Method for determining coal rock tar yield by using logging information
By establishing a multiple regression model based on well logging data and using well logging curves such as deep lateral resistivity, compensated density, and sonic transit time to calculate tar yield, the limitations of existing technologies in evaluating medium- to deep coal and rock resources have been overcome, and a simple, fast, and economical method for determining tar yield has been achieved.
Patent Information
- Application Number
- CN202410503278.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-04-25
- Publication Date
- 2025-10-28
AI Technical Summary
Existing methods for obtaining tar yield have limitations in evaluating medium- to deep oil-rich coal resources. In particular, they are complex to operate and rely on coal core sampling yield, making them difficult to apply effectively to areas with deep burial, thick coal seams, and large coal-bearing areas in the Mesozoic Jurassic period.
By utilizing well logging data, coal core data were obtained through low-temperature carbonization experiments using the Gekin method. A multiple regression model of tar yield and well logging curves was established. Using sensitive well logging curves such as deep lateral resistivity, compensated density, compensated neutron, and sonic transit time, a multiple linear regression model was established to calculate the tar yield of coal seams where carbonization experiments were not conducted.
It enables a simple, fast, and economical method to determine coal seam tar yield, reduces analysis costs, improves analysis efficiency, and allows for continuous evaluation of coal seam tar yield. It also compensates for insufficient coring and improves the accuracy of medium- to deep coal and rock resource evaluation.
Smart Images
Figure CN120845005A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of geophysical logging methods in petroleum geology, and is a method for determining coal tar production using logging data. Background Technology
[0002] The Gerkin low-temperature carbonization experiment (GB / 1341-2007) method for obtaining tar yield is currently the only method used both domestically and internationally. This method involves placing a coal sample in a carbonization tube within a Gerkin low-temperature carbonization furnace, heating it to a final temperature of 600℃ according to a prescribed heating program, and holding it at that temperature for a certain period. The yields of tar, pyrolysis water, and semi-coke are then measured. Simultaneously, the semi-coke is compared with a set of standard coke types to determine its type. For highly expansive coals, a certain amount of electrode carbon needs to be added to the coal sample. The coke type is expressed as the minimum amount of electrode carbon (integer grams) required to obtain a coke type consistent with the standard coke type G.
[0003] The Gekin low-temperature carbonization test relies heavily on factors such as coal core recovery rate, and the experimental operation is complex with subjective judgment of coke type. It is suitable for evaluating shallow, thin (10m to over 50m) and small-area Early and Late Tertiary shallow oil-rich coal resources that are easy to sample. However, it has significant limitations in evaluating deep, thick (20m to over 100m) and large-area medium- to deep oil-rich coal resources from the Mesozoic Jurassic period. Summary of the Invention
[0004] This invention provides a method for determining coal tar yield using well logging data, which overcomes the shortcomings of the existing technology and effectively solves the problem that the existing tar yield acquisition methods have significant limitations in evaluating medium- to deep oil-rich coal resources.
[0005] The technical solution of this invention is achieved through the following measures: a method for determining coal tar yield using well logging data, comprising the following steps: Step 1: A low-temperature dry distillation experiment was conducted on the coal core of the coal seam, and experimental data on tar yield, semi-coke yield, total moisture yield, coal gas content and loss content were obtained. Step 2: Obtain the well logging curve values for the coal seam depth segment corresponding to the coal core; Step 3: Through mathematical statistical analysis of the correlation between the tar yield of the coal core experiment and the logging curve, select the most sensitive logging curve that reflects the tar yield of the coal core. Step 4: Perform multiple regression analysis on each tar yield of the coal core and the sensitive logging curve to determine the correlation between the tar yield of the coal core and the sensitive logging curve, and establish a multiple linear regression model for tar yield. Step 5: Obtain the logging curve values of the coal seam section that has undergone conventional logging but not the low-temperature dry distillation experiment of coal rock genomics. Substitute the logging curve values into the multiple linear regression model to obtain the tar yield of the coal seam section.
[0006] The following are further optimizations and / or improvements to the above-mentioned technical solution: Furthermore, both the coal core and the coal rock are low-rank coals. The tar yield, semi-coke yield, and other contents were obtained through Gerkin's low-temperature carbonization experiment. Tar yield is a key parameter for determining whether the coal rock is oil-rich. The contents of other components are negligible, and their volume percentage relationships are as follows: V 焦油产率 +V 半焦产率 +V 总水分产率 +V 煤气+损失 =100 (1) In equation (1), V 焦油产率 This represents the volume percentage of tar yield. V 半焦产率 This represents the volume percentage of semi-coke yield. V 总水分产率 The total moisture content is expressed as a volume percentage. V 煤气+损失 This represents the volume percentage of gas plus losses.
