Method, device and related products for determining total organic carbon content in lacustrine mixed shale
By performing deep correction of the conventional and elemental logging curves of lake-phase mixed accumulation shale and using prediction model calculations, the problem of low calculation accuracy of total organic carbon content in the existing technology is solved, and higher calculation accuracy and better Accurate reservoir evaluation.
Patent Information
- Application Number
- CN202311836177.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-12-28
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2043-12-28
AI Technical Summary
In the prior art, when calculating the total organic carbon content in lake phase mixed accumulation shale, the ΔlogR method has a complex calculation process and low accuracy, and the Schmoker method cannot meet the reservoir evaluation needs.
By obtaining the conventional logging curve and element logging curve of the target oil and gas well, after depth correction, the logging parameters and element content are input into the trained prediction model to improve the calculation accuracy of the total organic carbon content.
The calculation accuracy of total organic carbon content is improved, the demand for reservoir evaluation is met, and the calculation error caused by depth mismatch is reduced.
Smart Images

Figure CN118014117B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of oil and gas exploration technology, and in particular to a method and device for determining the total organic carbon content in lacustrine mixed shale, and related products. Background Art
[0002] Organic carbon content is one of the key parameters for well logging evaluation of "source rock quality" in unconventional oil and gas reservoirs. The calculation accuracy of shale organic carbon content is related to oil and gas reservoir evaluation, development layer selection and reserve assessment. Therefore, when evaluating oil and gas reservoirs, selecting development layers and assessing reserves, it is necessary to calculate the total organic carbon content in lacustrine mixed shale.
[0003] At present, the ΔlogR method or the Schmoker method is generally used to calculate the total organic carbon content. ΔlogR can be the amplitude difference after the acoustic curve, density curve or neutron curve overlaps with the resistivity curve. However, in the actual calculation process of the ΔlogR method, there are often multiple baselines in a well, and it is often necessary to obtain ΔlogR in sections. The calculation process is relatively complicated, and the thermal variation index is not accurately estimated, resulting in low accuracy in the calculation results of the total organic carbon content. The Schmoker method is aimed at multi-mineral mixed shales, and the total organic carbon content is calculated only through the density curve. The calculation accuracy still cannot meet the needs of reservoir evaluation. Summary of the invention
[0004] The present application provides a method, device and related products for determining the total organic carbon content in lacustrine mixed shale, which are used to solve the technical problems in the prior art that the calculation process of the ΔlogR method is complicated and the calculation result of the total organic carbon content has low accuracy, and the calculation accuracy of the Schmoker method cannot meet the requirements of reservoir evaluation.
[0005] In a first aspect, the present application provides a method for determining the total organic carbon content in lacustrine mixed shale, comprising:
[0006] Obtaining a conventional logging curve and an element logging curve of a target oil and gas well; the conventional logging curve includes at least one curve showing the relationship between a logging parameter and a logging depth, and the element logging curve includes at least one curve showing the relationship between an element content and a logging depth;
[0007] Correcting the logging depth of the element logging curve according to the conventional logging curve to obtain a corrected element logging curve; the corrected element logging curve matches the conventional logging curve at each logging depth;
[0008] Obtaining at least one logging parameter and at least one element content at multiple same target logging depths in the conventional logging curve and the corrected element logging curve; the target logging depth is the depth at which the total organic carbon content is to be determined;
[0009] At least one logging parameter and at least one element content at multiple same target logging depths are input into a prediction model that has been trained to convergence, and the prediction model that has been trained to convergence is used to predict and output the total organic carbon content of multiple target logging depths.
[0010] In a possible design, the conventional logging curve and the element logging curve both include corresponding natural gamma curves, and the natural gamma curves represent the relationship between the natural gamma value and the logging depth;
[0011] The step of correcting the depth of the element logging curve according to the conventional logging curve to obtain a corrected element logging curve includes:
[0012] Determine a plurality of first extreme value points in a first natural gamma curve included in the conventional well logging curve, and determine a plurality of second extreme value points in a second natural gamma curve included in the element well logging curve;
[0013] Aligning a plurality of first extreme value points in the first natural gamma curve with a plurality of second extreme value points in the second natural gamma curve, and determining corresponding measurement depth offsets based on the aligned first extreme value points and second extreme value points;
[0014] The depth of at least one curve of the relationship between element content and logging depth included in the element logging curve is corrected according to each corresponding measurement depth offset to obtain a corrected element logging curve.
[0015] In one possible design, a preset prediction model is trained, including:
[0016] Acquire multiple training samples for training a preset prediction model, each of the training samples comprising: at least one logging parameter corresponding to a measurement depth, at least one element content, and annotated actual total organic carbon content, wherein the actual total organic carbon content is obtained by measuring the total organic carbon content of a core of lacustrine mixed shale;
[0017] Using each of the training samples to train a preset prediction model;
[0018] In response to satisfying a preset model convergence condition, the prediction model satisfying the preset model convergence condition is determined as the prediction model preliminarily trained to convergence.
[0019] In a possible design, after determining the prediction model that meets the preset model convergence condition as the prediction model that is initially trained to convergence, the method further includes:
[0020] Acquire multiple test samples for verifying the prediction model initially trained to convergence, each of the test samples comprising: at least one logging parameter corresponding to the measurement depth, at least one element content, and annotated actual total organic carbon content, wherein the actual total organic carbon content is obtained by measuring the total organic carbon content of the core of the lacustrine mixed shale;
[0021] Using the test sample to verify the prediction model that has been initially trained to convergence;
[0022] In response to the verification being passed, the prediction model that has been initially trained to convergence and has passed the verification is determined as the prediction model that has been trained to convergence.
[0023] In one possible design, the preset prediction model is a neural network model, and the neural network model includes at least two hidden layers.
[0024] In a possible design, after predicting and outputting the total organic carbon content of multiple target logging depths using the prediction model that has been trained to convergence, the method further includes:
[0025] Construct an initial relationship curve between total organic carbon content and target logging depth;
[0026] The initial relationship curve is smoothed using a preset filter window length to obtain a smoothed relationship curve.
[0027] In a possible design, the smoothed relationship curve has the highest correlation with the initial relationship curve.
[0028] In a second aspect, the present application provides a device for determining the total organic carbon content in lacustrine mixed shale, comprising:
[0029] An acquisition module, used to acquire conventional logging curves and element logging curves of a target oil and gas well; the conventional logging curves include at least one curve showing the relationship between a logging parameter and a logging depth, and the element logging curves include at least one curve showing the relationship between an element content and a logging depth;
[0030] A correction module, used for correcting the logging depth of the element logging curve according to the conventional logging curve to obtain a corrected element logging curve; the corrected element logging curve matches the conventional logging curve at each logging depth;
[0031] The acquisition module is further used to acquire at least one logging parameter and at least one element content at multiple same target logging depths in the conventional logging curve and the corrected element logging curve; the target logging depth is the depth of the total organic carbon content to be determined;
[0032] The prediction module is used to input at least one logging parameter and at least one element content at the same target logging depth into a prediction model that has been trained to convergence, and use the prediction model that has been trained to convergence to predict and output the total organic carbon content of multiple target logging depths.
