Electric imaging logging mineral component quantitative characterization method based on porosity spectrum correction
By preprocessing and porosity spectrum correction of electrical imaging logging data, combined with the optimal segmentation algorithm and core XRD data, the problems of insufficient resolution and pore fluid influence in the quantitative characterization of minerals in deep complex clastic reservoirs were solved, and high-precision quantitative characterization of mineral components and reservoir evaluation were achieved.
Patent Information
- Application Number
- CN202610147502.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-02-03
- Publication Date
- 2026-04-24
AI Technical Summary
Existing mineral calculation methods suffer from insufficient resolution and failure to effectively separate the influence of pore fluids in the quantitative characterization of minerals in deep and complex clastic reservoirs, resulting in large calculation errors. Traditional methods are costly and have complex interpretation models, while intelligent learning methods lack geological mechanism constraints.
By preprocessing the electrical imaging logging data, a high-density pixel image is generated, the apparent porosity is calculated and a porosity spectrum is generated, the influence of pore fluid is removed, the optimal segmentation algorithm is used to identify the mineral distribution cutoff threshold, and quantitative calibration is performed in combination with core XRD data to achieve accurate characterization of mineral components.
It significantly improves the accuracy and geological reliability of mineral content calculation, can clearly depict subtle changes in mineral content on the wellbore profile, identify thin interlayers and heterogeneous features, and provide high-resolution data support for reservoir evaluation.
Smart Images
Figure CN121918207A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a quantitative characterization method for mineral components in electrical imaging logging based on porosity spectrum correction, belonging to the field of mineral calculation and reservoir evaluation technology. Background Technology
[0002] As exploration and development deepen, the global oil industry is increasingly focusing on the development of unconventional oil and gas resources, among which deep complex clastic rocks (tight gas and tight oil) show broad exploration prospects. Deep clastic rock reservoirs are typically characterized by poor physical properties, complex pore structures, and strong heterogeneity. The vertical distribution of their mineral composition and content plays an important role in reservoir evaluation.
[0003] Quantitative characterization of minerals in deep, complex clastic reservoirs faces numerous technical challenges. Traditional core testing methods (such as XRD, XRF, thin sections, and scanning electron microscopy) can provide accurate mineral composition data, but they have limitations such as discontinuous sampling, high cost, and difficulty in reflecting in-situ formation conditions. Well logging techniques such as elemental capture spectroscopy (ECS) and lithological scanning (LithoScanner) can achieve continuous measurement of key elements, but the instruments are expensive, the interpretation models are complex, and the resolution and accuracy for low-content minerals are often insufficient (Li Zhaoping, 2023).
[0004] Existing mineral calculation methods are typically based on conventional well logging curves. Sima Liqiang et al. (2013) introduced the sensitivity of resistivity curves to lithology into the traditional dual-mineral model, achieving the calculation of calcite mineral content. However, this method has limited adaptability to complex mineral systems. Other methods combine various optimization and other mathematical methods with well logging evaluation. Lai Fuqiang et al. (2022) applied optimization methods to the study of the relationship between elements and mineral content, establishing a mineral inversion model based on elemental logging. Its accuracy is constrained by the quality of elemental logging data and the accuracy of the mineral-element correspondence. In recent years, intelligent learning methods such as neural networks and deep learning have gradually emerged, achieving certain results in the field of geophysical well logging. Li Lidong (2024) used deep learning technology to train a model of the relationship between well logging curves and various minerals, achieving high-precision prediction of mineral content. However, this method lacks geological mechanism constraints in model construction, resulting in weak interpretability. Summary of the Invention
[0005] This invention addresses the key issues of insufficient resolution in conventional well logging mineral calculations and the failure of existing electro-imaging mineral quantitative analysis to effectively separate the influence of pore fluids. It provides a method for quantitative characterization of mineral components in electro-imaging well logging based on porosity spectrum correction. This method effectively solves the calculation errors caused by the influence of fluids on mineral calculations in traditional imaging mineral calculations.
