes / dynamic phase difference threshold); for the edge blank zone of borehole distribution, consistency coefficient = 1 - (deviation rate between edge borehole and seismic horizon depth / dynamic deviation rate threshold); for the end area of ​​the roadway line extension, consistency coefficient = 1 - (offset between the end of the roadway line and geological trend / dynamic offset threshold); for the boundary area of ​​seismic data coverage times, consistency coefficient = 1 - (deviation between measured and designed coverage times). / Dynamic Coverage Frequency Deviation Threshold); S332: Construct a two-dimensional dynamic threshold calculation model of region type and influence factors to generate dynamic thresholds for each edge region. The specific steps are as follows: S3321: Determine the core influence factors and weights of each edge region. Among them, the influence factors of SGY splicing seams include data volume splicing accuracy (weight 0.4), seismic data frequency (weight 0.3), and phase axis continuity (weight 0.3); the influence factors of borehole edge blank zones include borehole density (weight 0.5) and borehole distribution uniformity (weight 0.5); the influence factors of tunnel end areas include tunnel measurement accuracy (weight 0.6) and geological trend fitting degree (weight 0.4); the influence factors of seismic coverage boundary areas include the design value of coverage frequency (weight 0.3) and the complexity of seismic acquisition environment (weight 0).7) S3322: Calculate the actual contribution value of each influencing factor. Contribution value = (Measured value of influencing factor / Standard value of influencing factor) × Influence factor weight. The larger the contribution value, the better the performance of the corresponding influencing factor. S3323: Dynamically generate the threshold for each edge region. Dynamic threshold = Preset basic threshold / (1 + Σ Influence factor contribution value). The preset basic threshold is determined based on historical high-quality geological data. When the performance of the influencing factor is better (the larger the Σ Influence factor contribution value), the dynamic threshold is smaller, and the judgment standard for the data quality of the edge connection area is more stringent. S3324: Set the qualified threshold of the edge connection consistency coefficient as the dynamic threshold of the corresponding area. When the consistency coefficient is lower than the qualified threshold, the data quality of the edge connection area is judged to be substandard, and the reasons for substandard performance are analyzed based on the contribution value of the influencing factor: If the SGY splicing seam threshold is substandard, it is due to the low contribution value of splicing accuracy (sponge... If the data splicing accuracy is poor, it is determined to be an error in the data splicing parameters; if the threshold of the blank zone at the borehole edge does not meet the standard due to a low borehole density contribution value (insufficient borehole density), it is determined to be insufficient borehole data. By refining the evaluation process for the edge connection zone, the problems of traditional edge area evaluation using fixed thresholds, lack of specific indicators, and unclear causes are solved: by calculating specific quality coefficients for different types of edge areas, the problem of traditional methods lacking targeted indicators and being unable to accurately evaluate edge area quality is solved; by constructing a two-dimensional dynamic threshold model, combining regional characteristics and influencing factors to generate dynamic thresholds, the problem of traditional fixed thresholds being unable to adapt to edge area differences is solved, making the thresholds match the actual situation of the region and allowing the severity to be adjusted according to the performance of influencing factors; by analyzing the contribution values ​​of influencing factors to clarify the causes of quality defects, the problem of traditional methods being unable to locate the causes of defects is solved, providing a clear direction for subsequent targeted optimization.

[0028] Specifically, step S6, outputting a geological data quality evaluation report, includes: S61: Generating a data quality heatmap, using color gradients to represent the distribution of comprehensive quality scores. The first color indicates a score below a first preset score threshold (poor quality), the second color indicates a score between the first and second preset score thresholds (medium quality), and the third color indicates a score above the second preset score threshold (excellent quality). The quality status of key evaluation areas is also marked, with the second preset score threshold being higher than the first preset score threshold. S62: Overlaying data quality defect location markers onto the heatmap, and combining defect causes, dynamic adaptation deviation thresholds, and dynamic thresholds for edge connection zones, providing targeted optimization suggestions: If the SGY splice quality is substandard, it is recommended to recalibrate the data splicing parameters; if the quality of the borehole edge blank zone is substandard, it is recommended to supplement borehole measurements in the edge area. S63: Integrating the optimization suggestions with the preprocessed multi-source data... By linking geological data, an iterative quality assessment closed loop is formed. When geological data for the same area is acquired again, historical dynamic thresholds and quality assessment reports are automatically retrieved for comparative analysis. The dynamic weight library, coupling coefficient, and marginal area influence factor weights are updated. By refining the assessment report output process, the problems of poor practicality and lack of iterative mechanism in traditional reports are solved: by generating a color-gradient quality heatmap and marking the quality of key areas, the problem of unintuitive quality distribution and difficulty in quickly grasping the overall quality in traditional reports is solved, achieving visualization of quality distribution; by overlaying defect location markers and combining thresholds and causes to provide specific optimization suggestions, the problems of incomplete defect information and generalized suggestions in traditional reports are solved, providing actionable directions for quality improvement; by establishing an iterative evaluation closed loop and calling historical data to update key parameters, the problem of lack of continuous optimization mechanism and inability to improve accuracy in traditional evaluations is solved, enabling continuous improvement in the adaptability and accuracy of subsequent evaluations.

