A method and apparatus for calculating oil saturation in shale.

Abstract

This application provides a method and apparatus for calculating shale oil saturation. A learning dataset is established using measured shale core analysis and geochemical analysis data, well logging data, and well logging interpretation data. This learning dataset is then sent to a machine learning algorithm model to obtain data learning results. Next, vitrinite reflectance and geochemical parameters are obtained from the core analysis and geochemical analysis data. The vitrinite reflectance is used to constrain the geochemical parameters to obtain their maximum values. Based on these maximum values, the data learning results are corrected to obtain the final learning result. Finally, the final learning result is fitted using sensitive parameters to obtain the shale oil saturation, thus solving the problems of high data acquisition costs and large calculation errors in existing shale oil saturation calculation methods.

Description

Technical Field

[0001] This application relates to the field of shale oil exploration and development technology, and in particular to a method and apparatus for calculating the oil saturation of shale. Background Technology

[0002] In recent years, the exploration and development of shale oil has been widely carried out both domestically and internationally. The oil saturation of shale is a key factor determining the economic viability of shale oil exploration and development. Therefore, the calculation of shale oil saturation has become a current research hotspot.

[0003] In existing technologies, conventional logging techniques commonly use Archie's formula to calculate shale oil saturation. Nuclear magnetic resonance (NMR) logging technology mainly calculates shale oil saturation using two methods: one is to convert the relaxation spectrum of NMR logging into a pseudo-capillary pressure curve, and calculate the oil saturation at each point given the oil-water interface and oil-water density difference; the other method is to assume a sufficiently large reservoir height and estimate the oil saturation using the bound water saturation obtained from NMR logging.

[0004] However, shale oil reservoirs have complex structures, making it difficult to determine the parameters of Archie's formula. While nuclear magnetic resonance (NMR) logging technology improves the accuracy of shale oil saturation calculations compared to conventional logging techniques, the data acquisition cost of NMR logging is far higher than that of conventional logging. Calculating shale oil saturation remains a critical problem that urgently needs to be solved in shale oil exploration and development. Summary of the Invention

[0005] This application provides a method and apparatus for calculating the oil saturation of shale, in order to solve the problems of high data acquisition cost and large calculation error in existing methods for calculating the oil saturation of shale.

[0006] On the one hand, this application provides a method for calculating the oil saturation of shale, including:

[0007] Collect and acquire core analysis and geochemical analysis data, well logging data and well logging interpretation data of shale;

[0008] Based on the core analysis and geochemical analysis data, well logging data and well logging interpretation data, a learning dataset is established, and the learning dataset is sent to a machine learning algorithm model to obtain data learning results;

[0009] Vitrin reflectance and geochemical parameters are obtained from the core analysis and geochemical analysis data, and the geochemical parameters are constrained by the vitrin reflectance to obtain the maximum value of the geochemical parameters.

[0010] The data learning results are corrected based on the maximum value of the geochemical parameters to obtain the final learning result;

[0011] The oil saturation of shale is obtained by fitting the sensitive parameters with the final learning result.

[0012] On the other hand, this application provides a method and apparatus for calculating the oil saturation of shale, including:

[0013] The acquisition module is used to collect core analysis and geochemical analysis data, well logging data, and well logging interpretation data of shale.

[0014] The analysis module is used to establish a learning dataset based on the core analysis and geochemical analysis data, well logging data and well logging interpretation data, and send the learning dataset to the machine learning algorithm model to obtain data learning results;

[0015] The constraint module is used to obtain vitrinite reflectance and geochemical parameters from the core analysis and geochemical analysis data, and to use the vitrinite reflectance to constrain the geochemical parameters to obtain the maximum value of the geochemical parameters.

[0016] The correction module is used to correct the data learning results based on the maximum value of the geochemical parameters to obtain the final learning result;

[0017] The fitting module is used to fit the final learning result with sensitive parameters to obtain the shale oil saturation.