[0033] In a third aspect, the present application provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor;
[0034] The memory stores computer-executable instructions;
[0035] The processor executes the computer-executable instructions stored in the memory to implement the method as described in any one of the first aspects.
[0036] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the method as described in any one of the first aspects.
[0037] In a fifth aspect, the present application provides a computer program product, including a computer program, which implements the method described in the first aspect when executed by a processor.
[0038] The present application provides a method, device and related products for determining the total organic carbon content in lacustrine mixed shale, which are obtained by obtaining conventional logging curves and element logging curves of target oil and gas wells; the conventional logging curves include at least one curve showing the relationship between logging parameters and logging depth, and the element logging curves include at least one curve showing the relationship between element content and logging depth; the logging depth of the element logging curves is corrected according to the conventional logging curves to obtain a corrected element logging curve; the corrected element logging curves match the conventional logging curves at each logging depth; at least one logging parameter and at least one element content at multiple same target logging depths are obtained from the conventional logging curves and the corrected element logging curves; the target logging depth is the depth at which the total organic carbon content is to be determined; at least one logging parameter and at least one element content at multiple same target logging depths are input into a prediction model that has been trained to convergence, and the prediction model that has been trained to convergence is used to predict and output the total organic carbon contents at multiple target logging depths. Since the logging depth of the element logging curve is corrected according to the conventional logging curve, the corrected element logging curve can be matched with the conventional logging curve at each logging depth, so that the logging parameters and element contents at the same depth can correspond one to one, reducing the calculation error caused by depth mismatch. In addition, the total organic carbon content is calculated through the prediction model that has been trained to convergence, and more curves related to the total organic carbon content are obtained. The total organic carbon content is calculated through at least one logging parameter and at least one element content at the same target logging depth, which effectively improves the calculation accuracy of the total organic carbon content. BRIEF DESCRIPTION OF THE DRAWINGS
[0039] 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.
[0040] Figure 1 An application scenario diagram of a method for determining total organic carbon content in lacustrine mixed shale provided in one embodiment of the present application;
[0041] Figure 2 A flow chart of a method for determining the total organic carbon content in lacustrine mixed shale provided in one embodiment of the present application;
[0042] Figure 3 A schematic diagram of the well logging depth correction between the second natural gamma curve and the first natural gamma curve provided in one embodiment of the present application;
[0043] Figure 4 A flow chart of a method for determining the total organic carbon content in lacustrine mixed shale provided in another embodiment of the present application;
[0044] Figure 5A schematic diagram of a prediction model provided in one embodiment of the present application;
[0045] Figure 6 A schematic diagram of a smoothed relationship curve obtained by smoothing an initial relationship curve using a preset filter window length provided in an embodiment of the present application;
[0046] Figure 7 A schematic diagram of a fitting curve between a calculated total organic carbon content and 200 actual total organic carbon contents obtained by smoothing an initial relationship curve using a preset filter window length provided in an embodiment of the present application;
[0047] Figure 8 A schematic diagram of the structure of a device for determining the total organic carbon content in lacustrine mixed shale provided in one embodiment of the present application;
[0048] Fig. 9 A schematic diagram of the structure of an electronic device provided in one embodiment of the present application.
[0049] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION
[0050] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.
[0051] It should be noted that the method, device and related products for determining the total organic carbon content in lacustrine mixed shale of the present application can be used in the field of oil and gas exploration technology, and can also be used in any field other than the field of oil and gas exploration technology. The application field of the method, device and related products for determining the total organic carbon content in lacustrine mixed shale of the present application is not limited.
[0052] In order to clearly understand the technical solution of the present application, the solution of the prior art is first introduced in detail.
[0053] At present, when calculating the total organic carbon content, there are generally two methods to achieve the purpose of calculating the total organic carbon content.
[0054] The first method is the ΔlogR method, which uses porosity logging curves, such as sonic curves, density curves or neutron curves, to overlap the porosity logging curves with the resistivity logging curves in the non-source rock section to determine the baseline, and then determine the thermal variation index through other experiments, and directly convert the ΔlogR into the total organic carbon content, and finally realize the calculation of the total organic carbon content. However, in the actual calculation process of the ΔlogR method, there are often multiple baselines in a well, and it is often necessary to obtain ΔlogR in sections. The calculation process is relatively complicated, and the thermal variation index is not accurately estimated, resulting in low accuracy in the calculation results of the total organic carbon content.
[0055] The second method is the Schmoker method, which calculates the total organic carbon content through the density curve. Since this method only calculates the total organic carbon content through the density curve and does not use other logging curves, this method cannot obtain more information related to the total organic carbon content for multi-mineral mixed shales, resulting in the calculation accuracy still not meeting the requirements of reservoir evaluation.
[0056] Therefore, when facing the technical problems in the prior art, since the prior art cannot obtain more information related to the total organic carbon content, the calculation accuracy still cannot meet the reservoir evaluation requirements. Therefore, at least one conventional logging curve and at least one element logging curve of the target oil and gas well can be obtained, so as to obtain at least one logging parameter and at least one element content at the same target logging depth, and the total organic carbon content of the target logging depth can be calculated by the prediction model that has been trained to convergence, so as to improve the calculation accuracy of the total organic carbon content. At the same time, since the conventional logging curve and the element logging curve are obtained by multiple measurements, there will be a phenomenon of mismatch in the logging depth in the measurement results. The logging depth of the element logging curve can be corrected according to the conventional logging curve, so that the corrected element logging curve matches the conventional logging curve at each logging depth, so as to further improve the calculation accuracy of the total organic carbon content.
[0057] The application scenarios of the embodiments of the present invention are introduced below.
[0058] Figure 1 An application scenario diagram of a method for determining the total organic carbon content in lacustrine mixed shale provided in one embodiment of the present application, such as Figure 1As shown, the specific application scenario of the present application can be the scenario of determining the total organic carbon content during oil and gas reservoir evaluation, development layer selection and reserve assessment. Then the system corresponding to the method for determining the total organic carbon content in lacustrine mixed shale provided by the present application includes: user terminal 1, electronic equipment corresponding to the determination of the total organic carbon content in lacustrine mixed shale, such as server 2 or server cluster. The user sends a total organic carbon content determination request to server 2 at the client of user terminal 1, and can include the access address of conventional logging curves and element logging curves in the total organic carbon content determination request. Server 2 can obtain conventional logging curves and element logging curves according to the access addresses of conventional logging curves and element logging curves, correct the logging depth of the element logging curve according to the conventional logging curve to obtain the corrected element logging curve, and obtain at least one logging parameter and at least one element content at multiple same target logging depths in the conventional logging curve and the corrected element logging curve, input at least one logging parameter and at least one element content at multiple same target logging depths into the prediction model that has been trained to convergence, and use the prediction model that has been trained to convergence to predict and output the total organic carbon content of multiple target logging depths. Server 2 finally sends the total organic carbon content of multiple target logging depths to the user terminal 1 client.