[0007] Furthermore, in step 3, the sensitive logging curves include deep lateral resistivity logging curves, compensated density logging curves, compensated neutron logging curves, and sonic transit time logging curves.
[0008] Furthermore, in step 4, the multiple linear regression model for tar yield includes an explanatory model for tar yield calculated using deep lateral resistivity (RD) and compensated neutron (CNL), an explanatory model for tar yield calculated using deep lateral resistivity (RD) and compensated density (DEN), and an explanatory model for tar yield calculated using deep lateral resistivity (RD) and acoustic transit time (AC).
[0009] Furthermore, the explanation model for calculating tar yield using deep lateral resistivity (RD) and compensated neutron (CNL) is as follows: V 焦油产率 =a+b*RD+c*CNL (2) In equation (2), RD represents deep lateral resistivity; CNL represents compensated neutron; a, b, and c are all undetermined coefficients, which are obtained by fitting through multivariate stepwise regression analysis.
[0010] Furthermore, the interpretation model for calculating tar yield using deep lateral resistivity (RD) and compensated density (DEN) is as follows: V 焦油产率 =d+e*RD+f*DEN (3) In equation (3), RD represents the deep lateral resistivity; DEN represents the compensation density; d, e, and f are all undetermined coefficients, which are obtained by fitting through multivariate stepwise regression analysis.
[0011] Furthermore, the model for calculating tar yield using deep lateral resistivity (RD) and acoustic transit time (AC) is explained as follows: V 焦油产率 =g+h*RD+i*AC (4) In equation (4), RD represents the deep lateral resistivity; AC represents the acoustic transit time; g, h, and i are all undetermined coefficients, which are obtained by fitting through multivariate stepwise regression analysis.
[0012] The method for determining coal seam tar yield using well logging data described in this invention only requires measuring the tar yield of coal cores within the same block. Using the well logging data, a strong correlation between the two sets of data is derived. The tar yield is then calculated by substituting the well logging curve values of the coal seam to be measured into this correlation (interpretive model). This method utilizes well logging curves, which contain abundant and readily available formation information, enabling continuous evaluation of different tar yields in coal seams. It compensates for insufficient core sampling, reduces analysis costs, and improves analysis efficiency. Determining tar yield through well logging data analysis offers advantages such as simplicity, speed, economy, and practicality.
[0013] This invention completely avoids the problems of low coring yield, high coring cost, and numerous controllable factors in pyrolysis experiments in medium-to-deep coal and rock formations. By fully utilizing the abundant single-well logging data from major oilfields in my country, the efficiency of predicting coal tar production parameters in the study area can be greatly improved. Researchers can construct predictive interpretation models that conform to the coal tar production of their study area based on the logging response characteristics of the coal reservoir, ensuring that sufficiently accurate tar production parameters can be obtained, thereby enabling rapid exploration and development research work such as general surveys or detailed surveys of oil-rich coal resources in the study area. Attached Figure Description
[0014] Appendix Figure 1 This is one of the schematic diagrams illustrating the relationship between the coal tar production rate and the corresponding depth-corrected well logging curve response, provided as an embodiment of the present invention.
[0015] Appendix Figure 2 This is the second schematic diagram illustrating the relationship between the coal and rock tar production rate and the corresponding depth-corrected well logging curve response, provided as an embodiment of the present invention.
[0016] Appendix Figure 3 A verification chart was created to show the correlation between the coal rock tar yield calculated using the RD-CNL logging curve and the measured coal core tar yield.
[0017] Appendix Figure 4A verification chart was created to show the correlation between the coal rock tar yield calculated from the RD-DEN logging curve and the measured coal core tar yield.
[0018] Appendix Figure 5 A verification chart was created to show the correlation between the coal rock tar yield calculated from the RD-AC logging curve and the measured coal core tar yield.
[0019] Appendix Figure 6 This is a data table for experimental testing and formula calculation of coal cores.
[0020] Appendix Figure 1 and attached Figure 2 The logging curves described herein are logging data from a certain coal seam. (Attached) Figure 6 Data extracted from the appendix Figure 1 and attached Figure 2 , attached Figure 3 To be continued Figure 5 It is attached Figure 6 Data correlation verification. Detailed Implementation
[0021] The present invention is not limited to the following embodiments, and the specific implementation can be determined according to the technical solution of the present invention and the actual situation.
