TOC prediction method for different lithologic source rocks

Abstract

The invention relates to a TOC prediction method for different lithologic source rocks, and belongs to the field of oil-gas exploration and development. Lithology of to-be-measured hydrocarbon source rock in the target area and logging data of a single well at the position where the to-be-measured hydrocarbon source rock is located are obtained, if the lithology of the current hydrocarbon source rock is dark mudstone, the TOC content of the dark mudstone layer of the single well is quantitatively predicted through a modeling formula determined through a delta logR method, and if the lithology of the current hydrocarbon source rock is carbonaceous mudstone or a coal seam, the TOC content of the dark mudstone layer of the single well is quantitatively predicted through a modeling formula determined through a delta logR method. The TOC of the to-be-measured hydrocarbon source rock is quantitatively predicted through a unary linear regression model, the unary linear regression model is obtained through the fitting relation between the actually-measured TOC of the carbonaceous mudstone or the coal seam and the corresponding interval transit time, the TOC content predicted through the delta logR method and the TOC content predicted through the second TOC prediction mode are summarized to serve as the final TOC content prediction result, and the TOC content prediction result is obtained. The method is more accurate in prediction result and better in prediction effect.

Description

Technical Field

[0001] The invention relates to a TOC prediction method for hydrocarbon source rocks of different lithology, and belongs to the field of oil and gas exploration and development. Background Art

[0002] The evaluation of source rocks is the basis of oil and gas exploration research. Coal-bearing source rocks mainly include dark mudstone, carbonaceous mudstone and coal according to the organic matter abundance from low to high. However, carbonaceous mudstone and coal seams have higher TOC and contribute greatly to coal-bearing source rocks, so their prediction results are particularly important. Due to limited funds and sample sources, it is not possible to conduct laboratory analysis on every well, which brings great limitations to source rock evaluation. In addition, conventional source rock evaluation often takes data such as organic matter abundance and averages it to represent the total organic carbon content of the entire source rock. The result conceals the important contribution of high-quality source rocks to oil and gas accumulation, which ultimately leads to the deviation of source rock evaluation results from the actual situation, misleading exploration decisions and hindering oil and gas exploration. Therefore, accurate and quantitative prediction of organic matter abundance in source rocks has become the primary problem to be solved in oil and gas exploration.

[0003] Du Wenfeng published "Log Response Characteristics and Quantitative Prediction Model of Organic Carbon Content in Coal-Bearing Source Rocks" and Wang Pan published "Quantitative Prediction of TOC in Coal-Bearing Source Rocks by Well Logging", which pointed out that multivariate regression models should be established for different lithologies in coal-bearing source rocks. The problem is that the same method is used to predict TOC in coal-bearing source rocks of different lithologies, which has certain limitations. Secondly, most wells in coal seams and carbonaceous mudstones cannot be used in the multivariate regression method due to wellbore collapse and distortion of neutron and density curves.

[0004] Zhu Wuquan published a paper titled "Research on the prediction method of organic matter abundance logging curves in coal-bearing hydrocarbon source rocks", which mainly used the improved △logR method to predict TOC. On the basis of the previous formula, the background influence factor K was introduced. The K value of different lithologies was different to establish a prediction model. This method is not suitable for areas with thin coal seams or few coal seam samples. It is difficult to model pure coal seam sections less than 50 meters or with less than 30 samples. This leads to low prediction results and poor results in areas with thin coal seams or few coal seam samples. The existing TOC prediction method △logR model method has a high prediction accuracy for dark mudstone, but the prediction results for high-abundance source rocks (carbonaceous mudstone and coal seams) are low and the effect is poor. Summary of the invention

[0005] The purpose of the present invention is to provide a TOC prediction method for source rocks of different lithologies, so as to solve the problem that the TOC prediction results of high-abundance source rocks are low and the effect is poor. By clarifying the inflection point values ​​of different TOC prediction methods for high-abundance source rocks (carbonaceous mudstone and coal seams) and medium- and low-abundance source rocks, a more accurate optimization method for the prediction method is provided.