[0018] This application provides a method and apparatus for calculating shale oil saturation. It calculates shale oil saturation by acquiring core analysis and geochemical analysis data, well logging data, and well logging interpretation data from conventional well logging data, without using nuclear magnetic resonance (NMR) well logging data, which currently has high acquisition costs, thus reducing data acquisition costs. Based on the core analysis and geochemical analysis data, well logging data, and well logging interpretation data, a machine learning algorithm model is used to obtain data learning results. These results are then corrected and fitted to obtain the shale oil saturation. Compared to the Archie formula based on conventional well logging data, the calculation method provided in this application allows for the determination of all formula parameters, solving the problem of large errors in calculating shale oil saturation using the Archie formula. Attached Figure Description

[0019] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0020] Figure 1 This is a schematic diagram of the pore structure and fluid distribution structure of the shale layer on which the embodiments of this application are based;

[0021] Figure 2 A flowchart illustrating a method for calculating shale oil saturation provided in this application embodiment;

[0022] Figure 3 A schematic diagram of the signaling interaction for the shale oil saturation calculation method provided in the embodiments of this application;

[0023] Figure 4 Distribution diagram of vitrinite reflectance and geochemical parameters provided for embodiments of this application;

[0024] Figure 5 The scatter plot of the relationship between shale oil saturation and porosity provided in the embodiments of this application;

[0025] Figure 6 The scatter plot of the relationship between shale oil saturation and geochemical parameters provided in the embodiments of this application;

[0026] Figure 7 The scatter plot of the relationship between shale oil saturation and sensitive parameters provided in the embodiments of this application;

[0027] Figure 8 A structural block diagram of the shale oil saturation calculation device provided in the embodiments of this application;

[0028] Figure 9 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0029] The accompanying drawings have illustrated specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to specific embodiments. Detailed Implementation

[0030] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0031] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation portals are provided for users to choose to authorize or refuse.

[0032] In recent years, shale oil exploration and development have been widely carried out both domestically and internationally. Shale oil saturation is a key factor determining the economic viability of shale oil exploration and development; therefore, its calculation has become a current research hotspot. Conventional logging techniques commonly use Archie's formula to calculate shale oil saturation. However, shale oil reservoirs have complex structures, making it difficult to determine the parameters of Archie's formula. Nuclear magnetic resonance (NMR) logging technology mainly estimates oil saturation by converting the relaxation spectrum of NMR logging into a pseudo-capillary pressure curve, or by using the bound water saturation obtained from NMR logging. Compared to conventional logging techniques, NMR logging technology improves the accuracy of shale oil saturation calculation, but the data acquisition cost of NMR logging is much higher than that of conventional logging. Therefore, the calculation of shale oil saturation remains a critical problem that urgently needs to be solved in shale oil exploration and development.

[0033] Figure 1 This is a schematic diagram of the pore structure and fluid distribution structure of the shale layer on which the embodiments of this application are based. See also: Figure 1 As shown, the mineral composition of shale layers is relatively complex, typically including clay (102), carbonate minerals, quartz (103), pyrite, etc. Shale oil (101) refers to the petroleum resources contained in shale-dominated strata, including petroleum in the pores and fractures of mudstone and shale, as well as petroleum resources in adjacent and interlayered layers of dense carbonate or clastic rocks within the mudstone and shale strata.

[0034] Considering the deep location and complex structure of shale formations, acquiring and analyzing data using nuclear magnetic resonance logging (NMR) technology is costly. Furthermore, shale has a unique source-reservoir integration characteristic, and simply using Archie's formula based on conventional logging techniques results in significant errors in calculating shale oil saturation. Therefore, this application provides a method and apparatus for calculating shale oil saturation based on conventional logging data. This addresses the problems of high data acquisition costs and large calculation errors in existing shale oil saturation calculation methods.