[0022] The present invention will be further described below with reference to embodiments: Example 1: A method for determining coal tar yield using well logging data, comprising the following steps: Step 1: A low-temperature dry distillation experiment was conducted on the coal core of the coal seam, and experimental data on tar yield, semi-coke yield, total moisture yield, coal gas content and loss content were obtained. Step 2: Obtain the well logging curve values for the coal seam depth segment corresponding to the coal core; Step 3: Through mathematical statistical analysis of the correlation between the tar yield of the coal core experiment and the logging curve, select the most sensitive logging curve that reflects the tar yield of the coal core. Step 4: Perform multiple regression analysis on each tar yield of the coal core and the sensitive logging curve to determine the correlation between the tar yield of the coal core and the sensitive logging curve, and establish a multiple linear regression model for tar yield. Step 5: Obtain the logging curve values of the coal seam section that has undergone conventional logging but not the low-temperature dry distillation experiment of coal rock genomics. Substitute the logging curve values into the multiple linear regression model to obtain the tar yield of the coal seam section.
[0023] Example 2: As an optimization of the above example, both the coal core and the coal rock are low-rank coal. Tar yield, semi-coke yield, and other contents are obtained through Gerkin low-temperature carbonization experiments. Tar yield is a key parameter for determining whether the coal rock is oil-rich. The contents of other components are negligible. The volume percentage relationship is as follows: V 焦油产率 +V 半焦产率 +V 总水分产率 +V 煤气+损失 =100% (1) In equation (1), V 焦油产率 This represents the volume percentage of tar yield. V 半焦产率 This represents the volume percentage of semi-coke yield. V 总水分产率 The total moisture content is expressed as a volume percentage. V 煤气+损失 This represents the volume percentage of gas plus losses.
[0024] Example 3: As an optimization of the above example, in step 3, the sensitive logging curves include deep lateral resistivity logging curves, compensated density logging curves, compensated neutron logging curves, and sonic transit time logging curves.
[0025] Example 4: As an optimization of the above example, in step 4, the multiple linear regression model for tar yield includes an interpretation model for tar yield calculated using deep lateral resistivity (RD) and compensated neutron (CNL), an interpretation model for tar yield calculated using deep lateral resistivity (RD) and compensated density (DEN), and an interpretation model for tar yield calculated using deep lateral resistivity (RD) and acoustic transit time (AC).
[0026] Example 5: As an optimization of Example 4 above, the interpretation model for calculating tar yield using deep lateral resistivity (RD) and compensated neutron (CNL) is as follows: V 焦油产率 =a+b*RD+c*CNL (2) In equation (2), RD represents deep lateral resistivity; CNL represents compensated neutron; a, b, and c are all undetermined coefficients, which are obtained by fitting through multivariate stepwise regression analysis.
[0027] Example 6: As an optimization of Example 4 above, the explanatory model for calculating tar yield using deep lateral resistivity (RD) and compensated density (DEN) is as follows: V 焦油产率 =d+e*RD+f*DEN (3) In equation (3), RD represents the deep lateral resistivity; DEN represents the compensation density; d, e, and f are all undetermined coefficients, which are obtained by fitting through multivariate stepwise regression analysis.
[0028] Example 7: As an optimization of Example 4 above, the interpretation model for calculating tar yield using deep lateral resistivity (RD) and acoustic transit time (AC) is as follows: V 焦油产率=g+h*RD+i*AC (4) In equation (4), RD represents the deep lateral resistivity; AC represents the acoustic transit time; g, h, and i are all undetermined coefficients, which are obtained by fitting through multivariate stepwise regression analysis.
[0029] Implementation Case: 1) A low-temperature dry distillation experiment was conducted on the coal core of a certain coal seam, and experimental data on tar yield, semi-coke yield, total moisture yield, and coal gas + loss were obtained. Figure 6 ).
[0030] The measurement area has a series that can represent the characteristics of the region. This invention conducts experiments according to the standard procedure of "Test Method for Low Temperature Dry Distillation of Coal (GB / 1341-2007)" to obtain the tar yield of each coal core.
[0031] 2) Obtain the logging curve values corresponding to the depth range of the mined coal core, including nine conventional logging curves such as acoustic transit time (AC), compensated density (DEN), compensated neutron (CNL), deep and shallow lateral resistivity (RD, RS), spontaneous potential (SP), borehole caliber (CAL), and natural gamma (GR). Perform environmental correction on these curves to obtain the corrected logging curves (see attached). Figure 1 Appendix Figure 2 ).
