A method and system for predicting gas content in coal-bearing formations through well logging and inversion.

Abstract

This invention provides a method and system for predicting gas content in coal-bearing formations through well logging and inversion. The method includes predicting coal seam gas content using sonic transit time, deep lateral resistivity, and natural gamma ray logging values, and establishing a gas content well logging prediction model through data regression analysis; validating the gas content well logging prediction model using measured coal seam gas content at sampling points; inverting the target exploration interval using seismic and borehole data to obtain a data volume of sonic transit time, deep lateral resistivity, and natural gamma ray logging values; extracting attribute values ​​from the above data volume to obtain attribute values ​​of the target layer plane; and substituting the attribute values ​​of the target layer plane into the validated gas content well logging prediction model to obtain the plane variation characteristics of gas content in the target exploration interval. Applying this invention overcomes the shortcomings of existing technologies, enabling accurate and effective prediction of gas content in coal-bearing formations with high speed and efficiency, allowing for large-scale application.

Description

Technical Field

[0001] This invention relates to the field of coal-bearing formation gas volume prediction technology, specifically to a method for predicting coal-bearing formation gas content through well logging and inversion, and a system for applying this method. Background Technology

[0002] Coal constitutes a large proportion of my country's energy resources and plays a significant role in total energy consumption. With continuous socio-economic development, the clean use of coal is increasingly becoming mainstream, and the use of coalbed methane is an important method of clean coal utilization. In this context, accurate prediction of coalbed methane demand has become a crucial element in the clean use of coal resources.

[0003] Currently, most methods for evaluating the gas content of coal-bearing strata use bivariate methods such as acoustic transit time, density, or deep lateral and spontaneous potential to predict the gas content at a single point.

[0004] Therefore, the main problems and shortcomings of existing technologies are as follows:

[0005] 1. Using only one or two logging methods to predict gas content results in a large error.

[0006] 2. Traditional methods only use boreholes to measure the gas content in the test area at points, and then use multiple points for interpolation to obtain planar characteristics. This ignores the heterogeneity of the strata and often results in a large deviation from the actual situation. Summary of the Invention

[0007] In order to overcome the shortcomings of the prior art, the purpose of this invention is to provide a method and system for predicting the gas content of coal-bearing formations through well logging and inversion. This method and system can solve the problems of large errors and low accuracy in the prior art. It can accurately and effectively predict the gas content of coal-bearing formations, and the prediction speed is fast and the efficiency is high, which can realize large-scale application.

[0008] To solve the above problems, the technical solution adopted by the present invention is as follows:

[0009] A method for predicting gas content in coal-bearing formations through well logging and inversion includes the following steps: predicting coal seam gas content using sonic transit time, deep lateral resistivity, and natural gamma ray logging values, and establishing a gas content well logging prediction model through data regression analysis; validating the gas content well logging prediction model using measured coal seam gas content at sampling points; inverting the target exploration interval using seismic and borehole data to obtain a data volume of sonic transit time, deep lateral resistivity, and natural gamma ray logging values; extracting attribute values ​​from the above data volume to obtain attribute values ​​of the target layer plane; and substituting the attribute values ​​of the target layer plane into the validated gas content well logging prediction model to obtain the plane variation characteristics of gas content in the target exploration interval.

[0010] A further approach is to predict coal seam gas content using sonic transit time, deep lateral resistivity, and natural gamma logging values, and to establish a gas content logging prediction model through data regression analysis. This includes: using borehole sonic transit time, deep lateral resistivity, and natural gamma logging values ​​as variables AC, RD, and GR, and a preset test gas content value as the target value; obtaining the coefficients a, b, and c of the three variables, as well as the constant d, through data regression analysis; and establishing a gas content logging prediction model, expressed as formula (1):

[0011] G=a*AC+b*RD+c*GR+d (1).

[0012] A further approach involves validating the gas content prediction model using the measured coal seam gas content at sampling points. This includes: validating the gas content prediction model using the measured coal seam gas content by substituting the sonic transit time, deep lateral resistivity, and natural gamma logging values ​​into the gas content prediction model to obtain the corresponding predicted gas content value; dividing the absolute value of the difference between the predicted and measured gas content values ​​by the measured gas content value to obtain the prediction error; if the prediction error is large, the model is reconstructed; if the prediction error is within the allowable range, the gas content prediction model passes the validation.