[0059] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.
[0060] Figure 2 A flow chart of a method for determining the total organic carbon content in lacustrine mixed shale provided in an embodiment of the present application, such as Figure 2 As shown, the execution subject of this embodiment is a device for determining the total organic carbon content in lacustrine mixed shale. The device for determining the total organic carbon content in lacustrine mixed shale can be implemented by a computer program, or by a medium storing relevant computer programs, such as a USB flash drive and / or a CD, etc.; or, it can also be implemented by a physical device integrated or installed with relevant computer programs, such as a chip, an electronic device, etc. The electronic device can be a server or a server cluster. The method for determining the total organic carbon content in lacustrine mixed shale provided in this embodiment includes the following steps:
[0061] Step 201, obtaining conventional logging curves and element logging curves of a target oil and gas well; the conventional logging curve includes at least one curve of the relationship between a logging parameter and a logging depth, and the element logging curve includes at least one curve of the relationship between an element content and a logging depth.
[0062] Among them, the target oil and gas well refers to the oil and gas well whose total organic carbon content is to be determined.
[0063] Among them, the conventional logging curve refers to a curve used to represent the relationship between logging parameters and logging depth. Its horizontal axis is the logging parameters and the vertical axis is the logging depth. The logging parameters refer to parameters related to the logging depth, which may specifically include: density, neutron porosity, sonic time difference, and formation resistivity. Therefore, the conventional logging curve may specifically include density curve, neutron curve, sonic wave curve and formation resistivity curve.
[0064] Among them, the element logging curve refers to a curve used to represent the relationship between element content and logging depth, with the horizontal axis being the element content and the vertical axis being the logging depth, wherein the elements may include aluminum, calcium, carbon, gadolinium, iron, magnesium, manganese, nitrogen, phosphorus, potassium, silicon, sodium, sulfur, titanium, etc. The element content refers to the content of each element related to the logging depth, including the content of aluminum, calcium, carbon, gadolinium, iron, magnesium, manganese, nitrogen, phosphorus, potassium, silicon, sodium, sulfur, titanium, etc. Therefore, the element logging curve may include the dry weight curve of the above elements.
[0065] Specifically, in this embodiment, the user sends a total organic carbon content determination request to the server at the client of the user terminal, and the total organic carbon content determination request may include the access addresses of the conventional well logging curve and the element well logging curve. The server may obtain the conventional well logging curve and the element well logging curve of the target oil and gas well according to the access addresses of the conventional well logging curve and the element well logging curve.
[0066] Step 202: Correct the logging depth of the element logging curve according to the conventional logging curve to obtain a corrected element logging curve; the corrected element logging curve matches the conventional logging curve at each logging depth.
[0067] The corrected element logging curve refers to an element logging curve obtained after depth correction that matches the conventional logging curve at each logging depth.
[0068] Specifically, in this embodiment, since the conventional logging curve and the element logging curve are the results of multiple measurements, the depth of the element logging curve can be corrected based on the logging depth in the conventional logging curve. Since the fluctuation characteristics of the conventional logging curve and the element logging curve are basically similar, but the logging depths corresponding to the extreme points are slightly different, the logging depths corresponding to the extreme points in the element logging curve can be adjusted to align the logging depths of the extreme points of the element logging curve with those of the conventional logging curve, thereby obtaining a corrected element logging curve.
[0069] Step 203, obtaining at least one logging parameter and at least one element content at multiple same target logging depths in the conventional logging curve and the corrected element logging curve; the target logging depth is the depth at which the total organic carbon content is to be determined.
[0070] The target logging depth refers to the depth at which the total organic carbon content is to be determined.
[0071] Specifically, in this embodiment, the horizontal axis of the conventional logging curve is the logging parameter, and the vertical axis is the logging depth; the horizontal axis of the element logging curve is the element content, and the vertical axis is the logging depth. And the conventional logging curve includes multiple curves such as density curve, neutron curve, acoustic wave curve and formation resistivity curve, and the element logging curve includes multiple curves such as dry weight curves of elements such as aluminum, calcium, carbon, gadolinium, iron, magnesium, manganese, nitrogen, phosphorus, potassium, silicon, sodium, sulfur, titanium, etc. Therefore, it is possible to obtain multiple logging parameters such as density, neutron porosity, acoustic wave time difference, formation resistivity and the content of elements such as aluminum, calcium, carbon, gadolinium, iron, magnesium, manganese, nitrogen, phosphorus, potassium, silicon, sodium, sulfur, titanium, etc. at the same target logging depth in the conventional logging curve and the corrected element logging curve.
[0072] Step 204: input at least one logging parameter and at least one element content at the same target logging depth into the prediction model that has been trained to convergence, and use the prediction model that has been trained to convergence to predict and output the total organic carbon content at the target logging depths.
[0073] The prediction model refers to a model used to predict the total organic carbon content. The prediction model that has been trained to convergence refers to a prediction model that uses training samples to train the initial prediction model to convergence.
[0074] Among them, the initial prediction model can be a deep learning model, a machine learning model, etc., which is not limited in this embodiment.
[0075] Specifically, in this embodiment, the values corresponding to the obtained logging parameters such as density, neutron porosity, acoustic wave time difference, formation resistivity, etc. at each target logging depth, and the contents of elements such as aluminum, calcium, carbon, gadolinium, iron, magnesium, manganese, nitrogen, phosphorus, potassium, silicon, sodium, sulfur, and titanium are respectively input into the prediction model that has been trained to convergence. The prediction model that has been trained to convergence performs feature extraction, prediction and other operations on these data, and predicts and outputs the total organic carbon content of each target logging depth.
[0076] The method for determining the total organic carbon content in lacustrine mixed shale provided in this embodiment is by obtaining a conventional logging curve and an element logging curve of a target oil and gas well; the conventional logging curve includes a curve showing the relationship between at least one logging parameter and the logging depth, and the element logging curve includes a curve showing the relationship between at least one element content and the logging depth; the logging depth of the element logging curve is corrected according to the conventional logging curve to obtain a corrected element logging curve; the corrected element logging curve matches the conventional logging curve at each logging depth; at least one logging parameter and at least one element content at multiple same target logging depths are obtained from the conventional logging curve and the corrected element logging curve; the target logging depth is the depth at which the total organic carbon content is to be determined; at least one logging parameter and at least one element content at multiple same target logging depths are input into a prediction model that has been trained to convergence, and the prediction model that has been trained to convergence is used to predict and output the total organic carbon contents at multiple target logging depths. Since the logging depth of the element logging curve is corrected according to the conventional logging curve, the corrected element logging curve can match the conventional logging curve at each logging depth, so that the logging parameters and element contents at the same depth can correspond one to one, reducing the calculation error caused by depth mismatch. In addition, the total organic carbon content is calculated through the prediction model that has been trained to convergence, and more curves related to the total organic carbon content are obtained. The total organic carbon content is calculated through at least one logging parameter and at least one element content at the same target logging depth, which effectively improves the calculation accuracy of the total organic carbon content.
