A method for evaluating oil content of continental shale

Abstract

The present invention provides a method for evaluating the oil content of continental shale, which relates to the technical field of exploration of continental shale oil and can quantitatively characterize the original oil content of shale. The present application obtains the original oil content parameter S1 of shale by cryo-pyrolysis, and obtains conventional logging series data in combination with conventional logging analysis. The key parameters characterizing the oil content are clarified through mathematical and statistical analysis. The S1 calculation function formula based on the key parameters is established. The present invention realizes the calculation of the original oil content parameters of shale, and can be further extended to areas where cryo-pyrolysis has not been carried out, providing more accurate parameters for shale oil exploration and development.

Description

Technical Field

[0001] The invention relates to a method for evaluating the oil content of continental shale, and belongs to the technical field of continental shale oil exploration. Background Art

[0002] Oil content is a very important indicator in the process of shale oil selection and sweet spot selection, which is directly related to the calculation of resources and the evaluation of target layers. Therefore, obtaining accurate shale oil content directly affects the shale oil exploration and development process. At present, the rock pyrolysis method has gradually become one of the main research methods for characterizing shale oil content due to its advantages of simple operation and rapid analysis. At the same time, geochemical indicators established based on pyrolysis S1 (free hydrocarbon) parameters, such as oil saturation index (OSI), are the mainstream methods for evaluating shale movable oil content and identifying shale oil exploitable layers. They have been promoted and applied in many mud shale layers, promoting shale oil exploration and evaluation research. However, the loss of S1 light hydrocarbons is ubiquitous, which is related to factors such as sample lithology, organic matter content, hydrocarbon composition, and storage time. The loss of rock samples under different conditions varies significantly. Cryo-pyrolysis freezes and seals the on-site core with liquid nitrogen, and then conducts rock pyrolysis under the condition of preserving the original oil content as much as possible. The obtained cryo-pyrolysis S1 is closer to the real state. However, cryo-pyrolysis requires the selection of core samples from newly drilled wells for testing, which limits the analysis of its formation scale. Summary of the invention

[0003] In view of the shortcomings of the prior art, in order to obtain more shale oil content data, the present invention provides a method for evaluating the oil content of continental shale. First, a newly drilled well is selected to carry out cryo-pyrolysis of core rocks to obtain S1 data, and then a conventional logging series data body is obtained through conventional logging means. Then, a correlation analysis is carried out between S1 and different series of logging data, and the logging series that can effectively characterize the oil content is screened, and the functional relationship between each logging series and S1 is determined, and then a comprehensive functional relationship between S1 and the logging data is established. The parameter values ​​are obtained through multivariate regression to obtain the final formula for calculating S1. Relying on this formula, continuous evaluation of S1 in the vertical direction can be achieved in a single well.

[0004] The technical solution of the present invention is as follows:

[0005] A method for evaluating the oil content of continental shale comprises the following steps:

[0006] In the first step, new wells were selected in the study area to carry out systematic sampling work, and closed coring was performed at the drilling site to prevent the loss of light hydrocarbons;

[0007] The second step is to use a cryogenic treatment method on the sealed core samples on site to crush the samples into 200-mesh powder samples under low temperature conditions. The whole process is kept sealed and liquid nitrogen is continuously injected to minimize the loss of light hydrocarbons.

[0008] The third step is to conduct rock pyrolysis analysis on all powder samples to obtain the sample oil content S1 data in mg / g, which is recorded as y;

[0009] The fourth step is to extract the corresponding depth logging data according to the depth of all samples for pyrolysis analysis. The logging data include natural gamma GR, natural potential SP, well diameter CAL, deep lateral resistivity RD, shallow lateral resistivity RS, acoustic time difference AC, neutron CNL and density DEN, which are recorded as x 1 , x 2 , x 3 …x i , i is the number of logging series;

[0010] Based on the above process, the oil content y of samples at the same depth and its corresponding logging data are combined into a set of related data (the oil content data is mainly the oil content y, and the logging data is mainly x). 1 , x 2 , x 3 …x i ), we can proceed to the next step, conduct correlation analysis, and select the main controlling factors of shale oil content through single factor analysis and multi-factor coupling;

[0011] The fifth step is to conduct correlation analysis between oil content and conventional logging data based on statistical software, that is, to use mathematical statistics software to calculate the linear correlation between oil content and different logging data in turn, and record the correlation coefficients between different logging data and oil content, which are respectively calculated as R 1 , R 2 , R 3 …R i , R is the correlation coefficient, i is the number of logging series;

[0012] The sixth step is single factor analysis. According to the correlation coefficient R between the recorded S1 and the logging data, the parameters with obvious correlation are selected as the characterization parameters that can characterize the original oil content of shale. Generally, the correlation coefficient above 0.7 indicates that the relationship is very close, so R is determined. 2 If the correlation is greater than 0.5, it is obvious, and the characterization parameter is recorded as z 1 , z 2 , z 3 …z j , j is the number of parameters screened;

