LgR model fitting parameter prediction method and shale oil resource quantity calculation method and system

Abstract

The invention belongs to the field of shale oil exploration evaluation, and discloses an LgR method logging source rock evaluation model fitting parameter prediction method and a shale type shale oil resource quantity calculation method and system. The prediction method comprises the following steps: screening mudstone data in a logging curve; determining a plurality of combinations of possible values of fitting parameters K, a and b, and calculating corresponding mudstone baseline parameters; aiming at different combinations of possible values of fitting parameters K, a and b and corresponding mudstone baseline parameters, according to the screened mudstone data, utilizing an LgR method well logging source rock evaluation model to respectively calculate and obtain organic matter abundance fitting values; and based on the organic matter abundance fitting value and the organic matter abundance measured value, evaluating a fitting effect, and screening out a parameter value combination with the optimal fitting effect as a predicted fitting parameter of the well with the organic matter abundance measured value in the work area. According to the method, all possible parameter value combinations can be comprehensively screened, the influence of subjective factors is eliminated, and undetermined parameters are objectively and accurately predicted.

Description

Technical Field

[0001] The present invention belongs to the field of shale oil exploration and evaluation, and particularly relates to a method for predicting fitting parameters of a logging source rock evaluation model by the △LgR method, a method for calculating shale-type shale oil resources and a system therefor. Background Art

[0002] Shale oil is an unconventional oil with the main oil-producing layer being shale. According to the data analysis of the U.S. Energy Information Administration (EIA) and Advanced Resources International (ARI) at the end of 2017, the global shale oil resources are abundant, with a total geological resource of 936.835×10⁸ t and a technically recoverable reserve of 618.47×10⁸ t. Among them, the technically recoverable reserve of shale oil in China is 44.8×10⁸ t. Shale oil is a new hot spot in the exploration and development of unconventional oil and gas resources after shale gas. Shale oil has achieved great success in North America, forming the second growth peak of U.S. crude oil production, greatly alleviating the U.S. energy pressure and making U.S. energy independence a reality. The successful commercial development of U.S. shale oil has profoundly affected the international energy pattern and reshaped the global energy industry. China is rich in shale oil resources, and abundant shale oil resources are developed in the Cretaceous of the Songliao Basin, the Permian of the Junggar Basin, the Triassic of the Ordos Basin, the Paleogene + Neogene of the Bohai Bay and Jianghan Basins, etc. Shale oil is a strategic field for the replacement of Chinese oil.

[0003] The shale oil resources refer to the total amount of shale oil contained in the shale formation. In the early stage of shale oil exploration and development, it is usually necessary to carry out resource evaluation on the research block. At present, the shale oil resource evaluation methods mainly include two categories: static evaluation and dynamic evaluation. The former mainly calculates its resources through static parameters by methods such as the volume / mass method series, the analogy method (abundance method series) and statistical models; the dynamic evaluation mainly calculates its resources by methods such as the material balance method, the decline method and numerical simulation according to the dynamic parameters in the shale oil development process. Since the volume method does not involve the change of dynamic parameters of oil well development, it is often used in the initial stage of exploration and development. Based on the volume model, the commonly calculated parameters include the thickness of shale, the effective area and the oil saturation (or oil content), etc.

[0004] In the calculation process of volume / mass method resources, the oil content value is the core parameter. Geochemical parameters such as chloroform bitumen "A" and pyrolysis S1 can be obtained through core measurement method as the oil content of shale. Although the data obtained by experimental analysis methods are relatively accurate, due to cost and other constraints, it is impossible to take a large number of cores and conduct analysis and testing for each well. Moreover, the oil content data obtained from discrete samples are difficult to accurately reflect the accurate distribution of strongly heterogeneous organic matter in the formation. Therefore, in production, the advantages of well logging data being longitudinally continuous and having high resolution are often utilized to calculate and determine the organic matter abundance profile through well logging data after lithology calibration, so as to make up for the problem of sparse experimental data. Among them, the △LgR method well logging source rock evaluation technology is simple, easy to implement, economical and practical, and is widely used in the evaluation of oil content of source rocks.

[0005] Due to the inherent physical and chemical properties of mud shale, it has certain well logging response characteristics. For rock layers with high organic matter abundance, the acoustic time difference, density well logging and resistivity well logging curves are abnormal. According to the ΔLgR model, a relationship model between well logging curves and measured data is established to obtain longitudinal continuity parameters.

[0006] However, in the actual evaluation work of resources, the following problems exist:

[0007] 1. The evaluation of organic matter abundance in source rocks needs to be realized by means of a fitting model. However, the model generally contains multiple undetermined parameters, which need to be assigned values through experience when used. The fitting effect is greatly affected by the subjectivity of researchers, and it is difficult to comprehensively consider all combinations of undetermined parameter values. The results have poor repeatability and complex operations, which restrict the accuracy of the evaluation results of organic matter content in source rocks;

[0008] 2. In the process of calculating resources, various types of geological data need to be used, and the amount of data is very large. Processes such as integrating data, adjusting fitting parameters and data statistics will occupy a large amount of human resources. The complicated work usually requires several weeks for the entire evaluation cycle, and the work efficiency is low;

[0009] 3. With the deepening of exploration and development, the resource evaluation work in the work area usually needs to be carried out in a rolling manner. Among them, processes such as adjusting fitting parameters and data statistics need to be carried out repeatedly, and the data in the previous evaluation process are difficult to reuse, which greatly limits the work efficiency of rolling evaluation.

[0010] In view of this, the existing technology needs to be further improved. Summary of the Invention

[0011] In view of at least one of the above problems, the present invention provides a method for predicting fitting parameters of a △LgR method logging source rock evaluation model, so as to objectively and accurately predict all undetermined parameters in the △LgR method logging source rock evaluation model (hereinafter also referred to as the △LgR model). The present invention also provides a method and system for calculating shale oil resource volume, so as to make the resource volume evaluation more accurate, convenient and efficient.

[0012] According to a first aspect of the present invention, there is provided a method for predicting fitting parameters of a △LgR method logging source rock evaluation model, comprising the following steps:

[0013] For wells in the work area with measured values of organic matter abundance (the concept of oil content hereinafter is similar to this concept):

[0014] Screen the shale data in the logging curves;

[0015] Determine several combinations of possible values of the fitting parameters K, a, b, and calculate the corresponding shale baseline parameter ΔLgR’ 基线 ;

[0016] For different combinations of possible values of the fitting parameters K, a, b and the corresponding ΔLgR’ 基线 , according to the screened shale data, use the △LgR method logging source rock evaluation model to calculate the fitting values of organic matter abundance respectively;

[0017] Based on the fitting values of organic matter abundance and the measured values of organic matter abundance, evaluate the fitting effect, and screen out the combination of parameter values with the best fitting effect as the fitting parameters of the △LgR model for the wells in the work area with measured values of organic matter abundance predicted;

[0018] For wells in the work area without measured values of organic matter abundance:

[0019] Use the predicted fitting parameters and well location coordinate data of the wells in the work area with measured values of organic matter abundance, and through the spatial interpolation model, predict the fitting parameters of the wells in the work area without measured values of organic matter abundance.

[0020] The specific form of the △LgR method logging source rock evaluation model described in the present invention is:

[0021] Fitting value = a × (LgR + K × Δt - ΔLgR’ 基线 ) + b

[0022] Wherein, R is the resistivity logging data, Δt is the acoustic time difference data, and K, a, b, ΔLgR’ 基线 are fitting parameters.

[0023] According to some embodiments of the present invention, screening the shale data in the logging curves includes: using lithology description data or shale GR logging data to screen the shale data in the logging curves.

[0024] According to some embodiments of the present invention, when screening shale data using lithology description data, if the lithology description of a certain section of well logging curve is shale and the lithology description does not contain sandstone, it is determined that the data of this section of well logging curve is shale data; otherwise, it is determined as non-shale data. When screening shale data using shale GR well logging data, if the well logging GR value of a certain section of well logging curve is higher than the GR value of a normal sandstone section, it is determined that the data of this section of well logging curve is shale data; otherwise, it is determined as non-shale data.

[0025] According to some embodiments of the present invention, when screening shale data using lithology description data, if the lithology description corresponding to a certain section of well logging curve contains any one of the keywords'shale', 'claystone', 'limestone', and does not contain the keyword'sand', it is determined that the data of this section of well logging curve is shale data; otherwise, it is determined as non-shale data. When screening shale data using shale GR well logging data, if the well logging GR value of a certain section of well logging curve is greater than 80 us / f, it is determined that the data of this section of well logging curve is shale data; otherwise, it is determined as non-shale data.

[0026] According to some embodiments of the present invention, determining several combinations of possible values of the fitting parameters K, a, b, including:

[0027] Determining the initial values and value ranges of the fitting parameters K, a, b for the entire area or a local area within the work area;

[0028] Based on the value ranges and value step sizes of the fitting parameters K, a, b, respectively determining several possible values of the fitting parameters K, a, b;

[0029] Based on several possible values of the fitting parameters K, a, b, determining several combinations of possible values of the fitting parameters K, a, b by the method of free permutation and combination.