[0077] As an optional implementation, based on the above embodiment, in this embodiment, both the conventional logging curve and the element logging curve include a corresponding natural gamma curve, and the natural gamma curve represents a curve showing the relationship between the natural gamma value and the logging depth.
[0078] The natural gamma curve refers to a curve used to represent the relationship between the natural gamma value and the logging depth, wherein the abscissa is the natural gamma value and the ordinate is the logging depth.
[0079] Accordingly, step 202 includes the following steps:
[0080] Step 301: determine a plurality of first extreme value points in a first natural gamma curve included in a conventional logging curve, and determine a plurality of second extreme value points in a second natural gamma curve included in an element logging curve.
[0081] The first natural gamma curve refers to a natural gamma curve measured in a conventional logging curve.
[0082] The first extreme point refers to the extreme point in the first natural gamma curve.
[0083] The second natural gamma curve refers to the natural gamma curve measured in the element logging curve.
[0084] The second extreme point refers to the extreme point in the second natural gamma curve.
[0085] Specifically, in this embodiment, the fluctuation characteristics of the first natural gamma curve included in the conventional logging curve and the second natural gamma curve included in the element logging curve are utilized to extract multiple first extreme points in the first natural gamma curve included in the conventional logging curve and multiple second extreme points in the second natural gamma curve included in the element logging curve.
[0086] Step 302: align a plurality of first extreme value points in the first natural gamma curve with a plurality of second extreme value points in the second natural gamma curve, and determine corresponding measurement depth offsets based on the aligned first extreme value points and second extreme value points.
[0087] The depth offset refers to the deviation of the logging depth between the first extreme point and the second extreme point at the corresponding logging depth, and can be expressed by the difference in logging depth between the first extreme point and the second extreme point at the corresponding logging depth.
[0088] Specifically, in this embodiment, illustratively, Figure 3 A schematic diagram of the well logging depth correction between the second natural gamma curve and the first natural gamma curve provided in an embodiment of the present application is shown in FIG. Figure 3 As shown, the horizontal axis is the natural gamma value, the vertical axis is the logging depth, and from left to right are respectively a schematic diagram of the second natural gamma curve, a schematic diagram of the first natural gamma curve, a schematic diagram of the comparison between the corrected second natural gamma curve and the first natural gamma curve, and a schematic diagram of the comparison between the second natural gamma curve and the corrected second natural gamma curve. Specifically, taking the first natural gamma curve measured in the conventional logging curve as a reference, for each logging depth, by adjusting the logging depth of the second extreme point on the corresponding logging depth in the element logging curve, the second extreme point in the second natural gamma curve measured in the element logging curve is aligned with the first extreme point in the first natural gamma curve measured in the conventional logging curve, and the logging depth corresponding to the second extreme point is subtracted from the logging depth corresponding to the first extreme point, so as to obtain the corresponding measured depth offset, and the measured depth offset corresponding to each logging depth is saved in a table to form a depth correction table.
[0089] Step 303: correct the depth of at least one element content-logging depth relationship curve included in the element logging curve according to each corresponding measurement depth offset to obtain a corrected element logging curve.
[0090] Specifically, in this embodiment, according to the depth correction table, the logging depth of at least one element content and logging depth relationship curve included in the element logging curve is summed with the corresponding measurement depth offset in the depth correction table to obtain the corrected logging depth, and the corrected element logging curves are formed by the element logging curves corresponding to the corrected logging depth.
[0091] The method for determining the total organic carbon content in lacustrine mixed shale provided in this embodiment, both the conventional logging curve and the element logging curve include corresponding natural gamma curves, and the natural gamma curve represents the curve of the relationship between the natural gamma value and the logging depth. Accordingly, when the depth of the element logging curve is corrected according to the conventional logging curve to obtain the corrected element logging curve, multiple first extreme value points in the first natural gamma curve included in the conventional logging curve are determined, and multiple second extreme value points in the second natural gamma curve included in the element logging curve are determined; the multiple first extreme value points in the first natural gamma curve are aligned with the multiple second extreme value points in the second natural gamma curve, and the corresponding measurement depth offset is determined based on the aligned first extreme value points and the second extreme value points; the depth of the curve of the relationship between at least one element content and the logging depth included in the element logging curve is corrected according to each corresponding measurement depth offset to obtain the corrected element logging curve. Since both conventional logging curves and element logging curves include natural gamma curves, logging depth correction can be performed on the natural gamma curves in the conventional logging curves and the natural gamma curves in the element logging curves, multiple first extreme points in the first natural gamma curve are aligned with multiple second extreme points in the second natural gamma curve, and the corresponding measurement depth offset is determined based on the aligned first extreme points and the second extreme points. The measurement depths corresponding to the extreme points in the first natural gamma curve and the second natural gamma curve can be determined more accurately and conveniently through the extreme points, so that the corresponding measurement depth offset can be quickly determined, and the logging depth correction can be performed on other logging curves in the element logging curve according to the corresponding measurement depth offset determined by the natural gamma curves included in the conventional logging curves and the element logging curves, so as to achieve correction between the logging depths of each element logging curve and each conventional logging curve, thereby solving the problem of mismatch between the conventional logging curve and the element logging curve at each logging depth.
[0092] Figure 4 A flow chart of a method for determining the total organic carbon content in lacustrine mixed shale provided in another embodiment of the present application is as follows: Figure 4 As shown, in this embodiment, training the preset prediction model includes the following steps:
[0093] Step 401, obtain multiple training samples for training a preset prediction model, each training sample includes: at least one logging parameter corresponding to the measurement depth, at least one element content, and annotated actual total organic carbon content, and the actual total organic carbon content is obtained by measuring the total organic carbon content of the core of the lacustrine mixed shale.
[0094] Among them, the training samples refer to the samples for training the preset prediction model, including at least one logging parameter corresponding to the measurement depth, at least one element content, and the marked actual total organic carbon content.
[0095] The actual total organic carbon content refers to the total organic carbon content obtained by measuring the core of the lacustrine mixed shale. For example, if the drilling depth is 500m, a core with a depth of 20m is taken out for measurement to obtain the actual total organic carbon content.
[0096] Specifically, exemplarily, in this embodiment, if there are 300 samples of actual total organic carbon content obtained by measuring the core of lacustrine mixed shale, the 300 actual total organic carbon contents are sampled through Excel. Specifically, the 300 actual total organic carbon contents measured at different logging depths are imported into a newly created Excel table, and a column of corresponding random numbers is established next to the 300 actual total organic carbon contents measured at different logging depths using the RAND() function, and the 300 actual total organic carbon contents are sampled using the INDEX(,RANK()) function, 200 actual total organic carbon contents are selected as training samples to train the preset prediction model, and 100 actual total organic carbon contents are selected as test samples to verify the prediction model that has been initially trained to convergence. At the same time, through the conditional selection function of Excel, at least one logging parameter and at least one element content at the same logging depth as the 200 actual total organic carbon contents in the training sample are selected from the conventional logging curves and the corrected element logging curves of the target oil and gas wells as training samples to train the preset prediction model.