[0032] The measurement of logging curves was carried out in accordance with the "Well Logging Engineering Design" document (Oil Exploration Document
[2017] No. 108) and the "Well Logging Design Compilation Specification" (Q / SY01156-2019). Environmental correction of logging curves involved identifying the coal seam using density, natural gamma, and resistivity logging curves, comparing the drill bit diameter and wellbore logging curves of the coal seam section, determining the degree of environmental impact such as enlargement, and selecting appropriate environmental impact correction charts for different series of logging instruments. This process corrected the coal seam logging values to the true standard coal seam logging values, ensuring the accuracy and reliability of the logging data used for coal seam calculations.
[0033] 3) Starting from the study of well logging response mechanism, the correlation between tar yield data and well logging curves was studied by experimental measurement. Four sensitive well logging curves reflecting coal core tar yield were selected from nine conventional well logging curves, namely deep lateral resistivity (RD), compensated density (DEN), compensated neutron (CNL), and acoustic transit time (AC). Using the correlation between these four well logging curves and coal core tar yield, a multiple linear regression model of tar yield was established. Advantages of multiple linear regression model analysis: (1) It can consider the influence of multiple variables at the same time, predict and estimate the dependent variable, which is more effective and more in line with reality than predicting and estimating with only one independent variable; (2) The model has stronger expressive power and can better fit the data.
[0034] By analyzing the correlation between the tar yield of the coal core experiment and the logging curve through mathematical statistics, a sensitive logging curve reflecting the tar yield of the coal core is selected.
[0035] 4) Substituting the environmentally corrected logging parameter values into the sensitive parameter multiple regression, the logging calculation formula for tar production is as follows: (1) Interpretive model for RD and CNL calculation of tar yield V 焦油产率 =a+b*RD+c*CNL (2) R 2 =0.79 Wherein, RD is the deep lateral resistivity; CNL - Compensated Neutron; R 2 - Correlation coefficient.
[0036] (2) Interpretive model for RD and DEN calculation of tar yield V 焦油产率 =d+e*RD+f*DEN (3) R 2 =0.785 Wherein, RD is the deep lateral resistivity; DEN - Compensation density; R 2 - Correlation coefficient.
[0037] (3) Interpretive model for RD and AC calculation of tar yield V 焦油产率 =g+h*RD+i*AC (4) R 2 =0.802 Wherein, RD is the deep lateral resistivity; AC - Acoustic wave time difference; R 2 - Correlation coefficient.
[0038] Where a, b, c, d, e, f, g, h, and i are all undetermined coefficients, which were obtained by fitting through multiple stepwise regression analysis: a=3.492, b=0.00023, c=3.628, d=4.53, e=0.00022, f=1.204, g=13.92, h=0.0002, and i=-0.0173.
[0039] 5) Obtain the logging curve values of the coal seam section that has undergone conventional logging but has not undergone low-temperature carbonization experiment of coal core grit, and substitute the logging curve values into the relationship to obtain the tar yield of the coal seam section.
[0040] Appendix Figure 1 ,2 In the experiment, the measured tar yields (red dots) all fell on or near the curves calculated from the well logging (blue dots, yellow dots, green dots) (see...). Figures 3 to 5 The calculated coal seam tar yield using well logging data from this invention shows a high degree of consistency with the values measured in coal core tests, demonstrating significant application effectiveness. Specific experimental measurements and calculation data can be found in [link to relevant documentation]. Figure 6 ,from Figure 6 As can be seen, the deviations in experimental measurement and calculation data are not significant.
[0041] Figure 3 , 4 Figure 5 shows the correlation verification chart of the coal rock tar yield determined by well logging data and the measured coal core tar yield provided by the present invention. It can be seen that the average correlation coefficient R between the measured and calculated tar yields reaches 0.915, which confirms that the well logging curve regression model has a good correlation in predicting tar yield. The stronger the correlation, the stronger the predictability.
[0042] The coal tar yield calculated using this invention can achieve the same results as those tested in laboratory coal cores according to the Gerkin low-temperature dry distillation test. Moreover, it is more convenient and less costly to obtain the coal tar yield of continuous well sections, and has better application effects.
[0043] The above technical features constitute the embodiments of the present invention, which have strong adaptability and implementation effect. Unnecessary technical features can be added or removed according to actual needs to meet the needs of different situations.