[0030] According to some embodiments of the present invention, determining the initial values and value ranges of the fitting parameters K, a, b for the entire area or a local area within the work area, including:

[0031] In the case where the background of the entire area within the work area is similar, determining the initial values and value ranges of the fitting parameters K, a, b for the entire area within the work area;

[0032] In the case where the background of the entire area within the work area is not similar, dividing the work area into several local areas, each local area being used as a sub-work area, and respectively determining the initial values and value ranges of the fitting parameters K, a, b,

[0033] wherein the background includes structure, sedimentation process of organic matter, organic matter abundance, and hydrocarbon generation evolution history.

[0034] According to some embodiments of the present invention, determining the initial values and value ranges of the fitting parameters K, a, and b for the entire area or a partial area in the work area includes:

[0035] Performing linear fitting with the acoustic travel time in the selected shale data as the independent variable and the logarithm of the resistivity as the dependent variable, taking the absolute value of the slope after linear fitting as the initial value of the fitting parameter K, and taking the range that fluctuates up and down by no less than 50% based on the initial value of the fitting parameter K as the value range of the fitting parameter K;

[0036] Based on the selected shale data and the fitting parameter K, calculating the corresponding shale baseline parameter ΔLgR'; 基线 ;

[0037] According to the comprehensive initial value of the fitting parameter K and ΔLgR'; 基线 Calculating the ΔLgR value corresponding to the measured value for the well logging data point, and the calculation formula is:

[0038] ΔLgR = LgR + K × Δt - ΔLgR'; 基线

[0039] wherein, R is the resistivity and Δt is the acoustic travel time;

[0040] Performing linear fitting with the ΔLgR value of the well logging data point as the independent variable and the measured value as the dependent variable, taking the fitted slope as the initial value of the fitting parameter a, and taking the range that fluctuates up and down by no less than 50% based on the initial value of the fitting parameter a as the value range of the fitting parameter a;

[0041] Selecting the non-shale data in the measured values, taking the maximum value of the organic matter abundance of the non-shale data as the initial value of the fitting parameter b. If the initial value of the fitting parameter b is higher than the maximum value of the organic matter abundance of the non-source rock, taking the maximum value of the organic matter abundance of the non-source rock as the initial value of the fitting parameter b, and taking the interval [0, b 初值 as the value range of the fitting parameter b.

[0042] According to some embodiments of the present invention, calculating the corresponding shale baseline parameter ΔLgR'; 基线 , including:

[0043] Based on the base factor values at each depth point on the well logging profile, determining the shale baseline;

[0044] For each specific K value, according to the well logging data within the shale baseline, calculating the ΔLgR' 基线 value at each depth point;

[0045] Taking the arithmetic mean of the ΔLgR' 基线 values at each depth point as the shale baseline parameter ΔLgR'; 基 line.

[0046] According to some embodiments of the present invention, calculate the corresponding shale baseline parameter ΔLgR’ 基线 , further comprising:

[0047] Sweep across the logging profile in a depth window of a fixed size, and calculate the basification factor value of each depth point within the depth window. The calculation formula for the basification factor value is:

[0048] ABS((LOG10(R)+1.523)*2.5-(10-Δt / 20)) / 10

[0049] where R is the resistivity and Δt is the acoustic travel time;

[0050] Take the arithmetic mean of the basification factor values of each depth point within each depth window as the basification factor value of the formation within the depth window;

[0051] If the basification factor value of the formation within a certain depth window is within the acceptable range, take this depth section as the shale baseline. If the basification factor values of the formations within all depth windows on the single-well profile exceed the acceptable range, then select several depth points with the smallest basification factor values in the shale of the whole well to form the shale baseline;

[0052] For each specific K value, calculate the ΔLgR’ of each depth point according to the logging data within the shale baseline 基线 value, and the calculation formula is:

[0053] ΔLgR’ 基线 =LgR 基线 +K×Δt 基线

[0054] where R 基线 is the resistivity within the determined shale baseline, and Δt 基线 is the acoustic travel time within the determined shale baseline;

[0055] Take the arithmetic mean of the ΔLgR’ values of each depth point as the shale baseline parameter ΔLgR’ 基线 line. 基 According to some embodiments of the present invention, based on the fitted value and the measured value of the organic matter abundance, evaluate the fitting effect and screen out the parameter value combination with the best fitting effect, including:

[0056] Set the first logging depth floating interval according to the accuracy of the logging data;

[0057] Set the second sampling depth floating interval according to the sampling situation and accuracy of the measured samples;

[0058]

[0059] ​The first logging depth floating interval and the second sampling depth floating interval are added together to form the depth floating interval;

[0060] The organic matter abundance fitting value with the highest matching degree within the depth floating interval corresponding to the measured value of organic matter abundance is taken as the organic matter abundance fitting value at the depth point;

[0061] The correlation coefficient R was calculated using the fitted value of organic matter abundance and the measured value of organic matter abundance. 2 , correlation coefficient R 2 The value of the fitting parameter K when it is the highest is taken as the preferred fitting parameter K;

[0062] Under the condition of the optimal fitting parameter K, the average value of the organic matter abundance fitting value is compared with the average value of the organic matter abundance measured value, and the values ​​of the fitting parameters a and b with the smallest deviation are taken as the optimal fitting parameters a and b;

[0063] The best parameter value combination selected is the preferred fitting parameter K, the preferred fitting parameters a and b, and the mudstone baseline parameter ΔLgR' calculated based on the preferred fitting parameter K. 基线 combination.

[0064] According to some embodiments of the present invention, when using the predicted fitting parameters and well coordinate data of wells with measured values ​​of organic matter abundance in the work area to predict the fitting parameters of wells without measured values ​​of organic matter abundance in the work area, unreliable data in the △LgR model fitting parameters predicted by the wells with measured values ​​of organic matter abundance are eliminated.

[0065] According to some embodiments of the present invention, removing unreliable data in the ΔLgR model fitting parameters predicted by wells with measured values ​​of organic matter abundance includes at least one of the following:

[0066] For the wells with measured values ​​of organic matter abundance in the work area, the fitting parameters of the wells with poor fitting effect and too few measured data are eliminated according to the fitting effect index and the number of measured core data. The fitting effect index includes the fitting correlation coefficient and the degree of deviation from the average value.

[0067] Eliminate abnormal outlier fitting parameters through outlier detection;

[0068] Use the fitting parameters K, a, b, ΔLgR' 基线 The spatial interpolation of the well coordinate data is performed separately, and then a convolution kernel of appropriate size and type is selected for smoothing filtering to remove the mutation noise in the data.

[0069] According to some embodiments of the present invention, removing abnormal outlier fitting parameters through outlier detection includes:

[0070] Calculate the fitting parameters K, a, b, ΔLgR' respectively基线 The average value μ and standard deviation σ;

[0071] Calculate the variable distribution interval (μ - 3σ, μ + 3σ);

[0072] Identify and remove the abnormal data outside the variable distribution interval.

[0073] According to some embodiments of the present invention, the spatial interpolation model uses the Kriging interpolation method, and its theoretical variogram model adopts a linear model.

[0074] According to a second aspect of the present invention, there is provided a method for calculating the shale-type shale oil resource amount, characterized by comprising the following steps:

[0075] Collect the experimental determination data of the oil content rate, logging data, horizon division data, lithologic profile description data, Ro maturity data, oil content rate recovery correction model data, well location coordinate data, and work area boundary data of each single well in the work area;

[0076] Use the method described in the first aspect of the present invention to predict the fitting parameters of the ΔLgR method logging source rock evaluation model for each single well in the work area, and fit the oil content rate profile data of each single well in the work area according to the fitting parameters and logging data;

[0077] According to the fitted oil content rate profile data and horizon division data, perform summary statistics according to stratification and self-defined grading to obtain the formation thickness and formation average oil content rate data of each single well in the work area under each stratification and each grading;

[0078] Calculate the shale-type shale oil resource amount in the work area according to the formation thickness, formation average oil content rate data, horizon division data, Ro maturity data, oil content rate recovery correction model data, well location coordinate data, and work area boundary data of each single well in the work area.

[0079] According to some embodiments of the present invention, calculating the shale-type shale oil resource amount in the work area according to the formation thickness, formation average oil content rate data, horizon division data, Ro maturity data, oil content rate recovery correction model data, well location coordinate data, and work area boundary data of each single well in the work area includes:

[0080] Construct a basic background grid covering the entire work area;

[0081] Calculate the resource amount of each single well in the work area by using the mass oil content rate model according to the formation thickness and formation average oil content rate data of each single well in the work area under each stratification and each grading;

[0082] Based on the resource quantity data and well position coordinate data within each layer and grade in the work area, a resource quantity grid model for each layer and grade based on the basic background grid is established through a spatial interpolation model, and the data of grids with formation thickness and average oil content rate in the formation lower than the industrial development marginal limit value and negative value grids are defined as 0;

[0083] Based on the Ro maturity data and well position coordinate data within each layer in the work area, an Ro grid model based on the basic background grid is established through a spatial interpolation model;

[0084] Based on the Ro grid model, the data within the Ro grid model are sequentially converted into recovery coefficients using an oil content rate recovery correction model to obtain a recovery correction coefficient grid model;

[0085] Based on the resource quantity grid model and the recovery correction coefficient grid model, the data in the corresponding grids of the two models are multiplied to obtain a recovered resource quantity grid model for each layer and grade;

[0086] According to the work area boundary data, it is judged whether each grid in the recovered resource quantity grid model for each layer and grade is within the work area boundary, and the total resource quantity of all grids under each layer and grade within the work area boundary is statistically calculated, and finally the shale-type shale oil resource quantity for each layer and grade within the work area is obtained.