[0006] The technical solution provided by this invention to solve the above-mentioned technical problems is: a method for quantitative characterization of mineral components in electrical imaging logging based on porosity spectrum correction, comprising the following steps: Step 1: Preprocess the raw electrical imaging logging data and generate a high-density pixel image; Step 2: Perform probability density statistical analysis on the pixel values at each depth point in the high-density pixel image to obtain the pixel waveform spectrum; Step 3: Calculate the apparent porosity of each pixel in the high-density pixel image to generate a porosity spectrum; Step 4: Correct the pore fluid effect on the original electro-imaging data based on the pore spectrum, and then perform image enhancement to map the image from the chromaticity value range of 0-255 to the range of 0-100. Then, re-perform probability density statistics on the pure mineral pixel values after correction of all pixels to generate pure mineral pixel waveform spectrum, and then obtain the imaging mineral probability map by equalization processing. Step 5: Based on the mineral content data obtained from the whole-rock XRD analysis of the core, the imaging mineral spectrum was quantitatively calibrated. By comparing and analyzing the imaging mineral spectrum of different samples from a single well with their corresponding experimental mineral content data, the optimal segmentation algorithm was used to identify and derive the distribution cutoff threshold and pixel value range for different minerals.
[0007] A further technical solution is that the preprocessing in step 1 includes normalization processing, velocity correction, image equalization processing, removal of bad resistivity values, and depth correction of natural gamma logging.
[0008] A further technical solution is that the high-density pixel image generation process is as follows: the pre-processed effective conductivity curve data is calibrated into two-dimensional color mark data according to different geological application requirements, mapped into a specific imaging color spectrum, and different resistivity values are calibrated into pixel data of different colors, thereby generating a visually intuitive high-density pixel image.
[0009] A further technical solution is that, in step 2, a statistical method is used to calculate the probability density of the pixel value at each depth point.
[0010] A further technical solution is that, in step 3, a specific image window is selected, and the porosity value of each imaging logging pixel is calculated according to the Archie formula. All pixels are traversed to generate a porosity data volume that completely corresponds to the spatial position of the original resistivity image, i.e., a porosity spectrum is generated.
[0011] A further technical solution is that the specific process of image enhancement in step 4 is as follows: first, the probability of a pixel appearing in each pixel value range is calculated, and a probability histogram is drawn; then, the histogram is adjusted so that the originally dense pixel range is stretched and the sparse range is compressed.
[0012] A further technical solution is that the specific process of setting the distribution cutoff threshold for different minerals in step 5 is as follows: collect paired data of different depth points of the target well in the study area. Each sample includes a normalized imaging mineral spectrum feature vector and the volume fraction of the XRD measured minerals at the corresponding depth point. The optimal segmentation algorithm is used to optimize the threshold.
[0013] A further technical solution is that the optimal segmentation algorithm specifically searches for n optimal segmentation points within the numerical space of the imaging mineral spectrum (0–100), thereby dividing the continuous spectral data into n+1 intervals for different minerals.
[0014] A further technical solution is that the optimal segmentation algorithm uses a step size of 1% to systematically traverse all possible segmentation schemes; for each group of candidate cutoff value combinations, the predicted mineral category divided by the imaging spectrum is compared with the actual mineral category identified by the XRD experiment to calculate the misclassification error; through the optimization process, the segmentation point that minimizes the misclassification error between the predicted category and the experimental category is finally selected, so as to achieve the best match between the imaging electrical response and the experimental mineral data, thereby determining the distribution cutoff threshold of different minerals.
[0015] The present invention has the following beneficial effects: 1. Compared with traditional electro-imaging mineral quantification methods, this invention is the first to calculate the porosity spectrum at the pixel level and accurately remove the contribution of pore fluid based on it. This makes the pure mineral spectrum used for mineral calculation more realistically reflect the inherent properties of the rock skeleton, significantly improving the accuracy and geological reliability of mineral content calculation. 2. This method is based on rock physics models and Archie's formula. Each step of the process has clear physical and geological significance, overcoming the weakness of interpretability in black-box intelligent algorithms. The obtained accurate mineral profile is a key parameter for calculating rock brittleness index, mechanical parameters, and evaluating reservoir "sweet spots," and has significant value for field applications. 3. This invention can clearly and continuously depict the subtle changes in mineral content on the wellbore profile. It has an excellent ability to identify microscopic heterogeneous features such as thin interlayers, calcareous (calcite) cemented bands or interlayers, and effectively captures abrupt changes in mineral content. It provides unprecedented high-resolution data support for the fine structural description of reservoirs and the evaluation of strong heterogeneity. Attached Figure Description
[0016] Figure 1 This is a schematic diagram illustrating the overall implementation principle of the method in this invention; Figure 2 The physical volumetric model of deep clastic reservoir rocks in this invention; Figure 3 To develop the probability density distribution of pixel values in electrical imaging logging; Figure 4This is a schematic diagram of the calculation of porosity spectrum in electrical imaging logging based on Archie's formula in this invention; Figure 5 This is a schematic diagram illustrating the principle of pixel-level porosity spectrum calculation and fluid contribution stripping in this invention; Figure 6 This is a diagram showing the pixel waveform after equalization processing in this invention. Figure 7 This is a schematic diagram of the cutoff threshold for the distribution of electro-imaging pixel values used to distinguish different minerals in this invention; Figure 8 This is a graph showing the calculated mineral composition of an FMI electrical imaging logging image according to an embodiment of the present invention. Figure 9 This is a comparison chart of the calculation accuracy of XRD measured content and calculated content in an embodiment of the present invention; Figure 10 This is an example of an imaging mineral probability spectrum from an embodiment of the present invention; Figure 11 This is a graph showing the cutoff values of different mineral contents in embodiments of the present invention; Figure 12 This is an application effect diagram of an embodiment of the present invention. Detailed Implementation
[0017] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the 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.