[0035] Figure 2 A flowchart illustrating a shale oil saturation calculation method provided in this application embodiment is shown below. Figure 2 As shown in the embodiments of this application, the shale oil saturation calculation method includes:

[0036] S201. Collect and obtain core analysis and geochemical analysis data, well logging data and well logging interpretation data of shale.

[0037] Before calculating the oil saturation of shale, researchers use specialized measuring instruments to measure the target shale layer, thereby obtaining core analysis and geochemical analysis data, well logging data, and well logging interpretation data of the target shale.

[0038] The data that professional measuring instruments can measure include: porosity, geochemical parameters, vitrinite reflectance, conventional logging curves, logging interpretation porosity curves, and total organic carbon content.

[0039] Conventional logging curves include: natural gamma, spontaneous potential, borehole diameter, sonic logging, density logging, neutron logging, deep lateral resistivity logging, shallow lateral resistivity logging, and microsphere focused resistivity logging.

[0040] In one implementation, if researchers want to obtain the total organic carbon content of a target shale, they can use a total organic carbon analyzer to measure a sample of the target rock layer, thereby obtaining the total organic carbon content of the target rock layer, for example: 2%.

[0041] In another approach, researchers aiming to obtain the natural gamma ray logging curves of a target shale can use both surface and downhole instruments, such as gamma detectors, amplifier circuits, and high-voltage power supplies, to measure the target rock formation. As the downhole instrument is raised from the bottom up within the well, the natural gamma rays from the rock formation pass through the drilling mud and the instrument casing to reach the detector. The detector converts the received gamma rays into individual electrical pulses, which are then amplified by the downhole amplifier and effectively transmitted to the surface via a cable. The surface instrument receives these electrical pulses from downhole, accumulates them using a counting circuit, and after simple transformation and calibration, continuously records the natural gamma ray logging curves of the target rock formation on the well profile.

[0042] This method acquires core analysis and geochemical analysis data, well logging data, and well logging interpretation data of the target shale using specialized measuring instruments, offering high accuracy and speed. Furthermore, the measuring instruments can be combined with microprocessors to achieve automated measurement, fault diagnosis, recording, data processing, and analysis.

[0043] S202. Based on core analysis and geochemical analysis data, well logging data and well logging interpretation data, establish a learning dataset and send the learning dataset to the machine learning algorithm model to obtain data learning results.

[0044] Geochemical parameters of the target shale are obtained from the acquired core analysis and geochemical analysis data; conventional logging curves of the target shale are obtained from the acquired well logging data and well logging interpretation data. The geochemical parameters and conventional logging curves are packaged together to create a learning dataset, which is then sent to a machine learning algorithm model to obtain the data learning results.

[0045] After obtaining the geochemical parameters and conventional logging curves of the target shale, the measurement data and depth need to be matched according to the depth comparison table to determine the true depth corresponding to the measurement data.

[0046] Conventional logging curves are continuous curves, while geochemical parameters are data points at certain depth intervals. Conventional logging curves have a certain influence on geochemical parameters; therefore, conventional logging curves can be considered as independent variables, and geochemical parameters as dependent variables. The resulting data learning outcome is the geochemical parameters.

[0047] The machine learning algorithm model can be an SVM algorithm model, a decision tree algorithm model, a kernel function algorithm model, an XGBoost algorithm model, or other algorithm models, without any limitation.

[0048] In one implementation, researchers obtain three data points corresponding to depths of 3000 meters, 5000 meters, and 7000 meters for the target rock strata's geochemical parameters. The resulting conventional logging curves for the target rock strata are continuous curves corresponding to depths of 3000 meters to 7000 meters. These three geochemical parameter data points and the conventional logging curves are then fed into a machine learning algorithm model. Through the model's autonomous learning, the geochemical parameter data points corresponding to a depth of 4000 meters can be obtained.

[0049] This method uses a machine learning algorithm model to make the discrete geochemical parameters of the target shale continuous, expanding the amount of geochemical parameter data of the target rock layer and providing data support for the subsequent scatter plot drawing.