[0087] According to some embodiments of the present invention, when conducting data stratification statistics, if the logging data of a certain well fails to completely cover the top or bottom surface of the layer, the data of this layer is directly abandoned for statistics.

[0088] According to the third aspect of the present invention, a shale-type shale oil resource quantity calculation system is provided, which is characterized in that the system includes:

[0089] A data acquisition module, configured to acquire the experimentally determined data of the sample oil content rate, logging data, layer division data, lithology profile description data, Ro maturity data, oil content rate recovery correction model data, well position coordinate data, and work area boundary data of each single well in the work area;

[0090] A data integration module, configured to summarize the experimentally determined data of the sample oil content rate, logging data, layer division data, and lithology profile description data of each single well together according to the sample depth;

[0091] A fitting parameter prediction module, configured to predict the fitting parameters of the △LgR method logging source rock evaluation model for each single well in the work area using the method described in the first aspect of the present invention;

[0092] An oil content rate profile module, configured to fit the oil content rate profile data of each single well in the work area according to the predicted fitting parameters and logging data;

[0093] The hierarchical and graded statistical module is used to perform summary statistics according to the oil content profile data and horizon division data obtained by fitting, by stratification and custom grading, to obtain the formation thickness and formation average oil content data of each single well in the work area under each stratification and each grading;

[0094] The single-well resource volume calculation module is used to calculate the resource volume of each single well in the work area by using the mass oil content model based on the formation thickness and formation average oil content data of each single well in the work area under each stratification and each grading;

[0095] The resource volume interpolation and prediction module is used to establish a resource volume grid model for each stratification and each grading based on the basic background grid by using the spatial interpolation model with the resource volume data and well location coordinate data in each stratification and grading in the work area as the basis;

[0096] The resource volume restoration and correction module is used to establish an Ro grid model based on the basic background grid by using the spatial interpolation model with the Ro maturity data and well location coordinate data in each stratification in the work area as the basis. Based on the Ro grid model, the data in the Ro grid model is sequentially converted into restoration coefficients by using the oil content restoration and correction model to obtain a restoration correction coefficient grid model. Based on the resource volume grid model and the restoration correction coefficient grid model, the data in the corresponding grids of the two models are multiplied to obtain a restored resource volume grid model for each stratification and each grading;

[0097] The resource volume summary and statistical module is used to determine whether each grid in the restored resource volume grid model for each stratification and each grading is within the work area boundary according to the work area boundary data, and to statistically sum up the resource volumes of all grids under each stratification and each grading within the work area boundary, and finally obtain the shale-type shale oil resource volume for each stratification and each grading in the work area.

[0098] According to some embodiments of the present invention, the fitting parameter prediction module includes:

[0099] The first sub-module is used to predict the fitting parameters of the △LgR model for wells with measured organic matter abundance values in the work area;

[0100] The screening module is used to eliminate unreliable data in the fitting parameters of the △LgR model predicted by the first sub-module;

[0101] The second sub-module is used to predict the fitting parameters of wells without measured organic matter abundance values in the work area by using the predicted fitting parameters and well location coordinate data of wells with measured organic matter abundance values in the work area after eliminating unreliable data through the spatial interpolation model.

[0102] Due to the above technical solutions, the present invention has the following beneficial effects:

[0103] The method for predicting fitting parameters of the ΔLgR method logging source rock evaluation model provided by the present invention screens out the parameter value combination with the best fitting effect by determining several combinations of possible values of the fitting parameters, calculating the fitting values for each combination and evaluating the fitting effect accordingly. It can comprehensively screen all possible parameter value combinations, eliminate the influence of subjective factors on the calculation results of organic matter abundance, objectively and accurately predict all undetermined parameters in the ΔLgR model, and help improve the data prediction accuracy and work efficiency.

[0104] The method for calculating shale-type shale oil resource quantity and the system for calculating shale-type shale oil resource quantity provided by the present invention use the method for predicting fitting parameters of the ΔLgR method logging source rock evaluation model provided by the present invention to predict the fitting parameters of each single well in the work area, and can objectively and accurately predict all undetermined parameters of each single well, improve the data prediction accuracy and work efficiency. In addition, each step of the method for calculating shale-type shale oil resource quantity provided by the present invention can be automatically implemented by a computer program, saving human resources, revolutionarily improving work efficiency, significantly shortening the evaluation cycle, and at the same time helping to efficiently carry out the rolling evaluation of resource quantity. BRIEF DESCRIPTION OF THE DRAWINGS

[0105] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0106] Figure 1 It is a flowchart of the method for predicting fitting parameters of the ΔLgR method logging source rock evaluation model provided by the present invention;

[0107] Figure 2 It is a flowchart of the method for calculating shale-type shale oil resource quantity provided by the present invention;

[0108] Figure 3 It is a flowchart of the method for evaluating shale-type shale oil resource quantity provided by an embodiment of the present invention;

[0109] Figure 4 It is a schematic diagram for evaluating the goodness of fit by correlation coefficient;

[0110] Figure 5 It is a schematic diagram for evaluating the fitting deviation by average value;

[0111] Figure 6 It is a block diagram of the system for calculating shale-type shale oil resource quantity provided by the present invention;

[0112] Figure 7The flowchart of the process executed by the shale-type shale oil resource volume calculation system provided by the present invention. Detailed implementation manners

[0113] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below in conjunction with embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0114] Unless the context clearly requires otherwise, words such as "including" and "comprising" in the entire specification and claims should be interpreted as having an inclusive meaning rather than an exclusive or exhaustive meaning; that is, it is the meaning of "including but not limited to".

[0115] According to the first aspect of the present invention, a method for predicting fitting parameters of a △LgR method logging source rock evaluation model is provided. As Figure 1 shown, the method includes the following steps:

[0116] S1: For wells with measured organic matter abundance values in the work area: screen the shale data in the logging curves; determine several combinations of possible values of the fitting parameters K, a, and b, and calculate the corresponding shale baseline parameter ΔLgR' 基线 ; for different combinations of possible values of the fitting parameters K, a, and b and the corresponding LgR' 基线 , according to the screened shale data, use the △LgR method logging source rock evaluation model to calculate the fitting values of the organic matter abundance respectively; based on the fitting values of the organic matter abundance and the measured values of the organic matter abundance, evaluate the fitting effect, and screen out the combination of parameter values with the best fitting effect as the fitting parameters of the wells with measured organic matter abundance values predicted in the work area.

[0117] The specific form of the △LgR method logging source rock evaluation model described in the present invention is:

[0118] Fitting value = a × (LgR + K × Δt - ΔLgR' 基线 ) + b

[0119] wherein, R is the resistivity logging data, Δt is the acoustic time difference data, and K, a, b, and ΔLgR' 基线 are fitting parameters.

[0120] The "measured value of organic matter abundance" mentioned in the present invention refers to the characterization value of the organic matter abundance obtained by actual measurement means. For example, the measured value of total organic carbon (TOC), the measured value of chloroform bitumen A, and the measured value of rock pyrolysis S1. These measured values can be obtained by methods known in the art, for example, by core measurement method. The "well with measured organic matter abundance value" mentioned in the present invention refers to a well that has obtained the measured value of the organic matter abundance.

[0121] Screen the shale data in the logging curves, including: using lithology description data or shale GR logging data to screen the shale data in the logging curves.

[0122] When using lithology description data to screen shale data, if the lithology description of a certain section of the logging curve is shale and the lithology description does not contain sandstone, it is judged that the data of this section of the logging curve is shale data, otherwise it is judged as non-shale data. Specifically, when using lithology description data to screen shale data, if the lithology description corresponding to a certain section of the logging curve contains any one of the keywords'mudstone','shale', 'limestone', and the lithology description does not contain the keyword'sand', it is judged that the data of this section of the logging curve is shale data, otherwise it is judged as non-shale data.

[0123] When using shale GR logging data to screen shale data, if the logging GR value of a certain section of the logging curve is higher than the GR value of the normal sandstone section, it is judged that the data of this section of the logging curve is shale data, otherwise it is judged as non-shale data. Specifically, when using shale GR logging data to screen shale data, if the logging GR value of a certain section of the logging curve is greater than 80 us / f, it is judged that the data of this section of the logging curve is shale data, otherwise it is judged as non-shale data.

[0124] The four key parameters (K, △LgR’ 基线 , a, b) included in the △LgR model formula are controlled by the geological environment. When the geological environments are similar, the values of each parameter are relatively stable. The parameter △LgR’ 基线 is controlled by the parameter K. Therefore, a broad range can be assumed to completely cover all possible values of K, a, and b, and each parameter is allowed to take values in the selected range in a certain step (precision) in turn, so as to determine several groups of K, a, b parameter combinations.

[0125] Determine several combinations of possible values of the fitting parameters K, a, b, including: determining the initial values of the K, a, b parameters in the work area (or local area within the work area) and the value ranges of the K, a, b parameters; based on the value ranges and value precisions (steps) of the fitting parameters K, a, b, respectively determine several possible values of the fitting parameters K, a, b; based on several possible values of the fitting parameters K, a, b, determine several combinations of possible values of the fitting parameters K, a, b by the free permutation and combination method.