[0097] For example, if 300 actual total organic carbon contents measured at different logging depths are located in column A in the table, let column B = RAND(), the first row of column C = INDEX($A$1:$A$300, RANK($B$1,$B$1:$B$300)), the second row of column C = INDEX($A$1:$A$300, RANK($B$2,$B$1:$B$300)), then column C is the extracted sample, and so on. 200 actual total organic carbon contents are extracted as training samples to train the preset prediction model, and 100 actual total organic carbon contents are extracted as test samples to verify the prediction model that has been initially trained to convergence.
[0098] It can be understood that by setting column B = RAND(), a set of random numbers ranging from 0 to 1 can be generated in column B.
[0099] Here, the RAND function refers to a function used to generate random numbers.
[0100] The RANK function is a function used for ranking.
[0101] Among them, the INDEX function refers to the function used for reference.
[0102] Step 402: Use each training sample to train the preset prediction model.
[0103] Specifically, in this embodiment, at least one logging parameter and at least one element content corresponding to the measurement depth in the conventional logging curve of the target oil and gas well and the corrected element logging curve are used as input samples of the preset prediction model, and 200 actual total organic carbon contents corresponding to the measurement depth are used as output samples of the preset prediction model to train the preset prediction model.
[0104] Step 403, in response to satisfying a preset model convergence condition, determining the prediction model satisfying the preset model convergence condition as the prediction model preliminarily trained to convergence.
[0105] Among them, the preset model convergence condition can be that the error curve between the output result of the prediction model and the output sample tends to be stable, or the number of iterations reaches a preset number of iterations, etc., which is not limited in this embodiment.
[0106] Specifically, in this embodiment, a preset prediction model is used to perform multiple data processing on each training sample, so that the preset prediction model is trained and learned until the error curve between the output results of the preset model tends to be stable. At this time, the preset prediction model meets the preset model convergence condition. At this time, the prediction model is a prediction model that has been initially trained to convergence, and this preset prediction model is saved.
[0107] The method for determining the total organic carbon content in lacustrine mixed shale provided in this embodiment trains a preset prediction model, including: obtaining multiple training samples for training the preset prediction model, each training sample including: at least one logging parameter corresponding to the measurement depth, at least one element content, and annotated actual total organic carbon content, the actual total organic carbon content being obtained by measuring the total organic carbon content of the core of the lacustrine mixed shale; using each training sample to train the preset prediction model; in response to satisfying the convergence condition of the preset model, determining the prediction model that satisfies the convergence condition of the preset model as the prediction model that has been initially trained to convergence. By training and learning the preset prediction model multiple times, the preset prediction model can be made more accurate, thereby improving the calculation accuracy of the organic carbon content by the preset prediction model.
[0108] As an optional implementation manner, based on the above embodiment, in this embodiment, after step 403, the following steps are further included:
[0109] Step 501, obtaining multiple test samples for verifying the prediction model that has been initially trained to convergence, each test sample includes: at least one logging parameter corresponding to the measurement depth, at least one element content, and annotated actual total organic carbon content, and the actual total organic carbon content is obtained by measuring the total organic carbon content of the core of the lacustrine mixed shale.
[0110] The test sample refers to the sample used to verify the prediction model that has been initially trained to convergence, including at least one logging parameter corresponding to the measurement depth, at least one element content, and the actual total organic carbon content marked.
[0111] Specifically, in this embodiment, the specific implementation method of obtaining multiple test samples for verifying the prediction model preliminarily trained to convergence is the same as the specific implementation method of step 401, which will not be repeated here.
[0112] Step 502: Use test samples to verify the prediction model that has been initially trained to convergence.
[0113] Specifically, in this embodiment, at least one logging parameter and at least one element content input corresponding to the measurement depth in the conventional logging curve of the target oil and gas well and the corrected element logging curve are input as test samples into the preset prediction model, and the preset prediction model predicts and outputs the total organic carbon content of the corresponding measurement depth. The total organic carbon content of the corresponding measurement depth predicted and output by the prediction model is linearly fitted with 100 actual total organic carbon contents in the test sample to obtain a correlation coefficient, and the preset prediction model is verified by the fitting coefficient. The closer the correlation coefficient is to 1, the more the total organic carbon content of the corresponding measurement depth predicted and output by the prediction model is correlated with the 100 actual total organic carbon contents in the test sample. The total organic carbon content of the corresponding measurement depth predicted and output by the prediction model is linearly fitted with 100 actual total organic carbon contents in the test sample to obtain a correlation coefficient, and the preset prediction model is verified by the fitting coefficient, which improves the credibility of the preset prediction model and ensures the accuracy of the total organic carbon content calculated by the preset prediction model.
[0114] Step 503: In response to the verification being passed, the prediction model that has been initially trained to convergence and has passed the verification is determined as the prediction model that has been trained to convergence.
[0115] The determination of whether the verification is passed may include: determining whether the correlation coefficient is within a preset correlation coefficient range.
[0116] Specifically, exemplarily, in this embodiment, if the correlation coefficient is greater than 0.7, it is considered that the verification has passed, and the prediction model that has been initially trained to convergence and has passed the verification is determined as the prediction model that has been trained to convergence; otherwise, the verification has failed.
[0117] The method for determining the total organic carbon content in lacustrine mixed shale provided in this embodiment, after determining the prediction model that meets the preset model convergence condition as the prediction model that has been initially trained to converge, obtains multiple test samples for verifying the prediction model that has been initially trained to converge, each test sample includes: at least one logging parameter corresponding to the measurement depth, at least one element content, and annotated actual total organic carbon content, and the actual total organic carbon content is obtained by measuring the total organic carbon content of the core of the lacustrine mixed shale; the prediction model that has been initially trained to converge is verified using the test sample; in response to the verification being passed, the prediction model that has been initially trained to converge is determined as the prediction model that has been trained to converge. By using the test sample to verify the prediction model that has been initially trained to converge, the credibility of the preset prediction model is improved, and the accuracy of the total organic carbon content calculated by the preset prediction model is ensured.
[0118] As an optional implementation, in this embodiment, the preset prediction model is a neural network model, and the neural network model includes at least two hidden layers.
[0119] Among them, the neural network model refers to the model used to calculate the total organic carbon content.
[0120] The hidden layer refers to the neuron layer between the input layer and the output layer, which serves as the output of the input layer and the input of the output layer.