[0050] S203. Obtain vitrinite reflectance and geochemical parameters from core analysis and geochemical analysis data, and use vitrinite reflectance to constrain the geochemical parameters to obtain the maximum value of the geochemical parameters.

[0051] A curve is plotted based on the vitrinite reflectance, and a scatter plot is plotted based on the geochemical parameters. When the outer envelopes of the curve and the scatter plot are consistent, a computer can be used to solve the transcendental equations to obtain the parameters of the equation corresponding to the vitrinite reflectance curve, thus obtaining the maximum value of the geochemical parameters. When the outer envelopes of the curve and the scatter plot are inconsistent, the parameters of the equation corresponding to the vitrinite reflectance curve need to be adjusted to make the outer envelopes of the curve and the scatter plot consistent. Then, a computer can be used to solve the transcendental equations to obtain the parameters of the equation corresponding to the vitrinite reflectance curve, thus obtaining the maximum value of the geochemical parameters.

[0052] Analysis based on actual data shows that the magnitude of vitrinite reflectance reflects the maturity of rocks, with a positive correlation between the two. Furthermore, the maximum value of geochemical parameters is related to vitrinite reflectance. When the target shale is in an immature stage, the amount of gas adsorbed by the rock is the primary indicator, and its change with vitrinite reflectance is not significant. Once the target shale enters the oil-generating stage, geochemical parameters increase rapidly with increasing vitrinite reflectance. When the target shale reaches its peak oil generation, the geochemical parameters reach their maximum value. As vitrinite reflectance continues to increase, the oil-generating capacity of the target shale weakens, and the geochemical parameters decrease accordingly.

[0053] Therefore, based on the above analysis, a relationship model can be established between vitrinite reflectance and the maximum value of geochemical parameters. The corresponding functional expression for this relationship model is:

[0054]

[0055] Where S1 is the geochemical parameter, TOC is the measured total organic carbon content of the target rock layer, and Ro is the vitrinite reflectance.

[0056]

[0057] α(Ro) = a × (Ro - Ro′) 2 ×e -(Ro+c) (3)

[0058] β(Ro)=b×(Ro-Ro′) (4)

[0059] Where A is the peak coefficient of geochemical parameter S1, and B is the baseline adsorbed gas volume; both A and B are empirical values ​​related to the target shale. α(Ro) is the function corresponding to the peak coefficient A, and β(Ro) is the function corresponding to the baseline adsorbed gas volume B; both are obtained by fitting measurement data of the target shale. a, b, and c are empirical coefficients, and Ro′ is the vitrinite reflectance of the target shale during its peak oil generation period.

[0060] See Figure 4 As shown, Figure 4 The distribution map of vitrinite reflectance and geochemical parameters provided in this application embodiment shows that when the outer envelope of the curve and the scatter plot are consistent, the transcendental equation can be solved by a computer to obtain A, B, a, b, c and Ro′ corresponding to the above functional relationship expression, thereby calculating the maximum value of the geochemical parameter S1max.

[0061] This method takes into account the heterogeneity of shale oil reservoirs, which means that even logging intervals with similar maturity will have different geochemical parameter values. By calculating different maximum geochemical parameter values ​​using different vitrinite reflectivities, the accuracy of geochemical parameters can be increased to a certain extent.

[0062] S204. Based on the maximum value of the localization parameter, correct the data learning results to obtain the final learning result.

[0063] The data learning result is compared with the maximum geochemical parameter S1max obtained in S203. When the data learning result is greater than S1max, S1max is determined as the final learning result; when the data learning result is less than or equal to S1max, the data learning result is determined as the final learning result.