[0126] Specifically, determine the initial values and value ranges of parameters K, a, and b for the work area (or a local area within the work area), including: if the work area is small and the backgrounds such as structure, organic matter deposition process, organic matter abundance, and hydrocarbon generation evolution history are similar, directly determine the initial values and value ranges of parameters K, a, and b for the work area (or a local area within the work area); if the work area is very large and there are obvious differences in backgrounds such as structure, organic matter deposition process, organic matter abundance, and hydrocarbon generation evolution history, the work area should be divided into several local areas, each area being regarded as a sub-work area, and the initial values and value ranges of parameters K, ΔLgR’ 基线 , a, and b are determined respectively; perform a linear fit with the acoustic travel time in the shale logging data as the independent variable and the logarithm of the resistivity as the dependent variable. The absolute value of the slope after the linear fit is used as the initial value of parameter K, and a range that fluctuates up and down by no less than 50% based on the initial value of parameter K is used as the value range of parameter K; based on the selected shale data and the fitting parameter K, calculate the corresponding shale baseline parameter ΔLgR’ 基线 ; according to the initial value of the fitting parameter K and ΔLgR’ 基线 , calculate the ΔLgR value corresponding to the measured value of the logging data point; perform a linear fit with the ΔLgR value of the logging data point as the independent variable and the experimentally measured value as the dependent variable. The fitted slope is used as the initial value of variable a, and a range that fluctuates up and down by no less than 50% based on the initial value of parameter a is used as the value range of parameter a; screen out the non-shale data in the experimentally measured values, and use the maximum value of the organic matter abundance of the non-shale data as the initial value of variable b. If the initial value of variable b is higher than the maximum value of the organic matter abundance of the non-source rock, use the maximum value of the organic matter abundance of the non-source rock as the initial value of variable b, and use the interval [0, b 初始取值 as the value range of variable b.

[0127] Specifically, the calculation formula for the ΔLgR value is:

[0128] ΔLgR = LgR + K×Δt - ΔLgR’ 基线

[0129] where R is the resistivity logging data and Δt is the acoustic travel time data;

[0130] For example, taking the acoustic travel time in the mudstone logging data as the independent variable and the logarithm of the resistivity as the dependent variable for linear fitting, the slope after linear fitting is -0.021. After taking the absolute value, 0.021 is used as the initial value of parameter K. After floating up and down, a range of 0.010 to 0.030 is selected as the value range of the fitting parameter K, and the step size is 0.001. Thus, it is determined that the fitting parameter K has 21 possible values; taking the ΔLgR value of the logging data points as the independent variable and the experimentally measured value as the dependent variable for linear fitting, the fitted slope of 4.5 is used as the initial value of variable a. Floating up and down based on the initial value of parameter a, a range of 0.1 to 10.0 is selected as the value range of parameter a, and the step size is 0.1. Thus, it is determined that the fitting parameter a has 100 possible values; screening the non-mudstone data in the experimentally measured values, taking the maximum value of 0.3 of the organic matter abundance of the non-mudstone data as the initial value of variable b, and the value range of the fitting parameter b is selected as 0 to 0.3, with a step size of 0.1. Thus, it is determined that the fitting parameter b has 4 possible values. Based on these possible values of the fitting parameters K, a, and b, combinations of the possible values of the fitting parameters K, a, and b are determined.

[0131] Calculate the corresponding mudstone baseline parameter ΔLgR’ 基线 , including: sliding a fixed-size depth window (depth segment) across the logging profile in sequence, calculating the basification factor values of each depth point within the depth window, and taking the arithmetic mean of the basification factor values of each depth point within each depth window as the basification factor value of the formation within that depth window; if the basification factor value of the formation within a certain depth window is within the acceptable range, then take this depth segment (and its corresponding logging data) as the mudstone baseline; if the basification factor values of the formations within all depth windows on the single-well profile exceed the acceptable range, then select several depth points (and their corresponding logging data) with the smallest basification factor values in the whole-well mudstone to form a "virtual mudstone baseline" (collectively referred to as the mudstone baseline hereinafter); for each specific value of K, calculate the ΔLgR’ 基线 value of each depth point based on the logging data within the said mudstone baseline; take the arithmetic mean of the ΔLgR’ 基线 values of each depth point as the mudstone baseline parameter ΔLgR’ 基线 .

[0132] For example, sliding a 20m depth window (depth segment) across the logging profile of a certain well in sequence, calculating the basification factor values of each depth point within the depth window, taking the arithmetic mean of the basification factor values of each depth point within each depth window as the basification factor value of the formation within that depth window, and selecting the depth segment from 1501.2m to 1521.2m with the smallest basification factor value in the whole-well mudstone as the mudstone baseline, and its basification factor value is less than the set threshold of 0.3; for each value within the value range of K from 0.010 to 0.030, calculate the ΔLgR’ of each depth point within 1501.2m to 1521.2m based on the logging data within the said mudstone baseline基线 values, and the arithmetic mean value of ΔLgR’ at each depth point within the depth segment 基线 The arithmetic mean value of 2.4 of the values is used as the shale baseline parameter ΔLgR’ 基 line.

[0133] Specifically, the calculation formula for the base factor value at each depth point within the depth window is:[[]]

[0134] ABS((LOG10(R)+1.523)*2.5-(10-Δt / 20)) / 10

[0135] where R is the resistivity logging data and Δt is the acoustic travel time data.

[0136] The ΔLgR’ values at each depth point within the depth segment of 1501.2m to 1521.2m 基线 The calculation formula is:[[]]

[0137] ΔLgR’ 基线 = LgR 基线 + K×Δt 基线

[0138] where R 基线 is the logging resistivity within the determined shale baseline, and Δt 基线 is the logging acoustic travel time data within the determined shale baseline.

[0139] Based on the fitted value and the measured value of organic matter abundance, evaluate the fitting effect, and screen out the combination of parameter values with the best fitting effect, including: calculating the correlation coefficient R using the fitted value and the measured value of organic matter abundance 2 , the correlation coefficient R 2 The value of the fitting parameter K with the highest value is used as the preferred fitting parameter K; compare the average value of the fitted value of organic matter abundance with the average value of the measured value of organic matter abundance, and the values of the fitting parameters a and b with the smallest deviation degree of the results are used as the preferred fitting parameters a and b; the combination of parameter values with the best fitting effect screened out is the preferred fitting parameter K, the preferred fitting parameters a and b, and the shale baseline parameter LgR’ calculated based on the preferred fitting parameter K 基线 combination.

[0140] Based on the fitted value and the measured value of organic matter abundance, evaluate the fitting effect, and screen out the parameter value combination with the best fitting effect, including: setting the first logging depth floating interval according to the accuracy of logging data to reflect the influence of factors such as cable stretching, buoyancy, friction, mud pressure and temperature change on the fitting effect; setting the second sampling depth floating interval according to the sampling situation and accuracy of measured samples to reflect the influence of factors such as sample type (core / cuttings), sampling volume, sampling (merging of multi-point samples, core recovery rate of drilling) on the fitting effect; adding the first logging depth floating interval and the second sampling depth floating interval as the depth floating interval; taking the organic matter abundance fitted value with the highest matching degree within the corresponding depth floating interval of the measured value of organic matter abundance as the organic matter abundance fitted value at this depth point; using different combinations of fitting parameters K, a, b and the corresponding ΔLgR’ 基线 Calculating the correlation coefficient R between several groups of calculated organic matter abundance fitted values and the measured values of organic matter abundance 2 , the correlation coefficient R 2 When the value of the fitting parameter K corresponding to the highest value is used as the preferred fitting parameter K; under the condition of the preferred fitting parameter K, comparing the average value of the organic matter abundance fitted value with the average value of the measured value of organic matter abundance, and taking the values of the fitting parameters a and b with the smallest deviation degree as the preferred fitting parameters a and b; the screened parameter value combination with the best fitting effect is the preferred fitting parameter K, the preferred fitting parameters a and b, and the mudstone baseline parameter ΔLgR’ calculated based on the preferred fitting parameter K 基线 Combination of

[0141] S2: For wells in the work area without measured values of organic matter abundance: Using the predicted fitting parameters and well location coordinate data of wells in the work area with measured values of organic matter abundance, through a spatial interpolation model, predict the fitting parameters of wells in the work area without measured values of organic matter abundance.

[0142] The "well without measured value of organic matter abundance" mentioned in the present invention refers to a well without available measured value of organic matter abundance.

[0143] Before using the predicted fitting parameters and well location coordinate data of wells in the work area with measured values of organic matter abundance to predict the fitting parameters of wells in the work area without measured values of organic matter abundance, eliminate the unreliable data in the △LgR model fitting parameters predicted by wells with measured values of organic matter abundance to improve the quality of data to be fitted; the wells with measured values of organic matter abundance whose data are eliminated are regarded as wells without measured values of organic matter abundance, and their fitting parameter prediction methods are the same as those of wells without measured values of organic matter abundance; using the predicted fitting parameters and well location coordinate data of the screened wells with measured values of organic matter abundance in the work area, through a spatial interpolation model, predict the fitting parameters of wells in the work area without measured values of organic matter abundance.