[0006] To achieve the above object, the solution of the present invention includes:

[0007] A TOC prediction method for source rocks of different lithologies of the present invention comprises the following steps:

[0008] 1) Obtaining the lithology of the source rock to be tested in the target area and the logging data of a single well at the location where the source rock to be tested is located;

[0009] 2) If the current source rock is dark mudstone, the ΔlogR method is used to quantitatively predict the TOC content of the dark mudstone layer of the source rock to be tested;

[0010] 3) If the current source rock is carbonaceous mudstone or coal seam, a univariate linear regression model is used to quantitatively predict the TOC of the source rock to be measured, wherein the univariate linear regression model is obtained by establishing a fitting relationship between the measured TOC of the carbonaceous mudstone or coal seam and the corresponding acoustic time difference;

[0011] 4) The TOC content predicted by the ΔlogR method and the TOC content predicted by the second TOC prediction method are summarized as the TOC content prediction result of the source rock to be tested in the target area.

[0012] Beneficial effect: The TOC prediction method of a source rock of the present invention adopts a combination of two prediction methods to predict the TOC of different lithologies. Different prediction methods are used for different lithologies. The ΔlogR method is used for quantitative prediction of dark mudstone, and the univariate linear regression model is used for quantitative prediction of carbonaceous mudstone or coal seam. The prediction results of source rocks by this method are accurate and the prediction effect is the best.

[0013] Furthermore, the prediction method also includes determining an inflection point value using core profile data, logging data and measured TOC data of the modeled well, wherein the inflection point value is used to indicate a critical value for switching the TOC content prediction method.

[0014] Beneficial effect: Whether to switch the prediction method is determined by different inflection point values, and the inflection point value is the critical value used to indicate the switching of the TOC content prediction method. According to the lithology of different abundance layers, two prediction methods are used to predict TOC, which makes the prediction more accurate and has better prediction effect.

[0015] Furthermore, the inflection point value is: selecting core profile data, logging data and measured TOC data of the modeling well, using the ΔlogR method to perform modeling, and obtaining a relationship between TOC and amplitude difference; using the modeling formula determined by the ΔlogR method to quantitatively predict the TOC content of the modeling well, and comparing the predicted result with the measured TOC content, and taking the value of the comparison result difference that exceeds the set range as the inflection point value.

[0016] Furthermore, the univariate linear regression model is:

[0017] Y=D*XE

[0018] Where Y is the TOC content of the carbonaceous mudstone and coal seam of the source rock to be tested, X is the acoustic time difference of the carbonaceous mudstone and coal seam of the source rock to be tested, and D and E are fitting coefficients.

[0019] Beneficial effect: The univariate linear regression method is used to predict the TOC of carbonaceous mudstone and coal seams with higher accuracy, which solves the problem that the results of predicting carbonaceous mudstone and coal seams using the ΔlogR model method in actual work are too low, and are much lower than the actual minimum lower limit of coal seams and carbonaceous mudstone.

[0020] Furthermore, the core profile data and measured TOC data are data after depth calibration and re-positioning.

[0021] Beneficial effect: The present invention simultaneously performs depth calibration on the measured TOC data of the single well and the core profile data of the single well model to improve the matching degree between the TOC data and the logging response characteristics, thereby making the TOC prediction result more accurate.

[0022] Furthermore, in step 2), the method for predicting TOC is: according to the relationship between the acoustic time difference and the resistivity in the logging data, the ΔlogR amplitude difference is obtained, a linear fitting relationship between the ΔlogR amplitude difference and TOC is established, and the TOC of the source rock is quantitatively predicted through the linear fitting relationship.

[0023] Beneficial effect: The organic matter abundance is characterized by the distance between the acoustic time difference and resistivity curves, namely ΔlogR. The larger the distance, the greater the TOC. The ΔlogR method is used to predict the TOC content of dark mudstone, and the prediction results are more accurate and reliable.

[0024] Furthermore, the inflection point value is set in the range of 6.0% to 8.0%.

[0025] Furthermore, the linear fitting relationship between the ΔlogR amplitude difference and TOC is:

[0026] TOC=A×ΔlogR+△TOC

[0027] Wherein, A is the coefficient of the fitting formula, ΔlogR is the distance after the resistivity and acoustic time difference curves are superimposed, ΔTOC is the organic carbon background value of the non-oil-generating layer, and the calculation formula of ΔlogR is:

[0028] ΔlgR=lg(R / R 基线 )+k(Δt-Δt 基线 )

[0029] Where k is the overlap coefficient; R is the resistivity; R 基线 is the resistivity value corresponding to the baseline; Δt is the acoustic wave time difference; Δt基线 is the corresponding acoustic time difference value on the baseline. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a flowchart of the technical solution research process in the method embodiment of the present invention;

[0031] Figure 2 Schematic diagram of TOC prediction process in the method embodiment of the present invention;

[0032] Figure 3 Schematic diagram of the ΔlogR curve overlap method in the method embodiment of the present invention;