[0018] like Figure 1 As shown, the present invention provides a quantitative characterization method for mineral components in electrical imaging logging based on porosity spectrum correction, comprising the following steps: Step 1: Preprocess the raw electrical imaging logging data (normalization, velocity correction, image equalization, removal of bad resistivity values, and depth correction for natural gamma logging) to eliminate abnormal features caused by non-geological factors, avoid misleading the interpretation work, and thus more accurately restore the true geological conditions of the wellbore, obtain effective conductivity curve data, and then scale the effective conductivity curve data into two-dimensional color mark data according to different geological application requirements, map it into a specific imaging color spectrum, and scale different resistivity values into pixel data of different colors, thereby generating a visually intuitive high-density pixel image; Step 2: Perform probability density statistical analysis on the pixel values at each depth point in the high-density pixel image (calculate the probability density of the pixel values at each depth point using statistical methods) to obtain the pixel waveform spectrum; Step 3: Calculate the apparent porosity of each pixel in the high-density pixel image to generate a porosity spectrum; Specifically, a specific image window is selected, and the porosity value of each imaging logging pixel is calculated according to the Archie formula. All pixels are traversed to generate a porosity data volume that completely corresponds to the spatial position of the original resistivity image, i.e., a porosity spectrum is generated. The formula for calculating its porosity is:
[0019]
[0020]
[0021] In the formula: Effective porosity, v / v; The resistivity of formation water is ohm / m. The true resistivity of the formation is ohm mm; To obtain the formation resistivity of the flushing zone, it was obtained from shallow lateral resistivity logging, ohm m; The resistivity of the slurry filtrate is given in ohms (m). The water saturation of the flushing belt is expressed as v / v. Water saturation, v / v; a , b , m , n These are stratigraphic factor coefficients, lithological coefficients, cementation values, and saturation index; i The effective porosity, v / v, is the value of each electrical imaging button electrode curve; using e The alternative effective porosity is determined using conventional well logging curves; R i The resistivity of the button electrode for electro-imaging is ohm m; Step 4: Correct the pore fluid effect on the original electro-imaging data based on the pore spectrum, and then perform image enhancement to map the image from the chromaticity value range of 0-255 to the range of 0-100. Then, re-perform probability density statistics on the pure mineral pixel values after correction of all pixels to generate pure mineral pixel waveform spectrum, and then obtain the imaging mineral probability map by equalization processing. The specific process of image enhancement is as follows: First, the probability of a pixel appearing in each pixel value range is calculated, and a probability histogram is plotted (in this probability density histogram, the horizontal axis represents the range of pixel values, the vertical axis represents the probability density corresponding to each pixel value, and the area enclosed by the curve and the horizontal axis is the total probability, which is always 1); then, the histogram is adjusted to stretch the originally dense pixel ranges and compress the sparse ranges. The normal distribution function is:
[0022] In the formula: x This refers to the pixel data of the electrode plate; μ , σ It represents the mean and standard deviation, and is dimensionless.
[0023] in μ This determines the location of the center of the data distribution in the graph. σ This determines the steepness of the peaks in the graph; Step 5: Based on the mineral content data obtained from the whole-rock XRD analysis of the core, the imaging mineral spectrum was quantitatively calibrated. By comparing and analyzing the imaging mineral spectrum of different samples from a single well with their corresponding experimental mineral content data, the optimal segmentation algorithm was used to identify and derive the distribution cutoff threshold and pixel value range for different minerals.