[0013] A further approach involves using seismic and borehole data to invert the target exploration section, including: creating a synthetic record using borehole logging data, establishing a low-frequency model, and directly inverting the target exploration section using inversion software to obtain three data volumes: sonic transit time, deep lateral resistivity, and natural gamma logging values.

[0014] A further approach involves extracting the attribute values ​​of the target layer plane from the aforementioned data volume, which includes: first, interpreting the target layer and difference values ​​in the seismic data volume; and then extracting the amplitude attribute values ​​of the target layer from the three data volumes obtained for sonic transit time, deep lateral resistivity, and natural gamma logging, respectively. These are the plane attribute values ​​of the sonic transit time, deep lateral resistivity, and natural gamma logging values ​​corresponding to the target layer.

[0015] A further proposed approach is to define the planar variation characteristics of gas content in the target exploration layer as including the distribution range of high gas content areas and the distribution range of areas with gas content below a preset lower limit.

[0016] A further proposed solution is that the gas content range of the high gas content area is 1.8-2.5 m3 / t, and the gas content range of the area with a gas content below the preset lower limit is 0.75-1.25 m3 / t.

[0017] Therefore, compared with the existing technology, the present invention integrates three logging methods closely related to gas content, and predicts the gas content of coal-bearing formations by using sonic transit time, deep lateral resistivity and natural gamma, as well as seismic inversion. Furthermore, by combining seismic inversion, the prediction method is changed from single-point prediction to area prediction, which can achieve high-precision prediction of gas content in coal-bearing formations.

[0018] A system for predicting gas content in coal-bearing formations through well logging and inversion is disclosed. This system is applied to the aforementioned method for predicting gas content in coal-bearing formations through well logging and inversion. The system includes: a model building unit for predicting coal seam gas content using sonic transit time, deep lateral resistivity, and natural gamma ray logging values, and establishing a gas content well logging prediction model through data regression analysis; a model validation unit for validating the gas content well logging prediction model using measured coal seam gas content at sampling points; an inversion unit for inverting the target exploration interval using seismic and borehole data to obtain a data volume of sonic transit time, deep lateral resistivity, and natural gamma ray logging values; an extraction unit for extracting attribute values ​​from the data volume to obtain attribute values ​​for the target exploration interval's plane; and a result output unit for substituting the attribute values ​​of the target layer plane into the validated gas content well logging prediction model to obtain the plane variation characteristics of gas content in the target exploration interval.

[0019] Therefore, it can be seen that the present invention uses a prediction system composed of a model building unit, a model verification unit, an inversion unit, an extraction unit, and a result output unit to predict the gas content of coal-bearing strata. It can accurately and effectively predict the gas content of coal-bearing strata, and the prediction speed is fast and the efficiency is high, which can realize large-scale application.

[0020] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments. Attached Figure Description

[0021] Figure 1 This is a flowchart of an embodiment of a method for predicting gas content in coal-bearing formations through well logging and inversion according to the present invention.

[0022] Figure 2 This is a planar variation characteristic map of gas content in a certain area in an embodiment of a method for predicting gas content in coal-bearing strata through well logging and inversion according to the present invention.

[0023] Figure 3 This is a schematic diagram of a system embodiment of the present invention for predicting gas content in coal-bearing formations through well logging and inversion. Detailed Implementation

[0024] An example of a method for predicting gas content in coal-bearing formations through well logging and inversion:

[0025] like Figure 1 As shown, a method for predicting gas content in coal-bearing formations through well logging and inversion includes the following steps:

[0026] Step S1: The gas content of the coal seam is predicted by sonic transit time (AC), deep lateral resistivity (RD), and natural gamma logging value (GR), and a gas content logging prediction model is established through data regression analysis.

[0027] Step S2: Validate the gas content prediction model for well logging by using the measured gas content of coal seams at the sampling points.

[0028] Step S3: Use seismic data and borehole data to invert the target stratum to obtain data volumes of sonic transit time, deep lateral resistivity and natural gamma logging values.

[0029] Step S4: Extract the attribute values ​​of the target layer plane from the above data volume through attribute extraction.

[0030] Step S5: Substitute the attribute values ​​of the target layer plane into the verified gas content logging prediction model to obtain the gas content plane variation characteristics of the target layer.