[0033] Figure 4 Schematic diagram of the prediction model of dark mudstone in the modeling well in area A in the method embodiment of the present invention;

[0034] Figure 5 Schematic diagram of the correlation between predicted TOC and measured TOC of dark mudstone in the modeling well in area A in the method embodiment of the present invention;

[0035] Figure 6 is a numerical comparison diagram between the measured TOC and the predicted TOC of the source rock calculated by the ΔLogR method in the method embodiment of the present invention;

[0036] Figure 7 Schematic diagram of the prediction model of carbonaceous mudstone and coal seam in area A in the method embodiment of the present invention;

[0037] Figure 8 It is a comparison chart of the measured TOC of carbonaceous mudstone and coal seam in area A and the TOC predicted by the △logR method in the method embodiment of the present invention;

[0038] Fig. 9 It is a comparison chart of the measured TOC of coal seams and carbonaceous mudstones in area A and the TOC predicted by the linear regression method in the method embodiment of the present invention;

[0039] Fig.10 This is a comparison diagram of two methods for carbonaceous mudstone and coal seam in area A in the method embodiment of the present invention. DETAILED DESCRIPTION

[0040] The present invention will be further described in detail below in conjunction with the accompanying drawings.

[0041] Example of TOC prediction method for source rocks of different lithologies:

[0042] By comparing the measured TOC and predicted TOC maps of the target area using the ΔlogR model method, the inflection point values ​​of different TOC prediction methods for high-abundance source rocks (carbonaceous mudstone and coal seams) and medium-low abundance source rocks (dark mudstone) are clarified. Figure 1The TOC prediction method of a source rock shown in the figure uses the inflection point value as the boundary, and uses different methods to predict TOC for high-abundance source rocks (carbonaceous mudstone and coal seams) and medium-low abundance source rocks (dark mudstone). The univariate linear regression method that should be used for quantitative prediction of high-abundance source rocks (carbonaceous mudstone and coal seams) is more conducive to the overall evaluation of the true accuracy of the source rock. The specific implementation method is as follows:

[0043] This plan mainly takes the typical area of ​​the Xujiahe Formation in Area A of the Sichuan Basin where continental coal-bearing source rocks are developed as an example for research. The specific research process is as follows:

[0044] First, a single-well geochemical columnar profile of the modeling well in area A is established, and the core profile and analytical samples of the modeling well are calibrated and reset at the same time to improve the matching degree between TOC data and logging response characteristics. During the coring process, logging and other operations will be performed while drilling, so the depth of a core section can be obtained, that is, the initial core depth. Due to the errors between the core section recovery rate, drilling, logging and logging depth, the initial core section depth obtained is often inaccurate and differs greatly from the actual depth of the core section. Therefore, the core section and analytical sample values ​​must be calibrated and reset at the same time to restore the actual depth of the core section and analytical samples.

[0045] Depend on Figure 3 The ΔlogR method shown in the figure is based on the relationship between the organic matter abundance of source rocks and the logging response characteristics. The organic matter abundance is characterized by the interval between the acoustic time difference and the resistivity curve, i.e., ΔlogR. The linear relationship between the ΔlogR amplitude difference and TOC is then obtained, and the relationship between them is fitted:

[0046] TOC=A×ΔlogR+△TOC

[0047] Where A is the coefficient of the fitting formula; ΔlogR is the distance after the resistivity and acoustic time difference curves are superimposed; △TOC is the organic carbon background value of the non-oil-generating layer.

[0048] The calculation formula of ΔlogR is:

[0049] ΔlgR=lg(R / R 基线 )+k(Δt-Δt 基线 )

[0050] Where k is the overlap coefficient; R is the resistivity; R 基线 is the resistivity value corresponding to the baseline; Δt is the acoustic wave time difference; Δt 基线 is the corresponding acoustic time difference value on the baseline.

[0051] Among them, the logging response characteristics of mature source rocks are high natural gamma, low density, high acoustic time difference, and high resistivity. Therefore, the greater the amplitude difference ΔlogR between the acoustic time difference AC and the resistivity Rt, the higher the organic matter abundance TOC.

[0052] The original data of all coal-measure source rocks of different lithologies in the modeling wells of the target block are screened, including: core profile data, well logging curve data mainly including resistivity and acoustic wave time difference, and measured TOC data. That is, in the A area of ​​the present embodiment, the original data of modeling the coal-measure source rock prediction model in the A area are screened, including: core profile data, well logging curve data mainly including resistivity and acoustic wave time difference, and measured TOC data.