[0024] The specific process for setting the distribution cutoff thresholds for different minerals is as follows: Paired data from different depths of the target well in the study area were collected. Each sample included a normalized imaging mineral spectrum feature vector and the volume fraction of minerals measured by XRD at the corresponding depth. The optimal segmentation algorithm was used to optimize the threshold.
[0025] The optimal segmentation algorithm specifically searches for n optimal segmentation points within the numerical space of the imaging mineral spectrum (0–100), thereby dividing the continuous spectral data into n+1 intervals for different minerals.
[0026] The optimal segmentation algorithm uses a step size of 1% to traverse all possible segmentation schemes. For each candidate cutoff value combination, the predicted mineral category divided by the imaging spectrum is compared with the actual mineral category identified by the XRD experiment to calculate the misclassification error. Through the optimization process, the segmentation point that minimizes the misclassification error between the predicted category and the experimental category is finally selected, achieving the best match between the imaging electrical response and the experimental mineral data, thereby determining the distribution cutoff threshold for different minerals.
[0027] Example For details on the practical application effect of the method of this invention in well X-2, please refer to [link / reference]. Figure 12 This figure comprehensively displays the logging lithology (channel 7), static electrical imaging logging images (channel 8), and continuous curves of felsic, calcite, and argillaceous mineral content calculated using the method of this invention (channels 10, 11, and 12). To verify the reliability of the calculation results, the figure also simultaneously marks the discrete core X-ray diffraction (XRD) whole-rock experimental analysis data points at the corresponding depths.
[0028] Threshold verification and optimization: To verify the rationality and optimality of the selected cutoff value, the system compared the computational effects of various candidate combinations. Based on the calibration data of 22 XRD data points from a well in the study area, the accuracy of 9 different cutoff value combinations was evaluated. Quantitative results show that when the calcite cutoff value is 15% and the felsic cutoff value is 70% (e.g., Figure 8 As shown), the mean absolute error (MAE) between the calculated mineral content and the XRD measured value is minimized (e.g., Figure 9 As shown, the MAE for calcite is 2.42%, for felsic materials it is 2.68%, for argillaceous materials it is 3.59%, and the overall MAE is 2.90%. The MAEs for the other combinations are significantly higher, thus confirming that this set of cutoff values has the best interpretable reliability and calculation accuracy. Based on this, the quantitative interpretation limits for the mineral probability maps of this study area are determined to be 15% for calcite and 70% for felsic materials.
[0029] Comparative analysis shows that the continuous mineral content curve calculated by the method of this invention is highly consistent with the longitudinal distribution trend of discrete core experimental data. Quantitative statistical analysis revealed that the correlation coefficients (R) between the calculated felsic, argillaceous, and calcite contents and the core XRD experimental results reached 0.9174, 0.8921, and 0.9058, respectively, confirming the strong correlation between the calculated and measured data. Furthermore, the average relative errors of the above three minerals were controlled within reasonable ranges of 7.24%, 12.93%, and 14.71%, respectively, demonstrating the high-precision predictive ability of the method of this invention for the mineral composition of complex clastic rocks.
[0030] Further analysis of core thin sections showed that the calculation results obtained by this method were consistent with the actual geological characteristics. At 3969 m and 4023 m, the results of this invention showed a high content of calcite (calcite) and relatively low contents of felsic and argillaceous materials, corresponding to lower porosity. Thin sections of rock fragments at this depth showed that the intergranular pores were strongly cemented and filled with calcite, resulting in a significant reduction in porosity. At 4059 m, the results of this invention showed a high content of felsic materials and relatively low contents of calcite and argillaceous materials, corresponding to relatively higher porosity. Thin section observation indicated that the rock composition in this layer was relatively pure, with less cementing material such as calcite.
[0031] This embodiment fully demonstrates that the mineral quantitative characterization method provided by the present invention not only meets the requirements in terms of overall statistical accuracy, but also fully leverages its high vertical resolution advantage, clearly depicting the continuous and subtle changes in mineral content on the wellbore profile, and accurately reflecting the strong heterogeneity of the formation. Particularly in identifying calcite (grayish) cemented layers or interlayers, this method can effectively capture abrupt changes in content, providing a reliable basis for the detailed evaluation of thin interbedded and heterogeneous reservoirs, and playing a crucial supporting role in determining reservoir effectiveness.
