A method, system, and apparatus for improving the resolution of logging data in thin sand layers.
By combining wavelet transform and chart correction with the K-MEANS clustering algorithm, the problem of insufficient resolution in thin sand layer logging data was solved, achieving high-resolution processing of thin sand layer logging data and improving the accuracy and reliability of logging interpretation.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHINA PETROLEUM & CHEMICAL CORP
- Filing Date
- 2024-12-30
- Publication Date
- 2026-06-30
AI Technical Summary
Under complex thin-layer geological conditions, conventional logging techniques are unable to accurately identify the true physical characteristics of thin sand layers, leading to discrepancies between logging interpretations and actual formation characteristics. This affects the interpretation of fluid properties in thin reservoirs and the accurate calculation of reservoir parameters.
Wavelet transform technology was used for multi-resolution analysis, combined with the chart correction method and K-MEANS clustering algorithm. The characteristic signals of thin sand layers were corrected through an iterative process to ensure that the logging data and core analysis results were consistent.
It enables high-resolution processing of logging data in thin sand layers, improving the accuracy and reliability of logging data and ensuring that logging results are highly consistent with actual formation conditions.
Smart Images
Figure CN122307732A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of well logging data processing technology, and in particular to a processing method, system and device for improving the resolution of well logging data in thin sand layers. Background Technology
[0002] The Tarim Carboniferous reservoirs in the Tarim Basin exhibit significant lithological variability, strong heterogeneity, and limited effective thickness, which significantly impacts the physical properties of oil and gas layers and water layers. Particularly under thin sandstone conditions, conventional logging techniques struggle to accurately capture formation characteristics, leading to discrepancies between logging interpretations and actual formation properties. Currently, while logging instruments from brands such as Jiai and Fenghe are widely used in oil and gas exploration in this region, they cannot accurately identify thin layers when acquiring sonic transit time and conventional induced resistivity curves. This situation makes the logging response of thin sandstone layers susceptible to interference from the surrounding rock medium, failing to accurately reflect their true physical characteristics and severely impacting the interpretation of fluid properties and the accurate calculation of reservoir parameters.
[0003] Existing technology CN114488293A discloses a high-resolution inversion method based on sensitive logging curves, including: acquiring well curve data; determining whether the well curve data is abnormal to obtain standardized curve data; analyzing and screening the standardized curve data to obtain sensitive curve data; reconstructing the sensitive curve data to form pseudo-acoustic curve data; and performing seismic inversion on the pseudo-acoustic curve data. By adding a sensitive curve that is more sensitive to lithology to the original curve to form a new pseudo-acoustic curve, and performing seismic inversion on it, a three-dimensional impedance volume that can reflect the reservoir can be obtained, thereby facilitating reservoir prediction.
[0004] However, the above methods do not solve the problems of insufficient adaptability and accuracy under complex thin geological conditions, thus the high-resolution processing of sonic logging data is not fast and accurate enough.
[0005] Therefore, there is an urgent need to provide a processing method, system, and device for improving the resolution of logging data in thin sand layers, and to perform high-resolution processing of acoustic logging data quickly and accurately compared to existing technologies. Summary of the Invention
[0006] This invention addresses the technical problems existing in the prior art and provides a processing method, system, and device for improving the resolution of logging data in thin sand layers.
[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows:
[0008] A method, system, and apparatus for improving the resolution of logging data in thin sand layers, comprising the following steps:
[0009] S1. Obtain well logging data from thin sandy strata in the area to be analyzed;
[0010] S2. Perform data preprocessing on the logging data obtained in step S1;
[0011] S3. For the logging data processed in step S2, wavelet transform technology is used for multi-resolution analysis to extract the characteristic signals of thin sand layers and enhance the resolution of the characteristic signals of thin sand layers.
[0012] S4. The characteristic signals of the thin sand layer enhanced in step S3 are processed using the plate correction method to improve accuracy.
[0013] S5. Evaluate whether the characteristic signal of the thin sand layer after processing in step S4 is qualified. If the evaluation is unqualified, iterate through steps S1-S5 until the evaluation is qualified. When the evaluation is qualified, it means that the characteristic signal of the thin sand layer after processing in step S4 is the logging data of the thin sand layer with improved resolution.
[0014] Furthermore, S4 specifically includes the following steps:
[0015] S41. Set the standard logging response chart for the thin sand layer in the area to be analyzed;
[0016] S42. Based on the characteristic signals of the thin sand layer enhanced in step S3, construct lithological maps of different lithological distributions at each depth, and simultaneously construct thickness maps of different thickness distributions corresponding to each lithology.