[0064] In one implementation, the data learning results are as follows: geochemical parameter at a depth of 3000 meters: 5 mg / g; geochemical parameter at a depth of 4000 meters: 4.5 mg / g; geochemical parameter at a depth of 5000 meters: 8 mg / g. The maximum geochemical parameter S1max calculated by S203 is 5.5 mg / g. Since the geochemical parameter at a depth of 3000 meters is less than S1max, the geochemical parameter at a depth of 4000 meters is less than S1max, and the geochemical parameter at a depth of 5000 meters is greater than S1max, the final learning results are: geochemical parameter at a depth of 3000 meters: 5 mg / g; geochemical parameter at a depth of 4000 meters: 4.5 mg / g; geochemical parameter at a depth of 5000 meters: 5.5 mg / g.

[0065] When using measurement data from mature areas to build machine learning algorithm models and calculate geochemical parameters for the entire area, inaccurate calculations of some geochemical parameter values ​​can occur because the measurement data from mature areas cannot reflect well logging information from immature or low-maturity areas. This method uses the maximum values ​​of geochemical parameters to constrain and correct the data learning results obtained from the machine learning algorithm model, effectively replacing unreasonable geochemical parameters.

[0066] S205. Fit the sensitive parameters with the final learning results to obtain the shale oil saturation.

[0067] The sensitive parameter is the product of the porosity obtained from measuring the target shale and the geochemical parameters. A scatter plot is drawn, with the sensitive parameter on the x-axis and the final learning result obtained in S204 on the y-axis, to show the relationship between the sensitive parameter and the final learning result. Based on the scatter plot, a functional relationship between the shale oil saturation and the sensitive parameter is obtained using a computer, thereby calculating the oil saturation of the target shale.

[0068] Porosity represents the target shale's capacity to store oil and gas, while geochemical parameters represent its oil-generating capacity. See also Figure 5 The scatter plot of the relationship between shale oil saturation and porosity provided in the embodiments of this application, Figure 6 The scatter plot of the relationship between shale oil saturation and geochemical parameters provided in this application embodiment shows that oil saturation increases with the increase of porosity and geochemical parameters. However, the data points are scattered on the scatter plot, and the fitting effect of each is not ideal. Therefore, the two can be combined by the scatter plot fitting method to calculate the shale oil saturation.

[0069] See Figure 7 The scatter plot showing the relationship between shale oil saturation and sensitive parameters provided in this application embodiment shows that the functional relationship between shale oil saturation and sensitive parameters, obtained by computer, is as follows:

[0070] So=29.403×ln(POR×S1pre)+7.369 (5)

[0071] Where So is the shale oil saturation, POR is the porosity of the target rock layer measured, and S1pre is the final learning result obtained in S204.

[0072] This method uses conventional well logging curves to calculate the oil saturation of shale, and combines porosity with geochemical parameters to fit the oil saturation of shale, thus solving the problems of high data acquisition costs and large calculation errors in existing methods for calculating shale oil saturation.

[0073] Figure 3 A schematic diagram of the signaling interaction for the shale oil saturation calculation method provided in this application embodiment is shown below. Figure 3 As shown, combined Figure 2 The shale oil saturation calculation method provided in this application includes the following steps:

[0074] S301 After measuring the target shale, the measuring instrument sends the core analysis and geochemical analysis data, well logging data and well logging interpretation data to the measurement server.

[0075] The core analysis and geochemical analysis data include: porosity, geochemical parameters, and vitrinite reflectance.

[0076] The logging data and logging interpretation data include: conventional logging curves, logging interpretation porosity curves, and total organic carbon content.

[0077] The conventional logging curves specifically include: natural gamma, spontaneous potential, borehole diameter, acoustic wave, density, neutron, deep lateral resistivity, shallow lateral resistivity, and microsphere focused resistivity conventional logging curves.

[0078] S302. The measurement server establishes a learning dataset based on core analysis and geochemical analysis data, well logging data, and well logging interpretation data.

[0079] The learning dataset includes geochemical parameters and conventional well logging curves.

[0080] S303, The measurement server sends the learning dataset to the computing server.