[0053] like Figure 4 As shown in Figure 1, the relationship between the measured TOC and the ΔlogR amplitude difference of the modeling wells in area A (excluding abnormal points) is:

[0054] Y=0.833X+0.1361

[0055] R 2 =0.866

[0056] Where, Y is TOC; X is the amplitude difference; R 2 is the correlation coefficient.

[0057] like Figure 5 As shown in the figure, from the relationship diagram between measured and calculated organic carbon content, the two have a high correlation (after screening out abnormal points), indicating that this method is feasible and reliable.

[0058] The original data of the prediction wells of the source rock prediction model of the target block are screened, including: logging lithology profile data, logging curve data mainly including resistivity and acoustic wave time difference; through logging lithology profile data and logging curve data (including resistivity and acoustic wave time difference), the logging organic phase model established by the ΔlogR method is used to predict the TOC of source rocks of different lithologies in the target area.

[0059] By comparing the data results after prediction of different lithology source rocks with the measured TOC data, the inflection point values ​​of the TOC prediction method for high-abundance source rocks and medium- and low-abundance source rocks in the target area can be clarified. Figure 6 As shown in the figure, the predicted value of medium-low abundance source rocks with TOC below 7.0% is consistent with the measured value, with an error rate of 14%; the predicted value of high-abundance source rocks with TOC greater than 7.0% is not consistent with the measured value, with an error rate of 82%. Therefore, the above prediction method is not suitable for TOC prediction of high-abundance source rocks. The inflection point value of the two is determined to be 7.0%.

[0060] The inflection point value of TOC is the critical value for indicating the switching of the TOC content prediction method, and the range is determined to be 6.0% to 8.0%. This embodiment is studied at 7.0%, so the set inflection point value of this embodiment is 7.0%. As other implementation methods, the set inflection point value can be any value between 6.0% and 8.0%, which is determined according to different engineering requirements in actual projects.

[0061] The inflection point value is obtained according to the above method: select wells with measured TOC values ​​and coring wells as modeling wells, screen the core profile data, logging data and measured TOC data of the modeling wells, use the ΔlogR method to model, and obtain the relationship between TOC and amplitude difference; use the modeling formula determined by the ΔlogR method to quantitatively predict the TOC content of the modeling well, and compare the predicted results with the measured TOC content, and take the value of the comparison result that exceeds the set range as the inflection point value. The ΔlogR method is suitable for the TOC prediction method of most dark mudstones, but for high-abundance source rocks such as coal seams and carbonaceous mudstones, the TOC values ​​are highly correlated with density and compensated neutron curve data in addition to acoustic time difference and resistance, so the ΔlogR method is not suitable for this method. Most dark mudstones are suitable for the ΔlogR method, and a few high-abundance lithology values ​​are screened out during modeling, retaining most of the medium and low-abundance dark mudstones. Therefore, the acoustic time difference, compensated neutron, density and other curves of the logging curves with high TOC sensitivity are selected for high-abundance source rocks (coal seams and carbonaceous mudstones) to establish parameters. Since the compensated neutron and density curves of high-abundance source rocks (carbonaceous mudstone and coal seam sections) are distorted due to wellbore collapse, the acoustic time difference curve data with the greatest TOC sensitivity is used to establish a univariate linear regression equation with TOC.

[0062] First, carbonaceous mudstone and coal seams are comprehensively and qualitatively identified based on logging curves. Coal seams have the logging response characteristics of "high neutrons, high acoustic time difference, high resistivity, low density, low natural potential, and low natural gamma". Carbonaceous mudstone and dark mudstone are rich in organic matter, showing the characteristics of high neutrons, high resistivity, high acoustic time difference, high natural gamma, high uranium content and low density.

[0063] The corrected core profiles and sample values ​​are used to screen the original data for modeling carbonaceous mudstone and coal seams in the target area, including core profile data, sonic time difference, and measured TOC data.

[0064] The univariate linear regression method was used to establish the high-abundance source rock logging organic phase model, that is, the measured TOC of carbonaceous mudstone and coal seams in the modeling well and the corresponding acoustic time difference AC were used to establish the correlation between the two:

[0065] Y=D*XE R2=F

[0066] In the formula, Y is TOC; X is AC; R2 is the correlation coefficient; D and E are factors.

[0067] like Figure 7 As shown, the measured TOC and corresponding acoustic time difference AC of carbonaceous mudstone and coal seams in the modeling well in area A.