[0121] Specifically, exemplarily, in this embodiment, Figure 5 A schematic diagram of a prediction model provided in an embodiment of the present application, such as Figure 5 As shown, the neural network model includes two hidden layers, the first hidden layer may include 48 nodes, the second hidden layer may include 36 nodes, or other numbers of nodes, which are not specifically limited here. At least one logging parameter and at least one element content at the same target logging depth are used as inputs of the neural network model, denoted as x. If there are 18 logging parameters and element contents at the same target logging depth, then x is a 18×1 matrix. Then the first hidden layer can be expressed as in, is the first hidden layer, which is a 48×1 matrix. x is an 18×1 matrix consisting of at least one logging parameter and at least one element content at the same target logging depth. w1 is an unknown parameter, which is a 48×18 matrix. f is the activation function. Here, the sigmoid function is selected. As the activation function. The output of the first hidden layer As the input of the second hidden layer, the second hidden layer can be expressed as in, is the second hidden layer, which is a 36×1 matrix, w2 is an unknown parameter, which is a 36×48 matrix, and f is the activation function. Here, the sigmoid function is selected. As the activation function. The output of the second hidden layer As the input of the output layer, the output layer can be expressed as in, is the total organic carbon content of multiple target logging depths predicted and output by the prediction model, w 输出 is an unknown parameter, a 1×36 matrix, and f is the activation function. Here, the sigmoid function is selected. as an activation function.
[0122] In the method for determining the total organic carbon content in lacustrine mixed shale provided in this embodiment, the preset prediction model is a neural network model, and the neural network model includes at least two hidden layers. Since the neural network model includes at least two hidden layers, the input data and the output data can be trained and learned through at least three activation functions, so the calculation accuracy of the neural network model can be improved, and the calculation result of the neural network model can be more accurate.
[0123] As an optional implementation manner, in this embodiment, after step 204, the following steps are further included:
[0124] Step 601, constructing an initial relationship curve between total organic carbon content and target logging depth.
[0125] Specifically, in this embodiment, the total organic carbon content of multiple target logging depths predicted and output by the prediction model that has been trained to convergence and the target logging depth are saved in a newly created table, and an initial relationship curve between the total organic carbon content of multiple target logging depths predicted and output by the prediction model that has been trained to convergence and the target logging depth can be constructed through data processing software such as Origin.
[0126] Step 602: Smoothing the initial relationship curve using a preset filter window length to obtain a smoothed relationship curve.
[0127] The preset filtering window length refers to the window length used for smoothing the initial relationship curve, and may include: 0.25m, 0.5m, 1m, etc.
[0128] The preset filtering window length is determined based on the smoothness of the curve and the correlation between the smoothed relationship curve and the initial relationship curve.
[0129] Specifically, in this embodiment, illustratively, Figure 6 A schematic diagram of a smoothed relationship curve obtained by smoothing an initial relationship curve using a preset filter window length provided in an embodiment of the present application is as follows: Figure 6 As shown in the figure, from left to right, they are the initial relationship curve, the smoothed relationship curve obtained by smoothing the initial relationship curve with a filter window length of 0.25m, the smoothed relationship curve obtained by smoothing the initial relationship curve with a filter window length of 0.5m, and the smoothed relationship curve obtained by smoothing the initial relationship curve with a filter window length of 1.0m. Specifically, since the neural network model often overfits during multiple iterations, the calculation results often have "burr" characteristics when displayed as curves (such as Figure 6 The initial relationship curve is smoothed by the preset filter window length. Users can use filter window lengths of 0.25m, 0.5m, and 1m according to their needs, and use the Gaussian function to perform convolution calculations on the initial relationship curve, thereby reducing the high-frequency components in the initial relationship curve and achieving smooth processing of the initial relationship curve. Figure 6 It can be seen that as the filter window length increases, the high-frequency components in the initial relationship curve gradually decrease, which means that the larger the filter window length, the better the effect of smoothing the initial relationship curve.
[0130] Specifically, exemplarily, Figure 7 A schematic diagram of a fitting curve between the calculated total organic carbon content and 200 actual total organic carbon contents obtained by smoothing the initial relationship curve using a preset filter window length provided in an embodiment of the present application, as shown in FIG. Figure 7 As shown in the figure, the correlation coefficients between the calculated total organic carbon content and the 200 actual total organic carbon contents obtained by smoothing the initial relationship curve using the preset filter window length are 0.8134, 0.7872, 0.7059, and 0.5907, respectively. The larger the correlation coefficient, the more correlated the two are, so the preset filter window length can be determined by the correlation coefficient. Figure 7 It can be seen that with the increase of the filter window length, the correlation coefficient between the calculated total organic carbon content and the 200 actual total organic carbon contents gradually decreases, which means that the larger the filter window length, the smaller the correlation between the calculated total organic carbon content and the 200 actual total organic carbon contents. In summary, users can select a suitable filter window length according to their own needs, so that the selected filter window length can not only meet the effect of smoothing the initial relationship curve, but also ensure that the calculated total organic carbon content has a certain correlation with the 200 actual total organic carbon contents.
[0131] The method for determining the total organic carbon content in lacustrine mixed shale provided in this embodiment predicts and outputs the total organic carbon content of multiple target logging depths using a prediction model that has been trained to convergence, and then constructs an initial relationship curve between the total organic carbon content and the target logging depth; and uses a preset filter window length to smooth the initial relationship curve to obtain a smoothed relationship curve. By using a preset filter window length to smooth the initial relationship curve, the high-frequency component in the initial relationship curve can be reduced, thereby eliminating the "burrs" in the initial relationship curve as much as possible, and providing the user with a smoothed relationship curve that can be used normally.
[0132] As an optional implementation, in this embodiment, the smoothed relationship curve has the highest correlation with the initial relationship curve.
[0133] The correlation refers to a parameter that measures the correlation between the smoothed relationship curve and the initial relationship curve.
[0134] Specifically, in this embodiment, the smoothed relationship curve obtained by using different preset filter window lengths is linearly fitted with the initial relationship curve to obtain different correlation coefficients. The user selects a smoothed relationship curve corresponding to a filter window length that meets his or her needs based on the correlation coefficient.
[0135] In the method for determining the total organic carbon content in the lacustrine mixed shale provided in this embodiment, the smoothed relationship curve has the highest correlation with the initial relationship curve. Since the smoothed relationship curve has the highest correlation with the initial relationship curve, it can be ensured that the smoothed relationship curve is not distorted and the smoothed relationship curve is most correlated with the initial relationship curve, so that it can be used by users.
[0136] Figure 8 A schematic diagram of the structure of a device for determining the total organic carbon content in lacustrine mixed shale provided in an embodiment of the present application is shown in FIG. Figure 8 As shown, the device for determining the total organic carbon content in lacustrine mixed shale provided in this embodiment is located in an electronic device, and the device 700 for determining the total organic carbon content in lacustrine mixed shale provided in this embodiment includes: an acquisition module 701, a correction module 702, and a prediction module 703.