[0081] S304. After obtaining the learning dataset, the computing server uses a machine learning algorithm model locally to obtain the data learning results.

[0082] The machine learning algorithm model can be an SVM algorithm model, a decision tree algorithm model, a kernel function algorithm model, an XGBoost algorithm model, or other algorithm models, and there is no limitation here.

[0083] S305. The measurement server sends the vitrinite reflectance to the calculation server.

[0084] S306. The calculation server uses vitrinite reflectance to constrain geochemical parameters and obtains the maximum value of geochemical parameters.

[0085] S307. The computing server corrects the data learning results based on the maximum value of the localization parameters to obtain the final learning result.

[0086] S308 The computing server uses sensitive parameters to fit the final learning results to obtain the shale oil saturation.

[0087] Figure 8 The structural block diagram of the shale oil saturation calculation device provided in the embodiments of this application is shown for ease of explanation, only the parts relevant to the embodiments of this application are shown. See also Figure 8 As shown, the vehicle start control device provided in this application embodiment includes: an acquisition module 801, an analysis module 802, a constraint module 803, a correction module 804, and a fitting module 805.

[0088] The acquisition module 801 is used to acquire core analysis and geochemical analysis data, well logging data and well logging interpretation data of shale.

[0089] Analysis module 802 is used to establish a learning dataset based on core analysis and geochemical analysis data, well logging data and well logging interpretation data, and send the learning dataset to the machine learning algorithm model to obtain data learning results;

[0090] The constraint module 803 is used to obtain vitrinite reflectance and geochemical parameters from core analysis and geochemical analysis data, and to constrain the geochemical parameters using vitrinite reflectance to obtain the maximum value of the geochemical parameters.

[0091] The correction module 804 is used to correct the data learning results based on the maximum value of the localization parameter to obtain the final learning result;

[0092] The fitting module 805 is used to fit the sensitive parameters with the final learning results to obtain the oil saturation of shale.

[0093] The shale oil saturation calculation device provided in this application acquires shale core analysis and geochemical analysis data, well logging data, and well logging interpretation data. Based on these data, a learning dataset is established and sent to a machine learning algorithm model to obtain the data learning results. Vitrin reflectance and geochemical parameters are obtained from the core analysis and geochemical analysis data, and vitrin reflectance is used to constrain the geochemical parameters to obtain their maximum values. The data learning results are then corrected based on these maximum values ​​to obtain the final learning result. Finally, the shale oil saturation is obtained by fitting sensitive parameters with the final learning result, thus solving the problems of high data acquisition costs and large calculation errors in existing shale oil saturation calculation methods.

[0094] Figure 9 See the schematic diagram of the electronic device provided in the embodiments of this application. Figure 9 As shown, the electronic device includes: a memory 901, a processor 902, and a computer program; wherein the computer program is stored in the memory 901 and configured to be executed by the processor 902. Figures 2 to 7 The various steps. Processor 902 is used to implement... Figure 8 Each module.

[0095] The memory 901 and the processor 902 are connected via a bus 903.

[0096] For relevant instructions, please refer to the corresponding text. Figures 2 to 8 The relevant descriptions and effects of the steps in the corresponding embodiments are understood, and will not be elaborated on here.

[0097] This application also provides a computer-readable storage medium including computer code that, when run on a computer, causes the computer to perform actions such as... Figures 2 to 9 The method provided by any of the corresponding implementation methods.

[0098] This application also provides a computer program product, including program code, which, when a computer runs the computer program product, executes as follows: Figures 2 to 9 The method provided by any of the corresponding implementation methods.