[0068] Y=0.9509*X-55.229R2=0.857

[0069] After qualitatively identifying the carbonaceous mudstone and coal seams of the prediction well using the logging curve, quantitative prediction is performed according to the formula in the above steps.

[0070] The prediction data obtained based on the ΔlogR method and the prediction data obtained by the univariate linear regression method are summarized to form complete prediction results data for medium- and low-abundance source rocks and high-abundance source rocks (carbonaceous mudstone and coal seams), which is convenient for the next overall evaluation of source rocks.

[0071] The complete prediction results data of medium- and low-abundance source rocks and high-abundance source rocks (carbonaceous mudstone and coal seams) in area A are shown in Table 1.

[0072] Table 1

[0073]

[0074] Using the data in Table 1, the results of the two methods for predicting TOC of high-abundance source rocks (coal seams and carbonaceous mudstones) were compared, and it was found that Figure 8 As shown in the figure, the ΔlogR model method predicts coal seams and carbonaceous mudstones much lower than the actual values, the effect is poor, and the degree of agreement is only 10%. Fig. 9 The method of univariate linear regression was used to compare the measured values ​​with the predicted values ​​for coal seams and carbonaceous mudstones, and the consistency reached more than 90%. Figure 5 The measured TOC values ​​of low-abundance source rock dark mudstone and the TOC values ​​predicted by the △logR method are consistent with each other, reaching 86%. Therefore, dark mudstone is suitable for TOC prediction by the △logR method, while high-abundance source rocks (coal seams and carbonaceous mudstone) are suitable for prediction by the univariate linear regression method.

[0075] Based on the above research results, the technical solution of this embodiment is obtained. This embodiment obtains the TOC prediction results of the source rock to be tested by obtaining three types of source rocks, namely dark mudstone, carbonaceous mudstone and coal seam, in the target area. Figure 2 As shown, the specific steps of this embodiment are as follows:

[0076] 1) Obtain the lithology of the source rock to be tested in the target area, as well as the logging data of a single well where the source rock to be tested is located.

[0077] First, the logging data used in this embodiment mainly include the resistivity and acoustic time difference curve of the single well to be tested, and the lithology of the source rock to be tested is comprehensively and qualitatively obtained by using the core profile data and / or the logging curve.

[0078] In order to improve the accuracy of prediction and the matching degree between TOC data and logging response characteristics, the present embodiment also includes calibrating and resetting the core profile data and measured TOC data to obtain an accurate depth, thereby more accurately determining the lithology of the formation.

[0079] 2) If the current source rock is dark mudstone, the ΔlogR method is used to quantitatively predict the TOC content of the dark mudstone layer of the source rock to be tested;

[0080] The current source rock is dark mudstone. Figure 2 The ΔlogR method shown in the figure is to establish the relationship between the organic matter abundance of the source rock and the logging response characteristics. The organic matter abundance is characterized based on the interval between the acoustic time difference and the resistivity curve, i.e., ΔlogR. Then, the linear relationship between the ΔlogR amplitude difference and TOC is obtained, and the relationship between them is fitted:

[0081] TOC=A×ΔlogR+△TOC

[0082] Where A is the coefficient of the fitting formula; ΔlogR is the distance after the resistivity and acoustic time difference curves are superimposed; △TOC is the organic carbon background value of the non-oil-generating layer.

[0083] The calculation formula of ΔlogR is:

[0084] ΔlgR=lg(R / R 基线 )+k(Δt-Δt 基线 )

[0085] Where k is the overlap coefficient; R is the resistivity; R 基线 is the resistivity value corresponding to the baseline; Δt is the acoustic wave time difference; Δt 基线 is the corresponding acoustic time difference value on the baseline.

[0086] 3) If the current source rock is carbonaceous mudstone or coal seam, a univariate linear regression model is used to quantitatively predict the TOC of the source rock to be measured. The univariate linear regression model is obtained by establishing a fitting relationship between the measured TOC of the carbonaceous mudstone or coal seam and the corresponding acoustic time difference;

[0087] The current source rock lithology is carbonaceous mudstone or coal seam. The established univariate linear regression model is used to quantitatively predict the TOC of the source rock to be tested. The univariate linear regression model is:

[0088] Y=D*XE R2=F

[0089] Where, Y is the TOC content of carbonaceous mudstone and coal seam, X is the acoustic time difference between carbonaceous mudstone and coal seam, R 2 is the correlation coefficient, and D and E are fitting coefficients, which are determined according to the modeling well.