[0017] S43. Compare the lithology charts with different lithological distributions at each depth and the thickness charts with different thickness distributions corresponding to each lithology with the standard logging response charts. For data points that are different from the standard logging response charts, modify them to be the same as the corresponding data points in the standard logging response charts, thereby obtaining the corrected characteristic signals of the thin sand layer.
[0018] Furthermore, lithological maps of different lithological distributions at each depth are constructed using the K-MEANS clustering algorithm.
[0019] Furthermore, the specific method for constructing lithological maps of different lithological distributions at each depth using the K-MEANS clustering algorithm is as follows: In the feature signal of the thin sand layer obtained in step S3, K1 data points representing different depths are first selected as initial centroids. For each data point in the feature signal of the thin sand layer, the distance from it to each initial centroid is calculated, and it is assigned to the cluster containing the nearest initial centroid. For each cluster, the average value of all data points assigned to that cluster is calculated, the centroid of that cluster is updated, and the iteration is performed until the centroid no longer changes or the first set number of iterations is reached. When the iteration stops, multiple clusters are obtained, each cluster representing a different depth classification. Lithological labels are applied to the data points in each cluster, and each cluster represents a lithological map of different lithological distributions at a depth.
[0020] Furthermore, thickness maps with different thickness distributions corresponding to each lithology are constructed using the K-MEANS clustering algorithm.
[0021] Furthermore, the specific method for constructing thickness maps with different thickness distributions corresponding to each lithology using the K-MEANS clustering algorithm is as follows: In the feature signal of the thin sand layer obtained in step S3, K2 data points representing different lithologies are first selected as initial centroids. For each data point in the feature signal of the thin sand layer, the distance from it to each initial centroid is calculated, and it is assigned to the cluster containing the nearest initial centroid. For each cluster, the average value of all data points assigned to that cluster is calculated, the centroid of that cluster is updated, and the iteration is performed until the centroid no longer changes or the second set number of iterations is reached. When the iteration stops, multiple clusters are obtained, each cluster representing a different lithology classification. The data points in each cluster are labeled with thickness values, and each cluster represents a thickness map with different thickness distributions under a certain lithology.
[0022] Furthermore, the method for evaluating whether the characteristic signal of the thin sand layer processed in step S4 is qualified in step S5 is as follows: core analysis is performed on the thin sand layer strata in the area to be analyzed, and the characteristic signal of the thin sand layer processed in step S4 is compared with the core analysis results. The distance between each corresponding data point is calculated, and a distance threshold is set. When there is a set of corresponding data points with a distance greater than the distance threshold, it indicates that the characteristic signal of the thin sand layer processed in step S4 is unqualified; otherwise, it is qualified.
[0023] Furthermore, S2 specifically includes the following steps:
[0024] S21. Noise filtering of well logging data;
[0025] S22. Perform depth correction on the logging data processed in step S22.
[0026] S23. Smooth the logging data after processing through steps S21 and S22.
[0027] Furthermore, in step S21, filtering techniques are used to remove noise from the logging data.
[0028] Furthermore, the specific method for depth correction in step S22 is as follows: core analysis is performed on the thin sand layer formation, and depth correction is performed on the logging data based on the core analysis results to ensure that the depth in the logging data corresponds one-to-one with the depth in the core analysis results.
[0029] Furthermore, in step S23, a smoothing algorithm is used to smooth the logging data after it has been processed by steps S21 and S22.
[0030] Furthermore, S3 specifically includes the following steps:
[0031] S31. Set the wavelet basis function and perform wavelet decomposition on the logging data processed in step S2 to obtain the wavelet decomposition results, which include high-frequency components and low-frequency components.
[0032] S32. Remove the high-frequency components from the wavelet decomposition results and retain the low-frequency components, which are the characteristic signals of the thin sand layer.
[0033] S33. The feature signals of the thin sand layer are processed by wavelet reconstruction technology to enhance the resolution of the feature signals of the thin sand layer.
[0034] A processing system for improving the resolution of logging data in thin sand layers includes a first module, a second module, a third module, a fourth module, a fifth module, and a sixth module. The first module is used to execute step S1, the second module is used to execute step S2, the third module is used to execute step S3, the fourth module is used to execute step S4, the fifth module is used to execute step S5, and the sixth module is used to execute step S6.
[0035] A processing device for improving the resolution of logging data in thin sand layers includes a storage medium and a processor. The storage medium stores a computer program, and the processor executes the computer program to implement a processing method for improving the resolution of logging data in thin sand layers.