[0099] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0100] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A method for calculating the oil saturation of shale, characterized in that, include: Collect and acquire core analysis and geochemical analysis data, well logging data and well logging interpretation data of shale; Based on the core analysis and geochemical analysis data, well logging data and well logging interpretation data, a learning dataset is established, and the learning dataset is sent to a machine learning algorithm model to obtain data learning results; Vitrin reflectance and geochemical parameters are obtained from the core analysis and geochemical analysis data, and the geochemical parameters are constrained by the vitrin reflectance to obtain the maximum value of the geochemical parameters. The data learning results are corrected based on the maximum value of the geochemical parameters to obtain the final learning result; The shale oil saturation is obtained by fitting the final learning result with a sensitive parameter, where the sensitive parameter is the product of porosity and geochemical parameters.

2. The method according to claim 1, characterized in that, The core analysis and geochemical analysis data of the shale specifically include: porosity, geochemical parameters, and vitrinite reflectance.

3. The method according to claim 2, characterized in that, The well logging data and well logging interpretation data specifically include: conventional well logging curves, well logging interpretation porosity curves, and total organic carbon content; The conventional logging curves include: natural gamma, spontaneous potential, borehole diameter, acoustic wave, density, neutron, deep lateral resistivity, shallow lateral resistivity, and microsphere focused resistivity conventional logging curves.

4. The method according to any one of claims 1-3, characterized in that, The learning dataset is established based on the core analysis and geochemical analysis data, well logging data, and well logging interpretation data, including: Geochemical parameters were obtained from the core analysis and geochemical analysis data; Obtain conventional logging curves from the logging data and logging interpretation data; The geochemical parameters and the conventional well logging curves are packaged together to create a learning dataset.

5. The method according to any one of claims 1-3, characterized in that, The process of obtaining vitrinite reflectance and geochemical parameters from the core analysis and geochemical analysis data, and using the vitrinite reflectance to constrain the geochemical parameters to obtain their maximum values, specifically includes: Plot a curve based on the reflectance of the vitrinite; Draw a scatter plot based on the geochemical parameters; Determine whether the outer envelopes of the curve and the scatter plot are consistent; If so, determine the parameters of the equation corresponding to the curve to obtain the maximum value of the geochemical parameter; If not, after determining the parameters of the equation corresponding to the curve, adjust the parameters of the equation corresponding to the curve to make the outer envelope of the curve consistent with that of the scatter plot, determine the parameters of the equation corresponding to the curve after adjustment, and obtain the maximum value of the geodesic parameter.

6. The method according to any one of claims 1-3, characterized in that, The step of correcting the data learning results based on the maximum value of the geochemical parameters to obtain the final learning result specifically includes: Determine whether the data learning result is greater than the maximum value of the localization parameter; If so, determine the maximum value of the geochemical parameters as the final learning result; If not, the data learning result is determined as the final learning result.

7. The method according to any one of claims 1-3, characterized in that, The process of fitting the sensitive parameters with the final learning result to obtain the shale oil saturation specifically includes: Plot a scatter plot showing the relationship between the sensitive parameters and the final learning results; Based on the aforementioned scatter plot, determine the relationship curve between shale oil saturation and sensitive parameters; The functional relationship corresponding to the relationship curve is determined to obtain the oil saturation of shale.

8. A shale oil saturation calculation device, comprising: The acquisition module is used to collect core analysis and geochemical analysis data, well logging data, and well logging interpretation data of shale. The analysis module is used to establish a learning dataset based on the core analysis and geochemical analysis data, well logging data and well logging interpretation data, and send the learning dataset to the machine learning algorithm model to obtain data learning results; The constraint module is used to obtain vitrinite reflectance and geochemical parameters from the core analysis and geochemical analysis data, and to use the vitrinite reflectance to constrain the geochemical parameters to obtain the maximum value of the geochemical parameters. The correction module is used to correct the data learning results based on the maximum value of the geochemical parameters to obtain the final learning result; The fitting module is used to fit the final learning result with a sensitive parameter to obtain the shale oil saturation. The sensitive parameter is the product of porosity and geochemical parameters.

9. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the shale oil saturation calculation method as described in any one of claims 1-7.

Images