[0090] 4) The TOC content predicted by the ΔlogR method and the TOC content predicted by the second TOC prediction method are summarized as the TOC content prediction result of the source rock to be tested in the target area.

[0091] This embodiment fully studies the different logging response characteristics of different lithologies of coal-bearing source rocks, resulting in different prediction results. Therefore, the inflection points of different TOC prediction methods for high-abundance source rocks (carbonaceous mudstone and coal seams) and medium-low abundance source rocks (dark mudstone) are clarified, and the inflection point value TOC is 7.0% as the boundary. The medium-low abundance source rocks use the ΔlogR model method, while the high abundance source rocks (carbonaceous mudstone and coal seams) use the univariate regression method to predict TOC, which solves the problem that the results of using the ΔlogR model method to predict carbonaceous mudstones and coal seams in actual work are too low, and the results are much lower than the actual minimum lower limit of coal seams and carbonaceous mudstones (such as Fig.10 The optimized quantitative prediction method can ensure that the source rock evaluation of the target block obtains the most realistic evaluation suggestions.

Claims

1. A TOC prediction method for source rocks of different lithologies, characterized in that: The following steps are involved: 1) Obtaining the lithology of the source rock to be tested in the target area and the logging data of a single well at the location where the source rock to be tested is located; 2) If the current source rock is dark mudstone, the ΔlogR method is used to quantitatively predict the TOC content of the dark mudstone layer of the source rock to be tested; 3) If the current source rock is carbonaceous mudstone or coal seam, a univariate linear regression model is used to quantitatively predict the TOC of the source rock to be measured, wherein the univariate linear regression model is obtained by establishing a fitting relationship between the measured TOC of the carbonaceous mudstone or coal seam and the corresponding acoustic time difference; 4) The TOC content predicted by the ΔlogR method and the TOC content predicted by the univariate linear regression are summarized as the TOC content prediction result of the source rock to be tested in the target area.

2. The TOC prediction method for source rocks of different lithologies according to claim 1, characterized in that: The prediction method also includes determining an inflection point value using core profile data, logging data and measured TOC data of the modeled well, wherein the inflection point value is used to indicate a critical value for switching the TOC content prediction method.

3. The TOC prediction method for source rocks of different lithologies according to claim 2, characterized in that: The inflection point value is: selecting the core profile data, logging data and measured TOC data of the modeling well, using the ΔlogR method to model, and obtaining the relationship between TOC and amplitude difference; using the modeling formula determined by the ΔlogR method to quantitatively predict the TOC content of the modeling well, and comparing the predicted result with the measured TOC content, and taking the value of the comparison result difference exceeding the set range as the inflection point value.

4. The TOC prediction method for source rocks of different lithologies according to claim 1, characterized in that: The univariate linear regression model is: Y=D*XE Where Y is the TOC content of the carbonaceous mudstone and coal seam of the source rock to be tested, X is the acoustic time difference of the carbonaceous mudstone and coal seam of the source rock to be tested, and D and E are fitting coefficients.

5. The TOC prediction method for source rocks of different lithologies according to claim 3, characterized in that: The core profile data and measured TOC data are the data after depth calibration.

6. The TOC prediction method for source rocks of different lithologies according to claim 1, characterized in that: The method for predicting TOC in step 2) is: according to the relationship between the acoustic time difference and resistivity in the logging data, the ΔlogR amplitude difference is obtained, a linear fitting relationship between the ΔlogR amplitude difference and TOC is established, and the TOC of the source rock is quantitatively predicted through the linear fitting relationship.

7. The TOC prediction method for source rocks of different lithologies according to claim 3, characterized in that: The set inflection point value ranges from 6.0% to 8.0%.

8. The TOC prediction method for source rocks of different lithologies according to claim 6, characterized in that: The linear fitting relationship between ΔlogR amplitude difference and TOC is: TOC=A×ΔlogR+△TOC Wherein, A is the coefficient of the fitting formula, ΔlogR is the distance after the resistivity and acoustic time difference curves of the source rock to be measured are superimposed, ΔTOC is the organic carbon background value of the non-oil-generating layer, and the calculation formula of ΔlogR is: ΔlgR=lg(R / R 基线 )+k(Δt-Δt 基线 ) Where k is the overlap coefficient; R is the resistivity; R 基线 is the resistivity value corresponding to the baseline; Δt is the acoustic wave time difference; Δt 基线 is the corresponding acoustic time difference value on the baseline.

Images