[0029] A geological data quality evaluation system for geological delineation specifically includes: a data preprocessing module for acquiring and preprocessing multi-source geological data of the area to be evaluated, wherein the multi-source geological data includes at least SGY seismic processing result data volume, borehole data, and coal mine roadway line data; during the preprocessing process, a dynamic adaptation deviation threshold calculated based on scene priority and data coupling degree is used to achieve preliminary screening and compensation of data quality; a key area identification module for identifying key evaluation areas in the multi-source geological data, wherein key evaluation areas include at least data loading edge connection areas, borehole data missing sections, and boundary areas of seismic data coverage times; and a quality index extraction module for extracting quality indicators from the global data and key evaluation areas. The system extracts multi-dimensional quality indicators from multi-source geological data. These indicators include at least parameters for evaluating data dynamic adaptability, subtle phase differences in stratigraphic interpretation, and stratigraphic interpretation accuracy. An AI-driven optimization module is used to optimize these multi-dimensional quality indicators by combining AI intelligent interpretation results, establishing a mapping relationship between AI interpretation confidence and data quality defects, and enabling the retrospective location of data quality defects. A quality calibration and scoring module is used to construct a true value standard based on borehole drilling data, calibrate the multi-dimensional quality indicators, and generate a comprehensive quality score. Finally, an evaluation report output module outputs a geological data quality evaluation report containing the quality status of key evaluation areas, the location of data quality defects, and optimization suggestions.

[0030] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A method for evaluating the quality of geological data used in geological delineation, characterized in that, Includes the following steps: S1: Obtain multi-source geological data of the area to be evaluated and perform preprocessing. The multi-source geological data includes SGY seismic processing results data, borehole results data and coal mine roadway data. During the preprocessing process, a dynamic adaptation deviation threshold based on scene priority and data coupling degree is used to achieve preliminary screening and compensation of data quality. S2: Identify key evaluation areas in the multi-source geological data, including data loading edge connection areas, missing borehole data segments, and boundary areas of seismic data coverage times; S3: For the global data and the key evaluation area, extract multi-dimensional quality indicators of the multi-source geological data. The multi-dimensional quality indicators include index parameters used to evaluate the dynamic adaptability of the data, the small phase difference in stratigraphic interpretation, and the accuracy of stratigraphic interpretation. S4: Combine the AI ​​intelligent interpretation results to optimize the multi-dimensional quality indicators, establish the mapping relationship between AI interpretation confidence and data quality defects, and realize the retrospective location of data quality defects; S5: Construct a true value standard based on actual drilling data, calibrate the multi-dimensional quality indicators, and generate a comprehensive quality score; S6: Output a geological data quality evaluation report that includes the quality status of key evaluation areas, the location of data quality defects, and optimization suggestions; The step S1, which involves acquiring and preprocessing multi-source geological data of the area to be evaluated, specifically includes: S11: Load the SGY seismic processing results data, borehole data and coal mine roadway line data, and extract the basic characteristic parameters of each data type. The basic characteristic parameters of SGY data include signal-to-noise ratio, sampling interval and phase continuity; the basic characteristic parameters of borehole data include depth error and lithological labeling integrity; and the basic characteristic parameters of roadway line data include spatial location accuracy and extension continuity. S12: Establish a dynamic weight library for basic feature parameters of geological classification scenarios. The geological classification scenarios include coal seam thickness classification scenarios and fault boundary classification scenarios. Dynamic weights are assigned to each data feature parameter for different scenarios. In the coal seam thickness classification scenario, the weight value of borehole data depth error is higher than that in the fault boundary classification scenario. In the fault boundary classification scenario, the weight value of SGY data phase continuity is higher than that in the coal seam thickness classification scenario. S13: Based on the priority of geological scene division and the coupling relationship between multi-source data features, a dynamic adaptation deviation threshold is generated. By comparing the actual adaptation deviation with this threshold, corresponding data adaptability judgment and compensation processing are performed. Specifically: when the actual adaptation deviation meets the threshold requirement, it is judged as adaptable; when the actual adaptation deviation exceeds the threshold but is within the preset range, it is compensated and corrected using associated data; when the actual adaptation deviation exceeds the preset range, it is judged as data mismatch and the corresponding data optimization processing mechanism is triggered to obtain usable data.