[0031] In step S1 above, the gas content of the coal seam is predicted by sonic transit time, deep lateral resistivity and natural gamma logging values, and a gas content logging prediction model is established through data regression analysis. This includes: using borehole sonic transit time, deep lateral resistivity and natural gamma logging values ​​as variables AC, RD and GR, and using a preset test gas content value as the target value, the coefficients a, b, and c of the three variables and the constant d are obtained through data regression analysis, and a gas content logging prediction model is established, expressed as formula (1):

[0032] G=a*AC+b*RD+c*GR+d (1)

[0033] Among them, AC, RD, and GR are the sonic transit time, deep lateral resistivity, and natural gamma logging values, respectively, while a, b, c, and d are empirical values ​​that are adjusted appropriately according to different regions.

[0034] In step S2 above, the gas content logging prediction model is validated using the measured gas content of the coal seam at the sampling point. This includes: validating the gas content logging prediction model using the measured gas content of the coal seam by substituting the sonic transit time, deep lateral resistivity, and natural gamma logging values ​​into the gas content logging prediction model to obtain the corresponding predicted gas content value; dividing the absolute value of the difference between the predicted gas content value and the measured gas content value by the measured gas content value to obtain the prediction error; if the prediction error is large, the model is reconstructed; if the prediction error is within the allowable range, the gas content logging prediction model passes the validation.

[0035] In step S3 above, seismic data and borehole data are used to invert the target exploration section, including: using borehole logging data to create a synthetic record, then establishing a low-frequency model, and directly inverting the target exploration section through inversion software to obtain three data volumes: sonic transit time, deep lateral resistivity, and natural gamma logging values.

[0036] In step S4 above, the attribute values ​​of the target layer plane are obtained by extracting the attributes of the data volume, including: first, interpreting the target layer and difference value in the seismic data volume, and extracting the amplitude attribute value of the target layer from the three data volumes of sonic transit time, deep lateral resistivity and natural gamma logging value respectively, which are the plane attribute values ​​of sonic transit time, deep lateral resistivity and natural gamma logging value corresponding to the target layer respectively.

[0037] In this embodiment, the planar variation characteristics of gas content in the target stratum include the distribution range of high gas content areas and the distribution range of areas with gas content below a preset lower limit.

[0038] Among them, the gas content range of the high gas content area is 1.8-2.5 m3 / t, and the gas content range of the area below the preset lower limit is 0.75-1.25 m3 / t.

[0039] In practical applications, this invention provides a method for predicting gas content in coal-bearing formations through well logging and inversion, which mainly includes the following steps:

[0040] First, the gas content of coal-bearing formations is predicted using sonic transit time, deep lateral resistivity, and natural gamma ray. The main method is as follows: using the sonic transit time, deep lateral resistivity, and natural gamma ray logging values ​​from borehole A1 as variables, and the gas content measured in the laboratory as the target value, the coefficients a1, b1, and c1 of the three variables, as well as the constant d1, are obtained through data regression analysis. This yields a predictive model for evaluating the gas content of coal-bearing formations using conventional logging data from borehole A1: G = a1*AC + b1*RD + c1*GR + d1.

[0041] Next, the model was validated using measured gas content. As shown in Table 3, the AC (acoustic transit time), RD (deep lateral resistivity), and GR (natural gamma) values ​​were substituted into the gas content logging prediction model to obtain the corresponding predicted gas content values. The prediction error was obtained by dividing the absolute value of the difference between the predicted and measured gas content by the measured gas content value. If the error was large, the model was reconstructed. If the error was within the allowable range, the model was used throughout the region.

[0042] Then, inversion is performed using seismic data and borehole data. First, a synthetic record is created using borehole logging data. Then, a low-frequency model is established. Using professional inversion software (with mature inversion methods), the inversion of the target borehole segment can be directly achieved, obtaining three data volumes: AC, RD, and GR.

[0043] Then, the three attribute values ​​of the target layer plane are obtained through attribute extraction: First, the target layer (such as T1) is interpreted and interpolated in the seismic data volume. The amplitude attribute values ​​of the T1 layer are extracted from the obtained AC, RD and GR data volumes, which are the AC, RD and GR plane attribute values ​​corresponding to the T1 layer.