[0013] The seventh step is multi-factor coupling. The functional relationship between oil content and characterization parameters is established through mathematical statistics software. The function is a linear function, a polynomial function, a power function, an exponential function or a combination of the above different types of functions. The following formula represents it:

[0014] y=f(z 1 , z 2, z 3 …z j , a, b, c…m)

[0015] Where a, b, c…m are the constants to be determined in the functional relationship;

[0016] Step 8: Based on the oil content and characterization parameter data of all groups, specifically y 1 , z 11 , z 21 , z 31 …z j1 ;y 2 , z 12 , z 22 , z 32 …z j2 ;y 3 , z 13 , z 23 , z 33 …z j3 ;…y k , z 1k , z 2k , z 3k …z jk ; k is the number of data groups, and the equation system is established:

[0017] y 1 =f(z 11 , z 21 , z 31 …z j1 , a, b, c…m)

[0018] y 2 =f(z 12 , z 22 , z 32 …z j2 , a, b, c…m)

[0019] y 3 =f(z 13 , z 23 , z 33 …z j3 , a, b, c…m)

[0020] …

[0021] y k =f(z 1k , z 2k , z 3k …z jk , a, b, c…m)

[0022] Iterative regression is used to obtain the unknown constants a, b, c…m in the function. Based on the known equations, the Macquart method is used to perform iterative regression, and the convergence judgment index is set to 10. -10 , set the maximum number of iterations to 1000, and obtain the undetermined constants in the equations through regression iteration to obtain the following functional relationship:

[0023] y=f(z 1 , z 2 , z 3 …z j )

[0024] Step 9: Based on the functional relationship determined in step 8, the different logging data at the same depth are substituted into the formula in turn to obtain the calculated oil content S1 at that depth. 计算1 , S1 计算2 , S1 计算3 …S1 计算k , k is the number of data groups;

[0025] Step 10: Use mathematical statistics software to calculate two columns of data, oil content S1 计算 The linear correlation between the measured oil content S1 verifies the reliability of the function, and the correlation coefficient > 0.9 is the verification standard for reliability.

[0026] Preferably, in the second step, the samples taken on site are placed in a portable low-temperature storage box, and the temperature inside the box is controlled to be below -40°C.

[0027] Preferably, in the third step, the sample after low temperature treatment is placed in a Rock-eval 6 rock pyrolysis instrument for testing.

[0028] Preferably, in the fifth step, the statistical software is EXCEL or SPSS mathematical statistics software.

[0029] Preferably, in step 8, the convergence judgment index is set to 10 -10 , the maximum number of iterations is 1000.

[0030] This application systematically analyzes the correlation between oil content and conventional logging, and then selects logging data that can characterize oil content to calculate oil content. The calculation is performed for objects with basically similar thermal evolution degree (Ro). The calculation result is not affected by Ro, and the continuous characterization of oil content can be achieved on a single well, rather than just calculation based on a small amount of data. This application explains how to achieve continuous testing of oil content in uncored sections and uncored wells by establishing a functional relationship between oil content S1 and conventional logging series data.

[0031] The beneficial effects of the present invention are:

[0032] The present invention obtains the original oil content parameter S1 of shale by cryo-pyrolysis, and obtains conventional logging series data by combining conventional logging analysis. The main controlling factors controlling oil content are clarified by mathematical and statistical analysis. The S1 calculation function formula based on the main controlling factors is established. The present invention realizes the calculation of the original oil content parameters of shale, and can be further extended to areas where cryo-pyrolysis has not been carried out, providing more accurate parameters for shale oil exploration and development. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is the intersection diagram of calculated S1 and measured S1. DETAILED DESCRIPTION

[0034] The present invention will be further described below by way of embodiments in conjunction with the accompanying drawings, but is not limited thereto.

[0035] Embodiment 1:

[0036] A method for evaluating the oil content of continental shale comprises the following steps:

[0037] In the first step, new wells were selected in the study area to carry out systematic sampling work, and closed coring was performed at the drilling site to prevent the loss of light hydrocarbons;

[0038] The second step is to use a low-temperature treatment method for the sealed core samples taken on site. The samples taken on site are placed in a portable low-temperature storage box with the temperature inside the box controlled below -40°C. The samples are crushed into 200-mesh powder samples under low-temperature conditions. The whole process is kept sealed and liquid nitrogen is continuously injected to minimize the loss of light hydrocarbons.

[0039] The third step is to conduct rock pyrolysis analysis on all powder samples. The samples after low temperature treatment are placed in Rock-eval 6 rock pyrolysis instrument for testing to obtain the oil content S1 data of the samples in mg / g, which is recorded as y.