[0032] The above description is not intended to limit the present invention in any way. Although the present invention has been disclosed through the above embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some changes or modifications to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall fall within the scope of the present invention.
Claims
1. A quantitative characterization method for mineral components in electrical imaging logging based on porosity spectrum correction, characterized in that, Includes the following steps: Step 1: Preprocess the raw electrical imaging logging data and generate a high-density pixel image; Step 2: Perform probability density statistical analysis on the pixel values at each depth point in the high-density pixel image to obtain the pixel waveform spectrum; Step 3: Calculate the apparent porosity of each pixel in the high-density pixel image to generate a porosity spectrum; Step 4: Correct the pore fluid effect on the original electro-imaging data based on the pore spectrum, and then perform image enhancement to map the image from the chromaticity value range of 0-255 to the range of 0-100. Then, re-perform probability density statistics on the pure mineral pixel values after correction of all pixels to generate pure mineral pixel waveform spectrum, and then obtain the imaging mineral probability map by equalization processing. Step 5: Based on the mineral content data obtained from the whole-rock XRD analysis of the core, the imaging mineral spectrum was quantitatively calibrated. By comparing and analyzing the imaging mineral spectrum of different samples from a single well with their corresponding experimental mineral content data, the optimal segmentation algorithm was used to identify and derive the distribution cutoff threshold and pixel value range for different minerals.
2. The method for quantitative characterization of mineral components in electrical imaging logging based on porosity spectrum correction according to claim 1, characterized in that, The preprocessing in step 1 includes normalization, velocity correction, image equalization, removal of bad resistivity values, and depth correction for natural gamma logging.
3. The method for quantitative characterization of mineral components in electrical imaging logging based on porosity spectrum correction according to claim 1, characterized in that, The process of generating the high-density pixel image is as follows: the pre-processed effective conductivity curve data is calibrated into two-dimensional color mark data according to different geological application requirements, mapped into a specific imaging color spectrum, and different resistivity values are calibrated into pixel data of different colors, thereby generating a visually intuitive high-density pixel image.
4. The method for quantitative characterization of mineral components in electrical imaging logging based on porosity spectrum correction according to claim 1, characterized in that, In step 2, a statistical method is used to calculate the probability density of the pixel value at each depth point.
5. The method for quantitative characterization of mineral components in electrical imaging logging based on porosity spectrum correction according to claim 1, characterized in that, In step 3, a specific image window is selected, and the porosity value of each imaging logging pixel is calculated according to the Archie formula. All pixels are traversed to generate a porosity data volume that completely corresponds to the spatial position of the original resistivity image, i.e., the porosity spectrum is generated.
6. The method for quantitative characterization of mineral components in electrical imaging logging based on porosity spectrum correction according to claim 1, characterized in that, The specific process of image enhancement in step 4 is as follows: First, the probability of a pixel appearing in each pixel value interval is calculated, and a probability histogram is drawn; then, the histogram is adjusted so that the originally dense pixel intervals are stretched and the sparse intervals are compressed.
7. The method for quantitative characterization of mineral components in electrical imaging logging based on porosity spectrum correction according to claim 1, characterized in that, The specific process for determining the distribution cutoff threshold of different minerals in step 5 is as follows: collect paired data of different depth points of the target well in the study area. Each sample includes a normalized imaging mineral spectrum feature vector and the volume fraction of the XRD measured minerals at the corresponding depth point. The optimal segmentation algorithm is used to optimize the threshold.
8. The method for quantitative characterization of mineral components in electrical imaging logging based on porosity spectrum correction according to claim 7, characterized in that, The optimal segmentation algorithm specifically searches for n optimal segmentation points within the numerical space of the imaging mineral spectrum (0–100), thereby dividing the continuous spectral data into n+1 intervals for different minerals.
9. The method for quantitative characterization of mineral components in electrical imaging logging based on porosity spectrum correction according to claim 8, characterized in that, The optimal segmentation algorithm uses a step size of 1% to traverse all possible segmentation schemes. For each candidate cutoff value combination, the predicted mineral category divided by the imaging spectrum is compared with the actual mineral category identified by the XRD experiment to calculate the misclassification error. Through the optimization process, the segmentation point that minimizes the misclassification error between the predicted category and the experimental category is finally selected, achieving the best match between the imaging electrical response and the experimental mineral data, thereby determining the distribution cutoff threshold for different minerals.