[0036] Compared with the prior art, the beneficial effects of the present invention are as follows:
[0037] The accuracy of the corrected well logging data is evaluated by comparing it with laboratory core analysis results. By adjusting wavelet transform and chart correction parameters during the iterative process, the final well logging data for the thin sand layer with improved resolution is ensured to highly match the actual formation conditions. Attached Figure Description
[0038] Figure 1 This is a flowchart of the method of the present invention.
[0039] Figure 2 This is a comparison chart of the well logging data processed by this invention before and after processing. Detailed Implementation
[0040] The technical solution of the present invention will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are not all embodiments of the present invention. All other embodiments obtained by those skilled in the art without creative effort are within the protection scope of the present invention.
[0041] like Figure 1 As shown, this invention provides a method for improving the resolution of logging data in thin sand layers, comprising the following steps:
[0042] S1. Obtain well logging data from thin sandy strata in the area to be analyzed. Figure 2 (A)) The logging data includes sonic transit time curves and ordinary induced resistivity curves; logging data in thin sand formations is obtained using existing logging instruments, with the existing logging instruments being logging equipment from the brands Jiai and Fenghe.
[0043] S2. Perform data preprocessing on the logging data obtained in step S1, specifically including the following steps:
[0044] S21. Noise filtering is performed on the logging data. Filtering technology is used to remove environmental noise and instrument errors from the logging data to ensure the purity of the data.
[0045] S22. Perform depth correction on the logging data processed in step S22, conduct core analysis on the thin sand layer formation, and perform depth correction on the logging data based on the core analysis results to ensure that the depth in the logging data corresponds one-to-one with the depth in the core analysis results.
[0046] S23. The logging data processed in steps S21 and S22 are smoothed. Specifically, a smoothing algorithm is used to process the logging data to reduce random fluctuations and improve the consistency and interpretability of the sonic transit time curve and the ordinary induced resistivity curve in the logging data.
[0047] S3. For the logging data processed in step S2, wavelet transform technology is used for multi-resolution analysis to extract the characteristic signals of thin sand layers and enhance the resolution of the extracted characteristic signals of thin sand layers. Specifically, this includes the following steps:
[0048] S31. Set the wavelet basis function and perform wavelet decomposition on the logging data processed in step S2 to obtain the wavelet decomposition results, which include high-frequency components and low-frequency components.
[0049] S32. Remove the high-frequency components from the wavelet decomposition results and retain the low-frequency components, which are the characteristic signals of the thin sand layer.
[0050] S33. The characteristic signals of the thin sand layer are processed by wavelet reconstruction technology to improve the resolution of the characteristic signals of the thin sand layer, so as to make it into high-resolution logging data.
[0051] S4. The characteristic signals of the thin sand layer obtained in step S3 are processed using the plate correction method to further improve accuracy. This includes the following steps:
[0052] S41. Set up a standard logging response chart for the thin sand layer in the area to be analyzed. The standard logging response chart is a cross-plot of lithology and thickness. The cross-plot of lithology and thickness is drawn based on the core analysis results and logging data.
[0053] S42. Based on the characteristic signals of the thin sand layer obtained in step S3, construct lithology maps for different lithological distributions at each depth, and simultaneously construct thickness maps for different thickness distributions corresponding to each lithology. This specifically includes the following steps:
[0054] S421. Using the K-MEANS clustering algorithm, in the feature signal of the thin sand layer obtained in step S3, first select K1 data points representing different depths as initial centroids. For each data point in the feature signal of the thin sand layer, calculate its distance to each initial centroid and assign it to the cluster where the nearest initial centroid is located. For each cluster, calculate the average value of all data points assigned to that cluster, update the centroid of that cluster, and iterate until the centroid no longer changes or the first set number of iterations is reached. When the iteration stops, multiple clusters are obtained, each cluster representing a classification at a different depth. Lithological labels are applied to the data points in each cluster, so each cluster represents a lithological map of different lithological distributions at a certain depth.
[0055] S422. Using the K-MEANS clustering algorithm, in the feature signal of the thin sand layer obtained in step S3, first select K2 data points representing different lithologies as initial centroids. For each data point in the feature signal of the thin sand layer, calculate its distance to each initial centroid and assign it to the cluster where the nearest initial centroid is located. For each cluster, calculate the average value of all data points assigned to that cluster, update the centroid of that cluster, and iterate until the centroid no longer changes or reaches the second set number of iterations. When the iteration stops, multiple clusters are obtained, each cluster representing a classification of different lithologies. Thickness values are marked for the data points in each cluster, so each cluster represents a thickness map of different thickness distributions under a certain lithology.