[0144] The geological environments formed by adjacent source rocks in the same work area have certain similarities, and their fitting parameters of K, LgR’, 基线 , a, and b are similar. However, for the fitting parameters of K, LgR’, 基线 , a, and b of the ΔLgR model of each well with measured values preliminarily selected in step S1, due to factors such as the number of core measured data, sample representativeness, and planar heterogeneity, the predicted results of the K, LgR’, 基线 , a, and b parameters of local adjacent well positions may have some outlier anomalies. To minimize the influence of outlier anomalies on the parameter prediction results, obvious outlier data can be removed through conditions such as the number of core measured data, fitting effect indicators, and parameter planar distribution to optimize the predicted parameters.

[0145] Remove the unreliable data in the fitting parameters of the △LgR model predicted by the wells with measured values of organic matter abundance. The methods include at least one of the following: removing the fitting parameters of wells with poor fitting effects and too few measured data through fitting effect indicators such as the fitting correlation coefficient and the deviation degree of the average value, and the number of core measured data; removing abnormal outlier fitting parameters through outlier detection; when necessary, spatial interpolation can be performed on the fitting parameters of K, a, b, and ΔLgR’ 基线 and well location coordinate data respectively, and then a suitable size and type of convolution kernel can be selected for smoothing filtering to remove the mutation noise points in the data. For example, a mean convolution kernel with a size of 10km×10km is selected to slide on the data surface after spatial interpolation in turn to filter out the local abnormal extreme values in the data.

[0146] Preferably, the data of the parameter prediction results with the number of core measured data less than 3, the fitting correlation coefficient less than 0.7, and the ratio of the average value of the fitting value to the average value of the measured value exceeding the range of 0.8 - 1.2 can be directly removed, that is, the predicted fitting parameters are ignored and do not participate in the calculation of the fitting parameters of the wells without measured values of organic matter abundance in the subsequent prediction work area. The fitting parameters of the well are predicted according to the determination method of the fitting parameters of the wells without measured values of organic matter abundance.

[0147] The method of removing abnormal outlier fitting parameters through outlier detection includes: calculating the average value μ and standard deviation σ of the parameters of K, a, b, and ΔLgR’ 基线 respectively; calculating the variable distribution interval (μ - 3σ, μ + 3σ); identifying and removing the abnormal data outside the variable distribution interval.

[0148] The spatial interpolation model uses the Kriging interpolation method, and its theoretical variogram model adopts the linear model.

[0149] The method for predicting fitting parameters of the ΔLgR method logging source rock evaluation model provided by the present invention is applicable to wells with measured organic matter abundance values in the work area. By determining several combinations of possible values of the fitting parameters, calculating the fitting values for each combination respectively and evaluating the fitting effect accordingly, the combination of parameter values with the best fitting effect can be screened out. All possible combinations of parameter values can be comprehensively screened, the influence of subjective factors on the calculation results of organic matter abundance can be eliminated, and all undetermined parameters in the ΔLgR method logging source rock evaluation model can be objectively and accurately predicted, which helps to improve the data prediction accuracy and work efficiency. The method for predicting fitting parameters of the ΔLgR method logging source rock evaluation model provided by the present invention predicts the fitting parameters of wells with measured organic matter abundance values in the work area based on the measured organic matter abundance values, and predicts the fitting parameters of wells without measured organic matter abundance values in the work area based on the fitting parameters of wells with measured organic matter abundance values in the work area, so as to realize the prediction of fitting parameters of all single wells in the same work area.

[0150] According to the second aspect of the present invention, there is provided a method for calculating the shale-type shale oil resource volume, as Figure 2 shown, the method comprising the following steps:

[0151] S10: Collect the experimental determination data of the oil content rate of samples, logging data, horizon division data, lithology profile description data, Ro maturity data, oil content rate recovery correction model data, well location coordinate data, and work area boundary data of each single well in the work area;

[0152] S20: Use the method described in the first aspect of the present invention to predict the fitting parameters of the ΔLgR method logging source rock evaluation model of each single well in the work area, and fit the oil content rate profile data of each single well in the work area according to the fitting parameters and the logging data;

[0153] S30: According to the fitted oil content rate profile data and the horizon division data, perform summary statistics by stratification and self-defined grading to obtain the formation thickness and formation average oil content rate data of each single well in the work area under each stratification and each grading;

[0154] S40: Calculate the shale-type shale oil resource volume in the work area according to the formation thickness, formation average oil content rate data, horizon division data, Ro maturity data, oil content rate recovery correction model data, well location coordinate data, and work area boundary data of each single well in the work area under each stratification and each grading.

[0155] Calculate the shale-type shale oil resource volume in the working area, including: according to the formation thickness and formation average oil content rate data of each single well in each layer and each grade in the working area, using the mass oil content rate model to calculate the resource volume of each single well in the working area; construct a basic background grid covering the entire working area; based on the resource volume data and well position coordinate data in each layer and grade in the working area, establish a resource volume grid model for each layer and each grade based on the basic background grid through a spatial interpolation model; based on the Ro maturity data and well position coordinate data in each layer in the working area, establish an Ro grid model based on the basic background grid through a spatial interpolation model; based on the Ro grid model, use the oil content rate recovery correction model to sequentially convert the data in the Ro grid model into recovery coefficients to obtain a recovery correction coefficient grid model; based on the resource volume grid model and the recovery correction coefficient grid model, multiply the data in the corresponding grids in the two models to obtain a resource volume grid model after recovery for each layer and each grade; according to the working area boundary data, determine whether each grid in the resource volume grid model after recovery for each layer and each grade is within the working area boundary, and count the total resource volume of all grids in each layer and each grade within the working area boundary, and finally obtain the shale-type shale oil resource volume for each layer and each grade in the working area.

[0156] Figure 3 This is a flowchart of the shale-type shale oil resource volume evaluation method provided by an embodiment of the present invention; as Figure 3 shown, the shale-type shale oil resource volume evaluation method described in the embodiments of the present disclosure includes: preparing and integrating basic data; fitting the oil content rate profile of each single well in the working area; statistically analyzing the oil content rate of each single well in the working area by layer and grade; calculating the resource volume.

[0157] I. Prepare and integrate basic data

[0158] The prepared data must include the acoustic time difference data, resistivity logging data, lithology description data or gamma logging data, well number index data of the logging file, core oil content rate experimental analysis data (core pyrolysis S1), horizon division data, well position coordinate data, core vitrinite reflectance measurement data, oil content rate recovery correction model data, working area boundary data, etc. After obtaining the prepared data, integrate all the logging data (acoustic, resistivity, gamma), horizon division data, lithology description data, shale TOC (total organic carbon) detection data, and shale S1 detection data of all wells by well.

[0159] The data should contain the following information:

[0160] ① Lithology description data: It should contain information such as well number, top depth of lithology, bottom depth of lithology, and lithology description;

[0161] ②Well number index data of logging files: It should at least include information such as the logging file path and the well number corresponding to the logging file;

[0162] ③Stratigraphic division data: It should include information such as well number, stratigraphic name, top boundary depth of the stratigraphic horizon, bottom boundary depth of the stratigraphic horizon, etc. (or other data from which the above information can be converted);

[0163] ④Well location coordinate data: It should include well number, longitude and latitude coordinate information;

[0164] ⑤Oil recovery correction model data: It should include maturity-corresponding recovery coefficient data or other models from which the corresponding relationship between maturity and recovery coefficient can be converted;

[0165] ⑥Work area boundary data: It is a series of coordinate point data enclosing the work area boundary, consisting of a series of longitude and latitude coordinates.

[0166] II. Fitting the oil content profile of each individual well in the work area

[0167] 1. Predict the fitting parameters of the ΔLgR model for each individual well.

[0168] 1.1 Predict the fitting parameters of the ΔLgR model for wells with measured values, including:'screening shale data in logging data', 'determining several combinations of possible values of fitting parameters K, a, b, and calculating the corresponding shale baseline parameter ΔLgR' 基线 ', 'profile fitting, evaluating the fitting effect and predicting parameters'.

[0169] ①Screening shale data in logging data. To screen shale data in logging data, 'lithology description data' or'shale GR logging data' can be used. When using 'lithology description data' to screen shale data, if the lithology description at a certain depth is shale and the lithology description does not contain sandstone, then judge that section of logging data as shale logging data, otherwise judge it as non-shale logging data; when screening shale data in logging curves according to'shale GR logging data', if the logging GR value is significantly higher than the normal GR value of the sandstone section, then judge that section of logging data as shale logging data, otherwise judge it as non-shale logging data.

[0170] Preferably, when using lithology description data to screen shale data in logging curves, if the lithology description corresponding to a certain section of logging data contains any one of the keywords'shale', 'claystone', 'limestone', and the lithology description does not contain the keyword'sand', then judge that section of logging data as shale logging data, otherwise judge it as non-shale logging data; when using GR logging data to screen shale data in logging curves, if the logging GR value is greater than 80 us / f, then judge that section of logging data as shale logging data, otherwise judge it as non-shale logging data.

[0171] ② Determine several combinations of possible values of the fitting parameters K, a, and b, and calculate the corresponding shale baseline parameter ΔLgR’ 基线 The main steps include: determining the initial values of the parameters K, a, and b in the work area and the value ranges of the parameters K, a, and b, and calculating the corresponding shale baseline parameter ΔLgR’ 基线 ; based on the value ranges and value precisions (step sizes) of the fitting parameters K, a, and b, respectively determine several possible values of the fitting parameters K, a, and b; based on the several possible values of the fitting parameters K, a, and b, determine several combinations of possible values of the fitting parameters K, a, and b by the free permutation and combination method.