[0040] The fourth step is to extract the corresponding depth logging data according to the depth of all samples for pyrolysis analysis. The logging data include natural gamma GR, natural potential SP, well diameter CAL, deep lateral resistivity RD, shallow lateral resistivity RS, acoustic time difference AC, neutron CNL and density DEN, which are recorded as x 1 , x 2 , x 3 …x i , i is the number of logging series;

[0041] Based on the above process, the oil content y of samples at the same depth and its corresponding logging data are combined into a set of related data (the oil content data is mainly the oil content y, and the logging data is mainly x). 1 , x 2 , x 3 …x i), we can proceed to the next step, conduct correlation analysis, and select the main controlling factors of shale oil content through single factor analysis and multi-factor coupling;

[0042] The fifth step is to carry out the correlation analysis between oil content and conventional logging data based on statistical software, such as EXCEL and SPSS mathematical statistics software, that is, to use mathematical statistics software to calculate the linear correlation between oil content and different logging data in turn, and record the correlation coefficients between different logging data and oil content, which are respectively calculated as R 1 , R 2 , R 3 …R i , R is the correlation coefficient, i is the number of logging series;

[0043] The sixth step is single factor analysis. According to the correlation coefficient R between the recorded S1 and the logging data, the parameters with obvious correlation are selected as the characterization parameters that can characterize the original oil content of shale. Generally, the correlation coefficient above 0.7 indicates that the relationship is very close, so R is determined. 2 If the correlation is greater than 0.5, it is obvious, and the characterization parameter is recorded as z 1 , z 2 , z 3 …z j , j is the number of parameters screened;

[0044] The seventh step is multi-factor coupling. The functional relationship between oil content and characterization parameters is established through mathematical statistics software. The function is a linear function, a polynomial function, a power function, an exponential function or a combination of the above different types of functions. The following formula represents it:

[0045] y=f(z 1 , z 2 , z 3 …z j , a, b, c…m)

[0046] Where a, b, c…m are the constants to be determined in the functional relationship;

[0047] Step 8: Based on the oil content and characterization parameter data of all groups, specifically y 1 , z 11 , z 21 , z 31 …z j1 ;y 2 , z 12 , z 22 , z 32 …z j2 ;y 3 , z 13 , z 23 , z 33 …z j3 ;…yk , z 1k , z 2k , z 3k …z jk ; k is the number of data groups, and the equation system is established:

[0048] y 1 =f(z 11 , z 21 , z 31 …z j1 , a, b, c…m)

[0049] y 2 =f(z 12 , z 22 , z 32 …z j2 , a, b, c…m)

[0050] y 3 =f(z 13 , z 23 , z 33 …z j3 , a, b, c…m)

[0051] …

[0052] y k =f(z 1k , z 2k , z 3k …z jk , a, b, c…m)

[0053] Iterative regression is used to obtain the unknown constants a, b, c…m in the function. Based on the known equations, the Macquart method is used to perform iterative regression, and the convergence judgment index is set to 10. -10 , set the maximum number of iterations to 1000, and obtain the undetermined constants in the equations through regression iteration to obtain the following functional relationship:

[0054] y=f(z 1 , z 2 , z 3 …z j )

[0055] Step 9: Based on the functional relationship determined in step 8, the different logging data at the same depth are substituted into the formula in turn to obtain the calculated oil content S1 at that depth. 计算1 , S1 计算2 , S1 计算3 …S1 计算k , k is the number of data groups;

[0056] Step 10: Use mathematical statistics software to calculate two columns of data, oil content S1 计算The linear correlation between the measured oil content S1 verifies the reliability of the function, and the correlation coefficient > 0.9 is the verification standard for reliability.

[0057] Experimental Example 1

[0058] In order to verify the feasibility of the method, 88 samples from the Funing Formation in the Subei Basin were selected for verification. The frozen S1 and conventional logging series data of the samples were obtained by cryo-pyrolysis and conventional logging analysis, as shown in Table 1.

[0059] Table 1 Statistics of Funing Formation sample S1 and logging data

[0060]

[0061]

[0062]

[0063] By analyzing the correlation between S1 and GR, RD, RS, and AC, it was found that GR and AC had a good correlation with S1, and R 2 All of them are greater than 0.5, which are the characterization parameters of oil content. The following functional relationship is established through mathematical statistics software:

[0064] S1=a+b*GR+c*GR^2+d*GR^3+e*GR^4+f*GR^5+g*AC+h*AC^2+i*AC^3

[0065] Based on the regression analysis of the existing data, the following parameter table is obtained:

[0066] a 379.2303583 b -28.55386458 c 0.763139537 d -0.009911951 e 6.27E-05 f -1.55E-07 g 0.366710462 h -0.001149854 i 1.18E-06

[0067] In order to verify the reliability of the test results, a comparative analysis was conducted between the calculated results and the measured results. It can be seen from the intersection diagram (such as Figure 1 As shown in the figure, the calculated S1 has a very good correlation with the measured S1, indicating that the original oil content parameters of shale can be effectively obtained by this method.