[0056] S43. Compare the lithology charts with different lithological distributions at each depth and the thickness charts with different thickness distributions corresponding to each lithology with the standard logging response charts. For data points that are different from the standard logging response charts, modify them to be the same as the corresponding data points in the standard logging response charts, thereby obtaining the characteristic signals of the corrected thin sand layers, which are represented by curves.
[0057] S5. Evaluate whether the characteristic signal of the thin sand layer after processing in step S4 is qualified. The specific method is as follows: compare the characteristic signal of the thin sand layer after processing in step S4 with the core analysis results, calculate the distance between each corresponding data point, and set a distance threshold. When there is a set of corresponding data points with a distance greater than the distance threshold, it means that the characteristic signal of the thin sand layer after processing in step S4 is unqualified, otherwise it is qualified.
[0058] When the evaluation is unqualified, steps S1-S5 are iteratively performed until the evaluation is qualified. When the evaluation is qualified, it means that the characteristic signal of the thin sand layer after step S4 is the logging data of the thin sand layer with improved resolution. Figure 2 (B) in the middle has a certain degree of reliability and accuracy.
[0059] S6. For logging data of thin sand layers with improved resolution, conduct formation analysis, reserve assessment and production capacity prediction of the thin sand layers, and provide more accurate basic data and decision-making basis for oilfield development.
[0060] The present invention also provides a processing system for improving the resolution of logging data in thin sand layers, comprising a first module, a second module, a third module, a fourth module, a fifth module, and a sixth module. The first module is used to execute step S1, the second module is used to execute step S2, the third module is used to execute step S3, the fourth module is used to execute step S4, the fifth module is used to execute step S5, and the sixth module is used to execute step S6.
[0061] The present invention also provides a processing device for improving the resolution of logging data in thin sand layers, comprising a storage medium and a processor. The storage medium stores a computer program, and the processor is used to implement the processing method for improving the resolution of logging data in thin sand layers provided by the present invention when executing the computer program.
[0062] The accuracy of the corrected logging data is evaluated by comparing it with the laboratory core analysis results. By adjusting the wavelet transform and chart correction parameters during the iteration process, the logging data of the thin sand layer with improved resolution is ensured to be highly consistent with the actual formation conditions.
[0063] Finally, it should be noted that the above content is only used to illustrate the technical solution of the present invention, and is not intended to limit the scope of protection of the present invention. Simple modifications or equivalent substitutions made by those skilled in the art to the technical solution of the present invention do not depart from the essence and scope of the technical solution of the present invention.
Claims
1. A method for improving the resolution of logging data in thin sand layers, characterized in that, Includes the following steps: S1. Obtain well logging data from thin sandy strata in the area to be analyzed; S2. Perform data preprocessing on the logging data obtained in step S1; S3. For the logging data processed in step S2, wavelet transform technology is used for multi-resolution analysis to extract the characteristic signals of thin sand layers and enhance the resolution of the characteristic signals of thin sand layers. S4. The characteristic signals of the thin sand layer enhanced in step S3 are processed using the plate correction method to improve accuracy. S5. Evaluate whether the characteristic signal of the thin sand layer after processing in step S4 is qualified. If the evaluation is unqualified, iterate through steps S1-S5 until the evaluation is qualified. When the evaluation is qualified, it means that the characteristic signal of the thin sand layer after processing in step S4 is the logging data of the thin sand layer with improved resolution.
2. The method for improving the resolution of logging data in thin sand layers according to claim 1, characterized in that, S4 specifically includes the following steps: S41. Set the standard logging response chart for the thin sand layer in the area to be analyzed; S42. Based on the characteristic signals of the thin sand layer enhanced in step S3, construct lithological maps of different lithological distributions at each depth, and simultaneously construct thickness maps of different thickness distributions corresponding to each lithology. S43. Compare the lithology charts with different lithological distributions at each depth and the thickness charts with different thickness distributions corresponding to each lithology with the standard logging response charts. For data points that are different from the standard logging response charts, modify them to be the same as the corresponding data points in the standard logging response charts, thereby obtaining the corrected characteristic signals of the thin sand layer.
3. The method for improving the resolution of logging data in thin sand layers according to claim 2, characterized in that, Lithological maps of different lithological distributions at each depth were constructed using the K-MEANS clustering algorithm.