[0172] Among them, when calculating the shale baseline parameter ΔLgR’ 基线 , for each K value, the data within the shale baseline can calculate the ΔLgR’ value of each point based on the logging acoustic travel time and resistivity data 基线 ; the average value of the ΔLgR’ values of each point 基线 is regarded as the ΔLgR’ of the shale baseline 基线 .

[0173] ③ Profile fitting, fitting effect evaluation, and parameter prediction. Each set of K, ΔLgR’ 基线 , a, and b parameters can be combined with logging data (acoustic travel time, resistivity) and the △LgR model formula to obtain an organic matter content (oil content) profile. By evaluating the fitting effect, the parameter combination with the best fitting effect can be screened out.

[0174] Preferably, the correlation coefficient R 2 is used to test the goodness of fit. The value of R 2 ranges from 0 to 1. The closer the value is to 1, the higher the degree of coincidence; the closer it is to 0, the lower the degree of coincidence. Since the linear correlation coefficient has linear invariance, the correlation coefficient remains unchanged after the translation and scaling (linear transformation) of the variables. Therefore, for the same acoustic travel time and resistivity data, R 2 is related to K and independent of the a and b parameters. After combining the fitting values and the measured values, the relationship curve of K and R 2 can be obtained (as shown in Figure 4 ), and the K value with the best goodness of fit can be predicted. In order to evaluate the deviation degree between the fitting value and the measured value of ΔLgR after the linear transformation of the a and b parameters, after using the correlation coefficient to predict the optimal K and the corresponding LgR’ 基线 , the a and b parameters can be further evaluated by the ratio of the average value of the fitting values to the average value of the measured values (or Euclidean distance, etc.) (as shown in Figure 5 ), and the parameters a and b that minimize the result deviation degree can be predicted.

[0175] Furthermore, the depth of logging data may have certain deviations due to factors such as cable expansion and contraction, measurement accuracy, etc.; the depth of experimental test data has certain deviations due to factors such as drilling coring, cuttings repositioning, and sampling position accuracy. Therefore, if the depth of core experimental test data is directly used to fit the logging data, both the fitting effect and the predicted values of single-well fitting parameters will be affected by the depth deviation factor. To reduce the influence brought by the fitting depth deviation, when evaluating the fitting effect, the first and second depth floating intervals can be set, and the experimental test data can be compared with the logging fitting value with the highest matching degree near the corresponding depth.

[0176] Preferably, during the offset optimization process of the data fitting depth, the logging fitting results corresponding to the measured point depth can be matched with the data within the range of 3 data (i.e., 6 * 0.125 m) above and below the logging data, and the closest fitting value can be found as the logging fitting value of this depth point.

[0177] Preferably, to improve the implementation efficiency of the method, processes such as ① - ③ can be executed in an appropriate method combination, cross (nested) manner.

[0178] 1.2 Predict the fitting parameters of all wells in the prediction area: The parameters predicted in the process of 'predicting the fitting parameters of the ΔLgR model for wells with measured values' above, after removing unreliable data and well location coordinate data, predict the fitting parameters of all wells in the prediction area through a spatial interpolation model.

[0179] Preliminarily optimize the K, LgR', 基线 a, b fitting parameters of the ΔLgR model for each well with measured values. Due to factors such as the number of core measured data, sample representativeness, and planar heterogeneity, the predicted results of the K, LgR' 基线 a, b parameters of locally adjacent well positions may have some outlier abnormal values. To minimize the influence of outlier abnormal values on the parameter prediction results, obvious outlier data can be removed through conditions such as the number of core measured data, fitting effect indicators, removing abnormal outlier fitting parameters, and parameter planar distribution, and the parameter value combination with the best fitting effect can be selected.

[0180] Preferably, the parameter prediction result data with the number of core measured data less than 3, the fitting correlation coefficient less than 0.7, and the ratio of the average fitting value to the average measured value exceeding the range of 0.8 - 1.2 can be removed. Then, abnormal outlier fitting parameters can be removed through outlier detection. When necessary, spatial interpolation can be performed on the K, a, b, ΔLgR' 基线 fitting parameters and well location coordinate data respectively, and then a convolution kernel with an appropriate size and type can be selected for smoothing filtering to remove the mutation noise points in the data.

[0181] Among them, the method of removing abnormal outlier fitting parameters through outlier detection includes:

[0182] Calculate K, a, b, and ΔLgR’ respectively 基线 The average value μ and standard deviation σ of the parameters

[0183] Calculate the variable distribution interval (μ - 3σ, μ + 3σ);

[0184] Identify and eliminate the abnormal data beyond the variable distribution interval

[0185] Preferably, the spatial interpolation model uses Kriging interpolation method, and its theoretical variogram model adopts a linear model

[0186] 2 Fit the oil-bearing rate profiles of each single well in the work area: According to the fitting parameters predicted in the process of 'predicting the fitting parameters of the ΔLgR model for each single well' and the logging data, the oil-bearing rate profiles of each single well in the work area can be fitted according to the ΔLgR model

[0187] III. Stratigraphic horizon, hierarchical statistics of the oil-bearing rate of each single well in the work area

[0188] According to the oil-bearing rate profile data and stratigraphic horizon division data fitted by the ΔLgR model, summary statistics can be carried out according to stratification and user-defined grading, and the formation thickness and formation average oil-bearing rate data of each single well under each stratification and each grading can be obtained

[0189] Preferably, when conducting data stratification statistics, if the logging data of a certain well fails to completely cover the stratigraphic horizon, the data of this stratigraphic horizon should be directly abandoned from statistics to avoid distortion of data such as formation thickness; when using S1 as the oil-bearing rate index, [0, 0.5], (0.5, 1], (1, 2], (2, ∞) can be used as grading

[0190] IV. Resource quantity calculation

[0191] Calculate the resource quantity in the work area according to the statistical oil-bearing rate data, stratigraphic horizon division data, Ro maturity data, oil-bearing rate recovery correction model, well location coordinate data, work area boundary data, etc. The specific steps include

[0192] 1. Calculate the resource quantity of each single well in the work area according to the mass oil-bearing rate model. Calculate the shale oil resource quantity according to the formation thickness and formation average oil-bearing rate data of each single well in the work area under each stratigraphic horizon and each grading. The calculation formula is

[0193] Q = S × h × ρ × K

[0194] Among them, Q is the resource quantity under stratification and grading; S is the area; h is the effective formation thickness under stratification and grading; ρ is the density of mud shale; K is the formation average oil-bearing rate under stratification and grading

[0195] Preferably, S takes a value of 1 square kilometer; ρ takes a value of 2.4 grams per cubic centimeter

[0196] 2. Construct a grid covering the entire work area. Establish a grid with fixed-size squares in the entire work area, and use this grid as the basic background grid for other grid models.

[0197] Preferably, the step size of the basic background grid is set to 1 kilometer.

[0198] 3. Establish a resource volume grid model. Based on the resource volume data and well location coordinate data within each layer and grade in the work area, establish resource volume grid models for each layer and grade based on the basic background grid through a spatial interpolation model. If the resource volume of a certain grid is negative after interpolation by the spatial interpolation model, then define the resource volume of this grid as 0.

[0199] Preferably, the spatial interpolation model uses the Kriging interpolation method, and its theoretical variogram model adopts a linear model.

[0200] 4. Establish a Ro grid model. Based on the Ro data and well location coordinate data within each horizon in the work area, establish a Ro grid model based on the basic background grid through a spatial interpolation model. If the Ro of a certain grid is negative after interpolation by the spatial interpolation model, then define the Ro of this grid as 0.

[0201] Preferably, the spatial interpolation model uses the Kriging interpolation method, and its theoretical variogram model adopts a linear model.

[0202] 5. Establish a recovery correction coefficient grid model. Based on the Ro grid model, use the oil content recovery correction model to sequentially convert the data in the Ro grid model into recovery coefficients to obtain the recovery correction coefficient grid model.

[0203] 6. Correct the recovered resource volume grid model. Based on the resource volume grid model and the recovery correction coefficient grid model, multiply the data in the corresponding grids of the two models to obtain the recovered resource volume grid models for each layer and grade.

[0204] 7. Statistically calculate the total resource volume of the recovered resource volume grid model. According to the work area boundary data, determine whether each grid in the recovered resource volume grid models for each horizon and grade is within the work area boundary, and statistically calculate the sum of the resource volumes of all grids under each horizon and grade within the work area boundary, and finally obtain the resource volumes for each horizon and grade within the work area.