Claims

1. A method for evaluating the oil content of continental shale, characterized in that: The steps include: In the first step, new wells were selected in the study area to carry out systematic sampling work and closed coring was performed at the drilling site; In the second step, the sealed core samples were treated with a low temperature method to crush the samples into 200-mesh powder samples under low temperature conditions. The whole process was kept sealed and continuously flushed with liquid nitrogen. The third step is to conduct rock pyrolysis analysis on all powder samples to obtain the sample oil content S1 data in mg / g, which is recorded as y; The fourth step is to extract the corresponding depth logging data according to the depth of all samples for pyrolysis analysis, and record them as x1, x2, x3…x i , i is the number of logging series; Based on the above process, the oil content y of samples at the same depth and their corresponding logging data are combined into a set of associated data; The fifth step is to carry out the correlation analysis between oil content and conventional logging data based on statistical software. The linear correlation between oil content and different logging data is calculated in sequence using mathematical statistics software. The correlation coefficients between different logging data and oil content are recorded as R1, R2, R3…R i , R is the correlation coefficient, i is the number of logging series; The sixth step is single factor analysis. According to the correlation coefficient R between the recorded S1 and the logging data, the parameters with obvious correlation are selected as the characterization parameters that can characterize the original oil content of shale. 2 The correlation is obvious when it is greater than 0.5, and the characterization parameters are recorded as z1, z2, z3…z j , j is the number of parameters screened; The seventh step is multi-factor coupling. The functional relationship between oil content and characterization parameters is established through mathematical statistics software. The function is a linear function, a polynomial function, a power function, an exponential function or a combination of the above different types of functions. The following formula represents it: y=f(z1,z2,z3…z j ,a,b,c…m) Where a, b, c…m are the constants to be determined in the functional relationship; Step 8: Based on the oil content and characterization parameter data of all groups, specifically y1, z 11 , z 21 , z 31 …z j1 ;y2,z 12 , z 22 , z 32 …z j2 ;y3,z 13 , z 23 , z 33 …z j3 ;…y k , z 1k , z 2k , z 3k …z jk ; k is the number of data groups, and the equation system is established: y1=f(z 11 ,from 21 ,from 31 ...from j1 ,a,b,c…m) y2=f(z 12 ,z 22 ,z 32 …z j2 ,a,b,c…m) y3=f(z 13 ,z 23 ,z 33 …z j3 ,a,b,c…m) … y k =f(z 1k ,z 2k ,z 3k …z jk ,a,b,c…m) Iterative regression is used to obtain the unknown constants a, b, c…m in the function. Based on the known system of equations, the McQuart method is used to perform iterative regression, set the convergence judgment index, set the maximum number of iterations, and obtain the unknown constants in the system of equations through regression iteration to obtain the following functional relationship: y=f(z1,z2,z3…z j ) Step 9: Based on the functional relationship determined in step 8, the different logging data at the same depth are substituted into the formula in turn to obtain the calculated oil content S1 at that depth. 计算1 , S1 计算2 , S1 计算3 …S1 计算k , k is the number of data groups; Step 10: Use mathematical statistics software to calculate two columns of data, oil content S1 计算 The linear correlation between the measured oil content S1 verifies the reliability of the function, and the correlation coefficient > 0.9 is the verification standard for reliability.

2. The method for evaluating the oil content of continental shale according to claim 1, characterized in that: In the second step, the samples taken on site are placed in a portable low-temperature storage box, and the temperature inside the box is controlled below -40°C.

3. The method for evaluating the oil content of continental shale according to claim 1, characterized in that: In the third step, the low-temperature treated samples were placed in a Rock-eval 6 rock pyrolyzer for testing.

4. The method for evaluating oil content of continental shale according to claim 1, characterized in that: In the fourth step, the logging data include natural gamma GR, natural potential SP, wellbore CAL, deep lateral resistivity RD, shallow lateral resistivity RS, acoustic time difference AC, neutron CNL and density DEN, which are recorded as x1, x2, x3…x i , i is the logging series number.

5. The method for evaluating the oil content of continental shale according to claim 1, characterized in that: In the fifth step, the statistical software is EXCEL and SPSS mathematical statistics software.

6. The method for evaluating the oil content of continental shale according to claim 1, characterized in that: In the eighth step, set the convergence judgment index to 10 -10 , the maximum number of iterations is 1000.

Images