Claims
1. A method for determining coal tar yield using well logging data, characterized in that... Includes the following steps: Step 1: A low-temperature dry distillation experiment was conducted on the coal core of the coal seam, and experimental data on tar yield, semi-coke yield, total moisture yield, coal gas content and loss content were obtained. Step 2: Obtain the well logging curve values for the coal seam depth segment corresponding to the coal core; Step 3: Through mathematical statistical analysis of the correlation between the tar yield of the coal core experiment and the logging curve, select the most sensitive logging curve that reflects the tar yield of the coal core. Step 4: Perform multiple regression analysis on each tar yield of the coal core and the sensitive logging curve to determine the correlation between the tar yield of the coal core and the sensitive logging curve, and establish a multiple linear regression model for tar yield. Step 5: Obtain the logging curve values of the coal seam section that has undergone conventional logging but not the low-temperature dry distillation experiment of coal rock genomics. Substitute the logging curve values into the multiple linear regression model to obtain the tar yield of the coal seam section.
2. The method for determining coal tar yield using well logging data according to claim 1, characterized in that... Both the coal core and the coal petrification were low-rank coals. The tar yield, semi-coke yield, and other contents were obtained through Gerkin's low-temperature carbonization experiment. The volume percentage relationship of these contents is as follows: V 焦油产率 +V 半焦产率 +V 总水分产率 +V 煤气+损失 =100 (1) In equation (1), V 焦油产率 This represents the volume percentage of tar yield. V 半焦产率 This represents the volume percentage of semi-coke yield. V 总水分产率 The total moisture content is expressed as a volume percentage of the total water yield. V 煤气+损失 This represents the volume percentage of gas loss.
3. The method for determining coal tar yield using well logging data according to claim 1 or 2, characterized in that... In step 3, the sensitive logging curves include deep lateral resistivity logging curves, compensated density logging curves, compensated neutron logging curves, and sonic transit time logging curves.
4. The method for determining coal tar yield using well logging data according to claim 1 or 2, characterized in that... In step 4, the multiple linear regression model for tar yield includes an explanatory model for tar yield calculated using deep lateral resistivity and compensated neutrons, an explanatory model for tar yield calculated using deep lateral resistivity and compensated density, and an explanatory model for tar yield calculated using deep lateral resistivity and acoustic transit time.
5. The method for determining coal tar yield using well logging data according to claim 3, characterized in that... In step 4, the multiple linear regression model for tar yield includes an explanatory model for tar yield calculated using deep lateral resistivity and compensated neutrons, an explanatory model for tar yield calculated using deep lateral resistivity and compensated density, and an explanatory model for tar yield calculated using deep lateral resistivity and acoustic transit time.
6. The method for determining coal tar yield using well logging data according to claim 4 or 5, characterized in that... The model for calculating tar yield using deep lateral resistivity and compensated neutrons is explained as follows: V 焦油产率 =a+b*RD+c*CNL (2) In equation (2), RD represents deep lateral resistivity; CNL represents compensated neutron; a, b, and c are all undetermined coefficients, which are obtained by fitting through multivariate stepwise regression analysis.
7. The method for determining coal tar yield using well logging data according to claim 4, characterized in that... The model for calculating tar yield using deep lateral resistivity and compensated density is explained below: V 焦油产率 =d+e*RD+f*DEN (3) In equation (3), RD represents the deep lateral resistivity; DEN represents the compensation density; d, e, and f are all undetermined coefficients, which are obtained by fitting through multivariate stepwise regression analysis.
8. The method for determining coal tar yield using well logging data according to claim 5, characterized in that... The model for calculating tar yield using deep lateral resistivity and compensated density is explained below: V 焦油产率 =d+e*RD+f*DEN (3) In equation (3), RD represents the deep lateral resistivity; DEN represents the compensation density; d, e, and f are all undetermined coefficients, which are obtained by fitting through multivariate stepwise regression analysis.
9. The method for determining coal tar yield using well logging data according to claim 6, characterized in that... The model for calculating tar yield using deep lateral resistivity and compensated density is explained below: V 焦油产率 =d+e*RD+f*DEN (3) In equation (3), RD represents the deep lateral resistivity; DEN represents the compensation density; d, e, and f are all undetermined coefficients, which are obtained by fitting through multivariate stepwise regression analysis.
10. The method for determining coal tar yield using well logging data according to any one of claims 4 to 9, characterized in that... The explanatory model for calculating tar yield using deep lateral resistivity and acoustic transit time is as follows: V 焦油产率 =g+h*RD+i*AC (4) In equation (4), RD represents the deep lateral resistivity; AC represents the acoustic transit time; g, h, and i are all undetermined coefficients, which are obtained by fitting through multivariate stepwise regression analysis.