[0205] According to the third aspect of the present invention, there is provided a shale-type shale oil resource volume calculation system, as shown in the block diagram of Figure 6 and the flowchart of Figure 7 This system includes:

[0206] A data acquisition module 61, configured to acquire experimental determination data of the oil content rate of samples of each single well in the work area, logging data, horizon division data, lithologic profile description data, Ro maturity data, oil content rate recovery correction model data, well location coordinate data, and work area boundary data;

[0207] A data integration module 62, configured to summarize the experimental determination data of the oil content rate of samples, logging data, horizon division data, and lithologic profile description data of each single well together according to the sample depth;

[0208] A fitting parameter prediction module 63, configured to predict the fitting parameters of the ΔLgR method logging source rock evaluation model for each single well in the work area by using the method described in the first aspect of the present invention;

[0209] An oil content rate profile module 64, configured to fit the oil content rate profile data of each single well in the work area according to the predicted fitting parameters and logging data;

[0210] A hierarchical and classified statistical module 65, configured to perform summary statistics according to the fitted oil content rate profile data and horizon division data according to stratification and custom classification, so as to obtain the formation thickness and formation average oil content rate data of each single well in the work area under each stratification and each classification;

[0211] A single well resource quantity calculation module 66, configured to calculate the resource quantity of each single well in the work area by using the mass oil content rate model according to the formation thickness and formation average oil content rate data of each single well in the work area under each stratification and each classification;

[0212] A resource quantity interpolation prediction module 67, configured to, based on the resource quantity data and well location coordinate data within each stratification and classification of the work area, establish a resource quantity grid model for each stratification and each classification based on the basic background grid through a spatial interpolation model;

[0213] A resource quantity recovery correction module 68, configured to, based on the Ro maturity data and well location coordinate data within each stratification of the work area, establish an Ro grid model based on the basic background grid through a spatial interpolation model, and based on the Ro grid model, use the oil content rate recovery correction model to sequentially convert the data within the Ro grid model into recovery coefficients to obtain a recovery correction coefficient grid model, and based on the resource quantity grid model and the recovery correction coefficient grid model, multiply the data in the corresponding grids of the two models to obtain a resource quantity grid model after recovery for each stratification and each classification;

[0214] A resource quantity summary statistics module 69, configured to determine whether each grid within the resource quantity grid model after recovery for each stratification and each classification is within the work area boundary according to the work area boundary data, and statistically sum up the resource quantities of all grids under each stratification and each classification within the work area boundary, and finally obtain the shale-type shale oil resource quantity for each stratification and each classification within the work area.

[0215] In one embodiment, the fitting parameter prediction module 63 includes: a first sub-module 631, configured to predict the △LgR model fitting parameters of wells with measured organic matter abundance values in the work area; a screening module 632, configured to eliminate unreliable data from the △LgR model fitting parameters predicted by the first sub-module; a second sub-module 633, configured to use the predicted fitting parameters and well location coordinate data of wells with measured organic matter abundance values in the work area after eliminating unreliable data, and through a spatial interpolation model, predict the fitting parameters of wells without measured organic matter abundance values in the work area.

[0216] By using the above system, the △LgR method logging source rock evaluation model fitting parameter prediction method provided in the first aspect of the present invention and the shale-type shale oil resource volume calculation method provided in the second aspect of the present invention can be automatically implemented through a program, so as to achieve automatic and rapid evaluation of resource volume. After implementation, it can save human resources, revolutionarily improve work efficiency, significantly shorten the evaluation cycle, and at the same time contribute to the efficient development of resource volume rolling evaluation.

[0217] The above-described embodiments are only for expressing the implementation manners of the present disclosure, and the description thereof is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present disclosure. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present disclosure, several deformations, equivalent replacements, improvements, etc. can be made, and these all belong to the protection scope of the present disclosure. Therefore, the protection scope of the patent of the present disclosure shall be subject to the appended claims.

Claims

1. A method for predicting fitting parameters of a logging source rock evaluation model using the △LgR method, characterized in that, Including the following steps: For wells with measured organic matter abundance values in the work area: Screen the shale data in the logging curves; Determine several combinations of possible values of the fitting parameters K, a, and b, and calculate the corresponding shale baseline parameter ΔLgR’ 基线 ; For different combinations of possible values of the fitting parameters K, a, and b and the corresponding ΔLgR’ 基线 , based on the selected mudstone data, using the ΔLgR method logging source rock evaluation model, the fitting values of organic matter abundance are calculated respectively; Based on the fitted values and measured values of organic matter abundance, evaluate the fitting effect, and screen out the combination of parameter values with the best fitting effect as the fitting parameters of the △LgR model for the wells with measured organic matter abundance values in the predicted work area; For wells without measured organic matter abundance values in the work area: Using the predicted fitting parameters and well location coordinate data of the wells with measured organic matter abundance values in the work area, through the spatial interpolation model, predict the fitting parameters of the wells without measured organic matter abundance values in the work area.

2. The method for predicting fitting parameters of a logging source rock evaluation model using the △LgR method according to claim 1, characterized in that, Screen the shale data in the logging curves, including: using lithology description data or shale GR logging data to screen the shale data in the logging curves.

3. The method for predicting fitting parameters of a logging source rock evaluation model using the △LgR method according to claim 2, characterized in that, When using lithology description data to screen shale data, if the lithology description of a certain section of the logging curve is shale and the lithology description does not include sandstone, it is judged that the data of this section of the logging curve is shale data, otherwise it is judged as non-shale data; when using shale GR logging data to screen shale data, if the logging GR value of a certain section of the logging curve is higher than the GR value of the normal sandstone section, it is judged that the data of this section of the logging curve is shale data, otherwise it is judged as non-shale data.

4. The method for predicting fitting parameters of a logging source rock evaluation model using the △LgR method according to claim 3, characterized in that, When using lithology description data to screen shale data, if the lithology description corresponding to a certain section of the logging curve contains any one of the keywords'mudstone','shale', 'limestone', and the lithology description does not contain the keyword'sand', it is judged that the data of this section of the logging curve is shale data, otherwise it is judged as non-shale data; when using shale GR logging data to screen shale data, if the logging GR value of a certain section of the logging curve is greater than 80 us / f, it is judged that the data of this section of the logging curve is shale data, otherwise it is judged as non-shale data.

5. The method for predicting fitting parameters of a logging source rock evaluation model using the △LgR method according to claim 1, characterized in that, Determine several combinations of possible values of the fitting parameters K, a, b, including: Determine the initial values and value ranges of the fitting parameters K, a, b for the entire area or local area in the work area; Based on the value ranges and value steps of the fitting parameters K, a, b, respectively determine several possible values of the fitting parameters K, a, b; Based on several possible values of the fitting parameters K, a, b, determine several combinations of possible values of the fitting parameters K, a, b by the free permutation and combination method.

6. The method for predicting fitting parameters of a logging source rock evaluation model using the △LgR method according to claim 5, characterized in that, Determine the initial values and value ranges of the fitting parameters K, a, b for the entire area or local area in the work area, including: Under the condition that the background of the entire area in the work area is similar, determine the initial values and value ranges of the fitting parameters K, a, b for the entire area in the work area; Under the condition that the background of the entire area in the work area is not similar, divide the work area into several local areas, each local area is used as a sub-work area, and the initial values and value ranges of the fitting parameters K, a, b are determined respectively, where the background includes structure, deposition process of organic matter, organic matter abundance, and hydrocarbon generation evolution history.

7. The method for predicting fitting parameters of a logging source rock evaluation model using the △LgR method according to claim 5, characterized in that, Determine the initial values and value ranges of the fitting parameters K, a, b for the entire area or local area in the work area, including: Using the acoustic travel time in the selected shale data as the independent variable and the logarithm of the resistivity as the dependent variable for linear fitting, the absolute value of the slope after linear fitting is used as the initial value of the fitting parameter K, and a range that fluctuates up and down by no less than 50% based on the initial value of the fitting parameter K is used as the value range of the fitting parameter K; Based on the selected shale data, calculate the corresponding shale baseline parameter ΔLgR' based on the fitting parameter K 基线 ; According to the initial value of the fitting parameter K and ΔLgR’ 基线 The ΔLgR value corresponding to the measured value of the logging data point is calculated, and the calculation formula is: ΔLgR = LgR + K×Δt - ΔLgR’ 基线 where, R is the resistivity and Δt is the acoustic travel time; Using the ΔLgR value of the logging data points as the independent variable and the measured value as the dependent variable for linear fitting, the fitted slope is used as the initial value of the fitting parameter a, and a range that fluctuates up and down by no less than 50% based on the initial value of the fitting parameter a is used as the value range of the fitting parameter a; Screen the non-shale data in the measured values, and use the maximum value of the organic matter abundance of the non-shale data as the initial value of the fitting parameter b. If the initial value of the fitting parameter b is higher than the maximum value of the organic matter abundance of the non-source rock, use the maximum value of the organic matter abundance of the non-source rock as the initial value of the fitting parameter b, and use the interval [0, b 初值 as the value range of the fitting parameter b.