[0044] Finally, by substituting the target layer's planar attribute values ​​into the aforementioned prediction model, the planar variation characteristics of gas content in the target layer are obtained, mainly including the distribution range of high gas content areas (e.g., Figure 2 The yellow and orange areas shown represent gas content ranging from 1.8 to 2.5 m³ / t, and the distribution range of areas with gas content below a certain lower limit (e.g., Figure 1 The dark green area shown has a gas content ranging from 0.75 to 1.25 m³ / t.

[0045] In summary, based on the above methods and using the existing data from borehole A1 for analysis and calculation, a well logging prediction model for gas content in a certain area was obtained. This prediction model is expressed as: G = -0.009064*AC + 0.020824*RD - 0.006207*GR + 5.8734223.

[0046] Tables 1 and 2 show the gas content and error statistics of borehole A1 calculated using conventional bivariate methods (AC and DEN, RD and SP), respectively. As can be seen from the two tables, the gas content calculated by these two methods has large errors, at 13.45% and 14.26% respectively, and the stability of the calculation results is also poor.

[0047] Table 1. Statistics on Gas Content Prediction and Errors Using Conventional Methods

[0048]

[0049] Table 2. Statistics on Predicted Gas Content and Errors Using Conventional Methods (Table 2)

[0050]

[0051] Table 3 shows the gas content and error statistics of borehole A1 calculated using the method of this invention, such as G = -0.009064*AC + 0.020824*RD - 0.006207*GR + 5.8734223. It can be seen from the calculation results and error statistics that the calculated gas content and error statistics for the same layer in borehole A1 show relatively small errors, with a maximum error of 14.45% and an average error of 6.85%.

[0052] Table 3. Predicted Gas Content and Error Statistics of the Invention

[0053]

[0054]

[0055] Figure 2 The figure shows the planar variation characteristics of gas content in a certain area obtained by the method of the present invention. As can be seen from the figure, the gas content in the area varies from 0.5 to 2.5 m3 / t, mainly concentrated in the range of 0.75 to 2.0 m3 / t. The planar distribution characteristics are clear, and the error with the measured gas content is small.

[0056] Therefore, compared with the existing technology, the present invention integrates three logging methods closely related to gas content, and predicts the gas content of coal-bearing formations by using sonic transit time, deep lateral resistivity and natural gamma, as well as seismic inversion. Furthermore, by combining seismic inversion, the prediction method is changed from single-point prediction to area prediction, which can achieve high-precision prediction of gas content in coal-bearing formations.

[0057] An example of a system for predicting gas content in coal-bearing formations through well logging and inversion:

[0058] A system for predicting gas content in coal-bearing formations through well logging and inversion is provided. This system is applied to the aforementioned method for predicting gas content in coal-bearing formations through well logging and inversion. Figure 3 As shown, the system includes:

[0059] Model building unit 1 is used to predict the gas content of coal seams by using sonic transit time, deep lateral resistivity and natural gamma logging values, and to establish a gas content logging prediction model through data regression analysis.

[0060] Model verification unit 2 verifies the gas content well logging prediction model by using the measured gas content of coal seams at sampling points.

[0061] Inversion Unit 3 uses seismic and borehole data to invert the target strata and obtain data volumes of sonic transit time, deep lateral resistivity, and natural gamma logging values.

[0062] Extraction unit 4 is used to extract the attribute values ​​of the target layer plane from the above data volume through attribute extraction.

[0063] Result output unit 5 is used to substitute the attribute values ​​of the target layer plane into the verified gas content logging prediction model to obtain the gas content plane variation characteristics of the exploration target layer.

[0064] In model building unit 1, the gas content of coal seams is predicted by sonic transit time, deep lateral resistivity and natural gamma logging values. A gas content logging prediction model is established through data regression analysis, including: using borehole sonic transit time, deep lateral resistivity and natural gamma logging values ​​as variables AC, RD and GR, and using the preset test gas content as the target value, the coefficients a, b, c of the three variables and the constant d are obtained through data regression analysis to establish the gas content logging prediction model, expressed as formula (1):

[0065] G=a*AC+b*RD+c*GR+d (1)

[0066] Among them, AC, RD, and GR are the sonic transit time, deep lateral resistivity, and natural gamma logging values, respectively, while a, b, c, and d are empirical values ​​that are adjusted appropriately according to different regions.