4. The method for improving the resolution of logging data in thin sand layers according to claim 3, characterized in that, The specific method for constructing lithological maps of different lithological distributions at each depth using the K-MEANS clustering algorithm is as follows: In the feature signal of the thin sand layer obtained in step S3, K1 data points representing different depths are first selected as initial centroids. For each data point in the feature signal of the thin sand layer, the distance from it to each initial centroid is calculated, and it is assigned to the cluster containing the nearest initial centroid. For each cluster, the average value of all data points assigned to that cluster is calculated, the centroid of that cluster is updated, and the iteration is performed until the centroid no longer changes or the first set number of iterations is reached. When the iteration stops, multiple clusters are obtained, each cluster representing a different depth classification. Lithological labels are applied to the data points in each cluster, and each cluster represents a lithological map of different lithological distributions at a depth.
5. The method for improving the resolution of logging data in thin sand layers according to claim 2, characterized in that, Thickness maps with different thickness distributions corresponding to each lithology were constructed using the K-MEANS clustering algorithm.
6. The method for improving the resolution of logging data in thin sand layers according to claim 5, characterized in that, The specific method for constructing thickness maps with different thickness distributions corresponding to each lithology using the K-MEANS clustering algorithm is as follows: In the feature signal of the thin sand layer obtained in step S3, K2 data points representing different lithologies are first selected as initial centroids. For each data point in the feature signal of the thin sand layer, the distance from it to each initial centroid is calculated, and it is assigned to the cluster containing the nearest initial centroid. For each cluster, the average value of all data points assigned to that cluster is calculated, the centroid of that cluster is updated, and the iteration is performed until the centroid no longer changes or the second set number of iterations is reached. When the iteration stops, multiple clusters are obtained, each cluster representing a different lithology classification. The data points in each cluster are labeled with thickness values, and each cluster represents a thickness map with different thickness distributions under a certain lithology.
7. The method for improving the resolution of logging data in thin sand layers according to claim 1, characterized in that, The method for evaluating whether the characteristic signal of the thin sand layer processed in step S4 is qualified in step S5 is as follows: core analysis is performed on the thin sand layer strata in the area to be analyzed. The characteristic signal of the thin sand layer processed in step S4 is compared with the core analysis results. The distance between each corresponding data point is calculated, and a distance threshold is set. When there is a set of corresponding data points with a distance greater than the distance threshold, it indicates that the characteristic signal of the thin sand layer processed in step S4 is unqualified; otherwise, it is qualified.
8. The method for improving the resolution of logging data in thin sand layers according to claim 1, characterized in that, S2 specifically includes the following steps: S21. Noise filtering of well logging data; S22. Perform depth correction on the logging data processed in step S22. S23. Smooth the logging data after processing through steps S21 and S22.
9. The method for improving the resolution of logging data in thin sand layers according to claim 8, characterized in that, In step S21, filtering techniques are used to remove noise from the logging data.
10. A method for improving the resolution of logging data in thin sand layers according to claim 8, characterized in that, The specific method for depth correction in step S22 is as follows: core analysis is performed on the thin sand layer formation, and depth correction is performed on the logging data based on the core analysis results to ensure that the depth in the logging data corresponds one-to-one with the depth in the core analysis results.
11. The method for improving the resolution of logging data in thin sand layers according to claim 8, characterized in that, In step S23, a smoothing algorithm is used to smooth the logging data after it has been processed by steps S21 and S22.
12. The method for improving the resolution of logging data in thin sand layers according to claim 1, characterized in that, S3 specifically includes the following steps: S31. Set the wavelet basis function and perform wavelet decomposition on the logging data processed in step S2 to obtain the wavelet decomposition results, which include high-frequency components and low-frequency components. S32. Remove the high-frequency components from the wavelet decomposition results and retain the low-frequency components, which are the characteristic signals of the thin sand layer. S33. The feature signals of the thin sand layer are processed by wavelet reconstruction technology to enhance the resolution of the feature signals of the thin sand layer.
13. A processing system for improving the resolution of logging data in thin sand layers, characterized in that, The processing method for improving the resolution of logging data in thin sand layers according to any one of claims 1-12 includes a first module, a second module, a third module, a fourth module, a fifth module, and a sixth module. The first module is used to perform step S1, the second module is used to perform step S2, the third module is used to perform step S3, the fourth module is used to perform step S4, the fifth module is used to perform step S5, and the sixth module is used to perform step S6.
14. A processing device for improving the resolution of logging data in thin sand layers, characterized in that, It includes a storage medium and a processor, wherein the storage medium stores a computer program, and the processor is used to implement, when executing the computer program, a processing method for improving the resolution of logging data in thin sand layers as described in any one of claims 1-12.