2. The geological data quality evaluation method for geological delineation according to claim 1, characterized in that, The identification of key evaluation areas in the multi-source geological data in step S2 specifically includes: S21: Identify the edge connection area of ​​the data loading. The edge connection area includes the splicing seam of multiple SGY data volumes, the edge blank zone of borehole distribution, the end area of ​​the roadway line extension, and the boundary area of ​​the seismic data coverage number. The boundary range of the edge connection area is determined by data coordinate overlap detection, data density gradient analysis, and seismic coverage number distribution calculation. S22: Identify missing sections of borehole data, locate the depth interval with data gaps by detecting the continuity of the borehole depth sequence, and record the valid borehole data above and below the missing section. The valid borehole data includes the stratigraphic depth and lithological information. S23: Mark the edge connection area and the missing section of borehole data to generate a key evaluation area distribution map.

3. The geological data quality evaluation method for geological delineation according to claim 1, characterized in that, Step S3, which extracts multi-dimensional quality indicators from the multi-source geological data for the global data and the key evaluation area, specifically includes: S311: Extract dynamic data adaptation indicators, and calculate the adaptation score for each data type based on the dynamic weight library and coupling coefficient. Adaptation score = Σ (measured value of basic feature parameter / standard value of feature parameter × dynamic weight × K), where K is the coupling coefficient. When the adaptation score is lower than the preset adaptation score threshold, it is determined that the dynamic adaptation indicator is not up to standard. S312: Extract the stratigraphic interpretation of small phase difference indicators, use wavelet transform algorithm to extract the phase features of the stratigraphic phase axis in the SGY data volume, capture the phase deviation within the predetermined small phase difference range, calculate the distribution density and continuous length of the small phase difference, and mark it as a phase abnormal region when the continuous length exceeds the preset sampling point threshold. S313: Extract the stratigraphic interpretation accuracy index, obtain the stratigraphic interpretation depth data on the SGY profile by AI or manual means, calculate the dispersion of the interpretation depth data, dispersion = (maximum interpretation depth - minimum interpretation depth) / average interpretation depth, when the dispersion exceeds the preset dispersion threshold, it is determined that the stratigraphic interpretation accuracy index does not meet the standard.

4. The geological data quality evaluation method for geological delineation according to claim 2, characterized in that, Step S3 further includes a step for extracting quality indicators for missing segments of borehole data, specifically: S321: Based on the effective borehole data above and below the missing section, a theoretical geological model of the missing section is constructed using a cubic polynomial interpolation algorithm. The theoretical geological model includes theoretical stratigraphic depth curves and theoretical lithological distribution. S322: Extract SGY data for the area corresponding to the missing section of the borehole, perform stratigraphic interpretation on the area, and obtain the actual stratigraphic interpretation results; S323: Calculate the degree of agreement between the actual stratigraphic interpretation results and the theoretical geological model. The degree of agreement = (1 - |actual stratigraphic depth - theoretical stratigraphic depth| / theoretical stratigraphic depth) × 100%. When the degree of agreement is lower than the preset degree of agreement threshold, it is determined that the SGY data quality of the area corresponding to the missing section of the borehole is substandard.

5. The geological data quality evaluation method for geological delineation according to claim 1, characterized in that, The step S4, which combines the AI ​​intelligent interpretation results to optimize the multi-dimensional quality indicators, specifically includes: S41: Obtain AI intelligent interpretation results, which include automatic layer tracking results and fault intelligent identification results. Simultaneously record the confidence value of each interpretation result. The confidence value is calculated based on the continuity of seismic phase axes, borehole data consistency, and coupling coefficient. S42: Establish a mapping model between AI interpretation confidence and data quality defects. When the confidence value is lower than the preset confidence threshold, it is determined that there are data quality defects in the corresponding area. When the confidence value is in the first preset confidence interval, it corresponds to data noise interference defects. When the confidence value is in the second preset confidence interval, it corresponds to data sampling accuracy defects. The upper limit of the second preset confidence interval is lower than the lower limit of the first preset confidence interval. S43: Based on the mapping model, backtrack to locate the specific location of data quality defects, trigger the corresponding data preprocessing optimization process according to the defect type, and optimize multi-dimensional quality indicators.

6. The geological data quality evaluation method for geological delineation according to claim 1, characterized in that, Step S5, which involves constructing a truth standard based on actual borehole drilling data and calibrating the multi-dimensional quality indicators, specifically includes: S51: Select the stratigraphic depth data verified in the field from the actual drilling data as the true value standard. The stratigraphic depth data includes the depth of the coal seam roof and floor and the depth of the marker layer. S52: Calculate the depth deviation between the interpreted depth data and the true standard depth. Depth deviation = |interpreted depth - true standard depth|. S53: Construct a data quality calibration coefficient based on the depth deviation value. The calibration coefficient = 1 - (depth deviation value / true layer depth). When the calibration coefficient is lower than the preset calibration coefficient threshold, the layer interpretation accuracy index is corrected. The corrected layer interpretation accuracy index = original layer interpretation depth × calibration coefficient. S54: Substitute the revised multi-dimensional quality indicators into the comprehensive quality scoring formula. The comprehensive quality score is calculated based on the quality indicators of each dimension and their preset weights. The quality indicators of each dimension include data dynamic adaptation indicators, layer interpretation micro phase difference indicators, layer interpretation accuracy indicators, and borehole missing segment matching indicators.