[0137] Among them, the acquisition module 701 is used to obtain the conventional logging curve and the element logging curve of the target oil and gas well; the conventional logging curve includes at least one curve of the relationship between the logging parameter and the logging depth, and the element logging curve includes at least one curve of the relationship between the element content and the logging depth; the correction module 702 is used to correct the logging depth of the element logging curve according to the conventional logging curve to obtain the corrected element logging curve; the corrected element logging curve matches the conventional logging curve at each logging depth; the acquisition module 701 is also used to obtain at least one logging parameter and at least one element content at multiple same target logging depths in the conventional logging curve and the corrected element logging curve; the target logging depth is the depth of the total organic carbon content to be determined; the prediction module 703 is used to input at least one logging parameter and at least one element content at multiple same target logging depths into the prediction model that has been trained to convergence, and use the prediction model that has been trained to convergence to predict and output the total organic carbon content of multiple target logging depths.
[0138] The device for determining the total organic carbon content in lacustrine mixed shale provided in this embodiment can be performed Figure 2 The specific implementation principles and technical effects of the method embodiments shown are similar and will not be repeated here.
[0139] Optionally, both the conventional logging curve and the element logging curve include a corresponding natural gamma curve, and the natural gamma curve represents a curve showing the relationship between the natural gamma value and the logging depth.
[0140] Accordingly, when the correction module 702 corrects the depth of the element logging curve according to the conventional logging curve to obtain the corrected element logging curve, it is specifically used to: determine multiple first extreme points in the first natural gamma curve included in the conventional logging curve, and determine multiple second extreme points in the second natural gamma curve included in the element logging curve; align the multiple first extreme points in the first natural gamma curve with the multiple second extreme points in the second natural gamma curve, and determine the corresponding measurement depth offset based on the aligned first extreme points and the second extreme points; correct the depth of at least one curve of the relationship between element content and logging depth included in the element logging curve according to each corresponding measurement depth offset to obtain the corrected element logging curve.
[0141] Optionally, when training a preset prediction model, the prediction module 703 is specifically used to: obtain multiple training samples for training the preset prediction model, each training sample including: at least one logging parameter corresponding to the measurement depth, at least one element content, and annotated actual total organic carbon content, where the actual total organic carbon content is obtained by measuring the total organic carbon content of the core of the lacustrine mixed shale; use each training sample to train the preset prediction model; in response to meeting the convergence condition of the preset model, determine the prediction model that meets the convergence condition of the preset model as the prediction model that has been initially trained to convergence.
[0142] Optionally, the device for determining the total organic carbon content in lacustrine mixed shale provided in this embodiment further includes: a verification module and a determination module.
[0143] Among them, the acquisition module 701 is also used for obtaining multiple test samples for verifying the prediction model initially trained to convergence after the prediction module 703 determines the prediction model that meets the preset model convergence conditions as the prediction model initially trained to convergence, each test sample includes: at least one logging parameter corresponding to the measurement depth, at least one element content, and annotated actual total organic carbon content, and the actual total organic carbon content is obtained by measuring the total organic carbon content of the core of the lacustrine mixed shale; the verification module is used to verify the prediction model initially trained to convergence using the test sample; the determination module is used to determine the prediction model initially trained to convergence that has passed the verification as the prediction model that has been trained to convergence in response to the verification passing.
[0144] Optionally, the preset prediction model is a neural network model, and the neural network model includes at least two hidden layers.
[0145] Optionally, the device for determining the total organic carbon content in lacustrine mixed shale provided in this embodiment further includes: a construction module and a smoothing module.
[0146] Among them, the construction module is used to construct an initial relationship curve between the total organic carbon content and the target logging depth after the prediction module 703 uses the prediction model that has been trained to convergence to predict and output the total organic carbon content of multiple target logging depths; the smoothing module is used to smooth the initial relationship curve using a preset filter window length to obtain a smoothed relationship curve.
[0147] Optionally, the smoothed relationship curve has the greatest correlation with the initial relationship curve.
[0148] The device for determining the total organic carbon content in lacustrine mixed shale provided in this embodiment can be performed Figure 4 The specific implementation principles and technical effects of the method embodiments shown are similar and will not be repeated here.
[0149] Fig. 9 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application is shown in FIG. Fig. 9 As shown, the electronic device 800 provided in this embodiment includes: a processor 801 and a memory 802 communicatively connected to the processor.
[0150] The memory 802 stores computer-executable instructions; the processor 801 executes the computer-executable instructions stored in the memory 802 to implement the method for determining the total organic carbon content in lacustrine mixed shale provided in any of the above embodiments. The relevant descriptions can be understood by referring to the relevant descriptions and effects corresponding to the steps in the accompanying drawings, and no further details are given here.
[0151] The program may include program code, and the program code includes computer-executable instructions. The memory 802 may include a high-speed RAM memory, and may also include a non-volatile memory, such as at least one disk memory.
[0152] In this embodiment, the memory 802 is connected to the processor 801 via a bus. The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig. 9 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.
[0153] The embodiment of the present application also provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed by a processor, they are used to implement the method for determining the total organic carbon content in lacustrine mixed shale provided in any of the above embodiments. For example, the computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.
[0154] An embodiment of the present application also provides a computer program product, including a computer program, which, when executed by a processor, implements the method for determining the total organic carbon content in lacustrine mixed shale provided in any of the above embodiments.
[0155] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required by the present application.
[0156] It should be further noted that, although the various steps in the flowchart are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowchart may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.
[0157] It should be understood that the above-mentioned device embodiments are only illustrative, and the device of the present application can also be implemented in other ways. For example, the division of units / modules in the above-mentioned embodiments is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units, modules or components can be combined, or can be integrated into another system, or some features can be ignored or not executed.
[0158] In addition, unless otherwise specified, each functional unit / module in each embodiment of the present application may be integrated into one unit / module, each unit / module may exist physically separately, or two or more units / modules may be integrated together. The above-mentioned integrated unit / module may be implemented in the form of hardware or in the form of a software program module.
[0159] If the integrated unit / module is implemented in the form of hardware, the hardware may be a digital circuit, an analog circuit, etc. The physical implementation of the hardware structure includes but is not limited to transistors, memristors, etc. If not specifically stated, the processor may be any appropriate hardware processor, such as a CPU, a GPU, an FPGA, a DSP, an ASIC, etc. If not specifically stated, the storage unit may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random-Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high-bandwidth memory HBM (High-Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc.
[0160] If the integrated unit / module is implemented in the form of a software program module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or all or part of the technical solution, can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.
[0161] In the above embodiments, the description of each embodiment has its own emphasis. For parts not described in detail in a certain embodiment, please refer to the relevant description of other embodiments. The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0162] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.
[0163] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.