8. The method for predicting fitting parameters of the logging source rock evaluation model by the △LgR method according to claim 1, characterized in that, Calculate the corresponding shale baseline parameter ΔLgR’ 基线 , including: Based on the base factor values at each depth point on the logging profile, determine the shale baseline; For each specific value of K, based on the logging data within the shale baseline, calculate the ΔLgR’ value at each depth point. 基线 value; Take the arithmetic mean of the ΔLgR’ values at each depth point 基线 as the shale baseline parameter ΔLgR’ 基线。 9. The method for predicting fitting parameters of the logging source rock evaluation model by the △LgR method according to claim 1, characterized in that, Calculate the corresponding shale baseline parameter ΔLgR’ 基线 , further comprising: Sweep the logging profile with a depth window of a fixed size in sequence, calculate the base factor value of each depth point within the depth window, and the calculation formula for the base factor value is: ABS((LOG10(R)+1.523)*2.5-(10-Δt / 20)) / 10 where, R is the resistivity and Δt is the acoustic travel time; Take the arithmetic mean of the base factor values of each depth point within each depth window as the base factor value of the formation within that depth window; If the base factor value of the formation within a certain depth window is within the acceptable range, take this depth segment as the shale baseline. If the base factor values of the formations within all depth windows on the single-well profile exceed the acceptable range, then select several depth points with the smallest base factor values in the whole-well shale to form the shale baseline; For each specific K value, calculate the ΔLgR’ value at each depth point based on the well logging data within the shale baseline. The calculation formula is as follows: 基线 value, and the calculation formula is: ΔLgR’ 基线 = LgR 基线 + K×Δt 基线 wherein, R 基线 is the resistivity within the determined shale baseline, and Δt 基线 is the acoustic travel time within the determined shale baseline; The arithmetic mean of the ΔLgR’ values at each depth point is used as the shale baseline parameter ΔLgR’ 基线 基线 .​ 10. The method for predicting fitting parameters of the logging source rock evaluation model by the △LgR method according to claim 1, characterized in that, Based on the fitted value of the organic matter abundance and the measured value of the organic matter abundance, evaluate the fitting effect, and screen out the combination of parameter values with the best fitting effect, including: Set the first logging depth floating interval according to the accuracy of the logging data; Set the second sampling depth floating interval according to the sampling situation and accuracy of the measured samples; Add the first logging depth floating interval and the second sampling depth floating interval to obtain the depth floating interval; Take the organic matter abundance fitted value with the highest matching degree within the depth floating interval corresponding to the measured value of the organic matter abundance as the organic matter abundance fitted value at this depth point; Calculate the correlation coefficient R using the fitted value of organic matter abundance and the measured value of organic matter abundance 2 , the correlation coefficient R 2 When it is the highest, take the value of the fitting parameter K as the preferred fitting parameter K; Under the condition of the optimized fitting parameter K, compare the average value of the organic matter abundance fitted value with the average value of the measured value of the organic matter abundance, and take the values of the fitting parameters a and b with the smallest deviation degree as the optimized fitting parameters a and b; The selected combination of parameter values with the best fitting effect is the preferred fitting parameter K, the preferred fitting parameters a and b, and the mudstone baseline parameter ΔLgR' calculated based on the preferred fitting parameter K 基线 combination.

11. The method for predicting fitting parameters of the logging source rock evaluation model by the △LgR method according to claim 1, characterized in that, Before predicting the fitting parameters of the wells without measured values of the organic matter abundance in the work area using the prediction fitting parameters and well location coordinate data of the wells with measured values of the organic matter abundance in the work area, eliminate the unreliable data in the △LgR model fitting parameters predicted by the wells with measured values of the organic matter abundance.

12. The method for predicting fitting parameters of the logging source rock evaluation model by the △LgR method according to claim 11, characterized in that, Eliminating the unreliable data in the △LgR model fitting parameters predicted by the wells with measured values of the organic matter abundance includes at least one of the following: For the wells with measured values of the organic matter abundance in the work area, eliminate the fitting parameters of the wells with poor fitting effect and too few measured data through the fitting effect index and the number of core measured data, where the fitting effect index includes the fitting correlation coefficient and the degree of deviation of the average value; Eliminate the abnormal outlier fitting parameters through outlier detection; Using the fitting parameters K, a, b, and ΔLgR’ 基线 Perform spatial interpolation on the well location coordinate data respectively, and then select a convolution kernel of appropriate size and type for smoothing filtering to remove the sudden noise points in the data.

13. The method for predicting fitting parameters of the logging source rock evaluation model by the △LgR method according to claim 12, characterized in that, Eliminating the abnormal outlier fitting parameters through outlier detection includes: Calculate the average values μ and standard deviations σ of the fitting parameters K, a, b, and ΔLgR' respectively 基线 ; Calculate the variable distribution interval (μ - 3σ, μ + 3σ); Identify and eliminate abnormal data that exceeds the variable distribution range.

14. The prediction method for fitting parameters of the △LgR method logging source rock evaluation model according to claim 1, characterized in that, The spatial interpolation model uses the Kriging interpolation method, and its theoretical variogram model adopts a linear model.

15. A calculation method for shale-type shale oil resource volume, characterized in that, It includes the following steps: Collect the experimental determination data of the oil content rate of samples from each single well in the work area, logging data, horizon division data, lithologic profile description data, Ro maturity data, oil content rate recovery correction model data, well location coordinate data, and work area boundary data; Use the method described in any one of the above claims 1-14 to predict the fitting parameters of the ΔLgR method logging source rock evaluation model for each single well in the work area, and fit the oil content rate profile data of each single well in the work area based on the fitting parameters and logging data; According to the fitted oil content rate profile data and horizon division data, perform summary statistics according to stratification and custom grading to obtain the formation thickness and formation average oil content rate data of each single well in the work area under each stratification and each grading; Calculate the shale-type shale oil resource volume in the work area based on the formation thickness, formation average oil content rate data, horizon division data, Ro maturity data, oil content rate recovery correction model data, well location coordinate data, and work area boundary data of each single well in the work area under each stratification and each grading.

16. The calculation method for shale-type shale oil resource volume according to claim 15, characterized in that, Calculate the shale-type shale oil resource volume in the work area based on the formation thickness, formation average oil content rate data, horizon division data, Ro maturity data, oil content rate recovery correction model data, well location coordinate data, and work area boundary data of each single well in the work area under each stratification and each grading, including: Construct a basic background grid covering the entire work area; Calculate the resource volume of each single well in the work area using the mass oil content rate model based on the formation thickness and formation average oil content rate data of each single well in the work area under each stratification and each grading; Based on the resource volume data and well location coordinate data in each stratification and grading in the work area, establish a resource volume grid model for each stratification and each grading based on the basic background grid through a spatial interpolation model, and define the data of grids with formation thickness and formation average oil content rate lower than the industrial development marginal limit value and negative value grids as 0; Based on the Ro maturity data and well location coordinate data in each stratification in the work area, establish an Ro grid model based on the basic background grid through a spatial interpolation model; Based on the Ro grid model, use the oil content rate recovery correction model to sequentially convert the data in the Ro grid model into recovery coefficients to obtain a recovery correction coefficient grid model; Based on the resource volume grid model and the recovery correction coefficient grid model, multiply the data in the corresponding grids of the two models to obtain a recovered resource volume grid model for each stratification and each grading; According to the work area boundary data, judge whether each grid in the recovered resource volume grid model for each stratification and each grading is within the work area boundary, and statistically sum up the resource volume of all grids under each stratification and each grading within the work area boundary, and finally obtain the shale-type shale oil resource volume for each stratification and each grading in the work area.

17. The calculation method for shale-type shale oil resource volume according to claim 15, characterized in that, When performing data stratification statistics, if the logging data of a certain well fails to completely cover the top or bottom surface of the horizon, directly abandon the statistics of the data of that horizon.

18. A calculation system for shale-type shale oil resource volume, characterized in that, The system includes: The data acquisition module is used to obtain the experimental determination data of the oil content of samples of each single well in the work area, well logging data, stratigraphic division data, lithology profile description data, Ro maturity data, oil content recovery correction model data, well location coordinate data, and work area boundary data; The data integration module is used to aggregate the experimental determination data of oil content of samples of each single well, well logging data, stratigraphic division data, and lithology profile description data according to the sample depth; A fitting parameter prediction module, used to predict the fitting parameters of the ΔLgR logging source rock evaluation model of each single well in the work area using the method described in any one of claims 1 to 14; The oil content profile module is used to fit the oil content profile data of each single well in the work area according to the predicted fitting parameters and logging data; The layered and graded statistics module is used to summarize and count the fitted oil content profile data and stratigraphic division data according to the layers and user-defined classifications, and obtain the formation thickness and average oil content data of each single well in the work area under each layer and classification; The single well resource calculation module is used to calculate the resource volume of each single well in the work area according to the formation thickness and average formation oil content data of each single well in each layer and each grade, using the mass oil content model; The resource interpolation prediction module is used to establish resource grid models of each layer and each level based on the basic background grid through a spatial interpolation model based on the resource data and well location coordinate data in each layer and level of the work area; The resource recovery and correction module is used to establish a Ro grid model based on the basic background grid through a spatial interpolation model based on the Ro maturity data and well location coordinate data in each layer of the work area, and to convert the data in the Ro grid model into recovery coefficients in sequence using the oil content recovery and correction model based on the Ro grid model to obtain a recovery correction coefficient grid model, and to multiply the data in the corresponding grids in the two models based on the resource grid model and the recovery correction coefficient grid model to obtain a resource grid model after recovery for each layer and each grade; The resource quantity summary statistics module is used to determine whether each grid in the resource quantity grid model after recovery of each layer and each grade is within the work area boundary according to the work area boundary data, and to count the total resource quantity of all grids in each layer and each grade within the work area boundary, and finally obtain the shale oil resource quantity of each layer and each grade in the work area.

19. The calculation system for shale-type shale oil resource volume according to claim 18, characterized in that, The fitting parameter prediction module comprises: The first submodule is used to predict the △LgR model fitting parameters of wells with measured values ​​of organic matter abundance in the work area; A screening module, used to eliminate unreliable data in the △LgR model fitting parameters predicted by the first submodule; The second submodule is used to predict the fitting parameters of wells without measured values ​​of organic matter abundance in the work area through a spatial interpolation model by using the predicted fitting parameters and well location coordinate data of wells with measured values ​​of organic matter abundance in the work area after removing unreliable data.

Images