7. A method for evaluating the quality of geological data for geological delineation according to claim 2, characterized in that, Step S3 further includes a quality index extraction step for the edge connection area, which uses a region type-influence factor dual-dimensional dynamic threshold model to calculate the qualified threshold of the edge connection consistency coefficient, specifically: S331: Calculate the edge connection consistency coefficient of the edge connection area, where: For the splicing seam of SGY data volumes, the consistency coefficient = 1 - (mean phase difference of in-phase axis of adjacent data volumes / dynamic phase difference threshold). For the edge blank zone of borehole distribution, the consistency coefficient = 1 - (the deviation rate of edge borehole depth from seismic horizon / dynamic deviation rate threshold). For the terminal area of ​​the tunnel line extension, the consistency coefficient = 1 - (the offset between the tunnel line end and the geological trend / the dynamic offset threshold). For the boundary area of ​​seismic data coverage, the consistency coefficient = 1 - (the deviation between the measured and designed coverage values ​​ / the dynamic coverage deviation threshold). S332: Construct a dynamic threshold calculation model with two dimensions of region type and impact factor to generate dynamic thresholds for each marginal region, including: determining the core impact factor and weight according to the marginal region type, calculating the actual contribution value of each impact factor, and dynamically generating the threshold of the corresponding region based on the contribution value; setting the dynamic threshold as the qualified threshold of the marginal connection consistency coefficient, and judging the data quality as substandard when the consistency coefficient is lower than the qualified threshold, and analyzing the reasons for the substandard based on the contribution value of the impact factor.

8. The geological data quality evaluation method for geological delineation according to claim 1, characterized in that, The geological data quality evaluation report output in step S6 specifically includes: S61: Generate a data quality heatmap, using color gradients to represent the distribution of comprehensive quality scores, where the first color indicates a score below the first preset score threshold, the second color indicates a score between the first and second preset score thresholds, and the third color indicates a score above the second preset score threshold. The quality status of key evaluation areas is also marked, and the second preset score threshold is higher than the first preset score threshold. S62: Overlay data quality defect location markers on the heat map, and provide targeted optimization suggestions based on defect causes, dynamic adaptation deviation thresholds, and dynamic thresholds of edge connection areas: If the SGY splice seam quality is substandard, it is recommended to recalibrate the data splicing parameters; if the quality of the blank zone at the borehole edge is substandard, it is recommended to supplement the borehole measurement in the edge area. S63: The optimization suggestions are associated with the preprocessed multi-source geological data to form an iterative quality evaluation closed loop. When the geological data of the area to be evaluated is acquired again, the historical dynamic threshold and quality evaluation report are automatically called for comparative analysis, and the dynamic weight library, coupling coefficient and marginal area influence factor weight are updated.

9. A geological data quality evaluation system for geological delineation, characterized in that, The system includes the geological data quality evaluation method for geological delineation as described in any one of claims 1-8, specifically including: The data preprocessing module is used to acquire and preprocess multi-source geological data of the area to be evaluated. The multi-source geological data includes SGY seismic processing results data, borehole results data and coal mine roadway line data. During the preprocessing process, a dynamic adaptation deviation threshold based on scene priority and data coupling degree is used to achieve preliminary screening and compensation of data quality. The key area identification module is used to identify key evaluation areas in the multi-source geological data. The key evaluation areas include data loading edge connection areas, missing borehole data segments, and boundary areas of seismic data coverage times. The quality index extraction module is used to extract multi-dimensional quality indicators from the multi-source geological data for the global data and the key evaluation area. The multi-dimensional quality indicators include index parameters used to evaluate the dynamic adaptability of the data, the small phase difference in stratigraphic interpretation, and the accuracy of stratigraphic interpretation. The AI ​​feedback optimization module is used to combine the AI ​​intelligent interpretation results to provide feedback optimization for the multi-dimensional quality indicators, establish a mapping relationship between the confidence of AI interpretation and data quality defects, and realize the retrospective location of data quality defects. The quality calibration and scoring module is used to construct a true value standard based on actual drilling data, calibrate the multi-dimensional quality indicators, and generate a comprehensive quality score. The evaluation report output module is used to output a geological data quality evaluation report that includes the quality status of key evaluation areas, the location of data quality defects, and optimization suggestions.

Images