Claims
1. A method for determining the total organic carbon content in lacustrine mixed shale, It is characterized in that include: Obtaining conventional logging curves and element logging curves of the target oil and gas well; the conventional logging curves include at least one curve showing the relationship between logging parameters and logging depth, and the element logging curves include at least one curve showing the relationship between element content and logging depth; both the conventional logging curves and the element logging curves include corresponding natural gamma curves, and the natural gamma curves represent the relationship between natural gamma values and logging depth; the logging parameters include at least one of density, neutron porosity, acoustic wave time difference, and formation resistivity; Determine a plurality of first extreme value points in a first natural gamma curve included in the conventional well logging curve, and determine a plurality of second extreme value points in a second natural gamma curve included in the element well logging curve; Aligning a plurality of first extreme value points in the first natural gamma curve with a plurality of second extreme value points in the second natural gamma curve, and determining corresponding measurement depth offsets based on the aligned first extreme value points and second extreme value points; Correcting the depth of at least one curve of the relationship between element content and logging depth included in the element logging curve according to each corresponding measurement depth offset to obtain a corrected element logging curve; Obtaining at least one logging parameter and at least one element content at multiple same target logging depths in the conventional logging curve and the corrected element logging curve; the target logging depth is the depth at which the total organic carbon content is to be determined; Inputting at least one well logging parameter and at least one element content at a plurality of the same target well logging depths into a prediction model that has been trained to convergence, and using the prediction model that has been trained to convergence to predict and output the total organic carbon content at the plurality of target well logging depths; Methods for obtaining the prediction model that has been trained to convergence include: A plurality of training samples for training a preset prediction model are obtained, each of the training samples comprising: at least one logging parameter corresponding to a measurement depth, at least one element content, and annotated actual total organic carbon content, wherein the actual total organic carbon content is obtained by measuring the total organic carbon content of a core of a lacustrine mixed shale; wherein the preset prediction model is a neural network model, the neural network model comprises at least two hidden layers, the at least two hidden layers comprise a first hidden layer and a second hidden layer, the first hidden layer comprises N nodes, the second hidden layer comprises M nodes, and the first hidden layer is represented as: , in, is the first hidden layer, which is an N×1 matrix. is a K×1 matrix consisting of at least one logging parameter and at least one element content at the same target logging depth, w 1 is a matrix in N×K format, f is an activation function, and the activation function is a sigmoid function ; The second hidden layer is expressed as: , in, is the second hidden layer, which is an M×1 matrix. w 2 is a matrix of M×N format; N, M, and K are all integers greater than or equal to 1; Using each of the training samples to train the preset prediction model; In response to satisfying a preset model convergence condition, determining the prediction model satisfying the preset model convergence condition as the prediction model preliminarily trained to convergence; Acquire multiple test samples for verifying the prediction model initially trained to convergence, each of the test samples comprising: at least one logging parameter corresponding to the measurement depth, at least one element content, and annotated actual total organic carbon content, wherein the actual total organic carbon content is obtained by measuring the total organic carbon content of the core of the lacustrine mixed shale; Using the test sample to verify the prediction model that has been initially trained to convergence; In response to the verification being passed, the prediction model that has been initially trained to convergence and has passed the verification is determined as the prediction model that has been trained to convergence.
2. The method according to claim 1, It is characterized in that After the prediction model trained to convergence is used to predict and output the total organic carbon content of multiple target logging depths, the method further includes: Construct an initial relationship curve between total organic carbon content and target logging depth; The initial relationship curve is smoothed using a preset filter window length to obtain a smoothed relationship curve.
3. The method according to claim 2, It is characterized in that The smoothed relationship curve has the greatest correlation with the initial relationship curve.
4. A device for determining the total organic carbon content in lacustrine mixed shale, It is characterized in that include: An acquisition module, used to acquire conventional logging curves and element logging curves of a target oil and gas well; the conventional logging curve includes at least one curve showing the relationship between a logging parameter and a logging depth, and the element logging curve includes at least one curve showing the relationship between an element content and a logging depth; both the conventional logging curve and the element logging curve include corresponding natural gamma curves, and the natural gamma curve represents a curve showing the relationship between a natural gamma value and a logging depth; the logging parameters include at least one of density, neutron porosity, acoustic time difference, and formation resistivity; A correction module, used to determine a plurality of first extreme value points in a first natural gamma curve included in the conventional logging curve, and to determine a plurality of second extreme value points in a second natural gamma curve included in the element logging curve; Aligning a plurality of first extreme value points in the first natural gamma curve with a plurality of second extreme value points in the second natural gamma curve, and determining corresponding measurement depth offsets based on the aligned first extreme value points and second extreme value points; Correcting the depth of at least one curve of the relationship between element content and logging depth included in the element logging curve according to each corresponding measurement depth offset to obtain a corrected element logging curve; The acquisition module is further used to acquire at least one logging parameter and at least one element content at multiple same target logging depths in the conventional logging curve and the corrected element logging curve; the target logging depth is the depth of the total organic carbon content to be determined; A prediction module, for inputting at least one well logging parameter and at least one element content at a plurality of the same target well logging depths into a prediction model that has been trained to convergence, and using the prediction model that has been trained to convergence to predict and output the total organic carbon content at the plurality of target well logging depths; The prediction module is specifically used for: A plurality of training samples for training a preset prediction model are obtained, each of the training samples comprising: at least one logging parameter corresponding to a measurement depth, at least one element content, and annotated actual total organic carbon content, wherein the actual total organic carbon content is obtained by measuring the total organic carbon content of a core of a lacustrine mixed shale; wherein the preset prediction model is a neural network model, the neural network model comprises at least two hidden layers, the at least two hidden layers comprise a first hidden layer and a second hidden layer, the first hidden layer comprises N nodes, the second hidden layer comprises M nodes, and the first hidden layer is represented as: , in, is the first hidden layer, which is an N×1 matrix. is a K×1 matrix consisting of at least one logging parameter and at least one element content at the same target logging depth, w 1 is a matrix in N×K format, f is an activation function, and the activation function is a sigmoid function ; The second hidden layer is expressed as: , in, is the second hidden layer, which is an M×1 matrix. w 2 is a matrix of M×N format; N, M, and K are all integers greater than or equal to 1; Using each of the training samples to train the preset prediction model; In response to satisfying a preset model convergence condition, determining the prediction model satisfying the preset model convergence condition as the prediction model preliminarily trained to convergence; The acquisition module is further used to acquire a plurality of test samples for verifying the prediction model initially trained to convergence, each of the test samples comprising: at least one logging parameter corresponding to the measurement depth, at least one element content, and annotated actual total organic carbon content, wherein the actual total organic carbon content is obtained by measuring the total organic carbon content of the core of the lacustrine mixed shale; A verification module, used to verify the prediction model preliminarily trained to convergence using the test sample; The determination module is used to determine the prediction model that has been initially trained to convergence and has passed the verification as the prediction model that has been trained to convergence in response to the verification passing.
5. An electronic device, It is characterized in that include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 3.
6. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 3 when executed by a processor.
7. A computer program product comprising a computer program, It is characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 3 is implemented.
Citation Information
Patent Citations
Method, device and equipment for determining total organic carbon content and storage medium
CN113027432A
Stratum element logging curve reconstruction method and device
CN114075973A
Shale gas total organic carbon prediction method based on deep learning and interpolation regression
CN114943060A
Method, device, equipment and medium for predicting total organic carbon data of shale reservoir
CN116927771A