[0067] In model verification unit 2, the gas content logging prediction model is verified using the measured gas content of the coal seam at the sampling points. This includes: verifying the gas content logging prediction model using the measured gas content of the coal seam by substituting the sonic transit time, deep lateral resistivity, and natural gamma logging values ​​into the gas content logging prediction model to obtain the corresponding predicted gas content value; dividing the absolute value of the difference between the predicted gas content value and the measured gas content value by the measured gas content value to obtain the prediction error; if the prediction error is large, the model is reconstructed; if the prediction error is within the allowable range, the gas content logging prediction model passes the verification.

[0068] In inversion unit 3, seismic data and borehole data are used to invert the target exploration section, including: using borehole logging data to create a synthetic record, then establishing a low-frequency model, and directly inverting the target exploration section through inversion software to obtain three data volumes: sonic transit time, deep lateral resistivity, and natural gamma logging values.

[0069] In extraction unit 4, attribute values ​​of the target layer plane are obtained by extracting the above data volume through attribute extraction, including: first, interpreting the target layer and difference value in the seismic data volume, and extracting the amplitude attribute value of the target layer in the three data volumes of sonic transit time, deep lateral resistivity and natural gamma logging value respectively, which are the plane attribute values ​​of sonic transit time, deep lateral resistivity and natural gamma logging value corresponding to the target layer respectively.

[0070] In this embodiment, the planar variation characteristics of gas content in the target stratum include the distribution range of high gas content areas and the distribution range of areas with gas content below a preset lower limit.

[0071] Among them, the gas content range of the high gas content area is 1.8-2.5 m3 / t, and the gas content range of the area below the preset lower limit is 0.75-1.25 m3 / t.

[0072] Therefore, it can be seen that the present invention uses a prediction system composed of a model building unit 1, a model verification unit 2, an inversion unit 3, an extraction unit 4, and a result output unit 5 to predict the gas content of coal-bearing strata. It can accurately and effectively predict the gas content of coal-bearing strata, and the prediction speed is fast and the efficiency is high, which can realize large-scale application.

[0073] The above embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of protection of the present invention. Any non-substantial changes and substitutions made by those skilled in the art based on the present invention shall fall within the scope of protection claimed by the present invention.

Claims

1. A method for predicting gas content in coal-bearing formations through well logging and inversion, characterized in that, Includes the following steps: The gas content of coal seams is predicted by sonic transit time, deep lateral resistivity and natural gamma logging values. A gas content logging prediction model is established through data regression analysis. The model includes: using borehole sonic transit time, deep lateral resistivity and natural gamma logging values ​​as variables AC, RD and GR, and a preset test gas content value as the target value. Through data regression analysis, the coefficients a, b, c of the three variables and the constant d are obtained to establish the gas content logging prediction model, which is expressed as formula (1): （1） The gas content logging prediction model is validated using the measured gas content of coal seams at sampling points. This includes: using the measured gas content of coal seams to validate the gas content logging prediction model, substituting the sonic transit time, deep lateral resistivity, and natural gamma logging values ​​into the gas content logging prediction model to obtain the corresponding predicted gas content value, and dividing the absolute value of the difference between the predicted gas content value and the measured gas content value by the measured gas content value to obtain the prediction error. If the prediction error is large, the model is reconstructed; if the prediction error is within the allowable range, the gas content logging prediction model passes validation. Inversion of the target exploration section using seismic and borehole data includes: creating a synthetic record using borehole logging data, establishing a low-frequency model, and directly inverting the target exploration section using inversion software to obtain data volumes of sonic transit time, deep lateral resistivity, and natural gamma logging values. The attribute values ​​of the target layer plane are obtained by extracting the above data volumes, including: interpreting the target layer and difference value in the seismic data volume, and extracting the amplitude attribute value of the target layer from the three data volumes of sonic transit time, deep lateral resistivity and natural gamma logging value respectively. These are the plane attribute values ​​of sonic transit time, deep lateral resistivity and natural gamma logging value corresponding to the target layer. Substitute the attribute values ​​of the target layer plane into the verified gas content logging prediction model to obtain the gas content plane variation characteristics of the target exploration layer; wherein, the gas content plane variation characteristics of the target exploration layer include the distribution range of high gas content areas and the distribution range of areas with gas content below the preset lower limit.

2. The method according to claim 1, characterized in that: The gas content range of the high gas content area is 1.8-2.5 m3 / t, and the gas content range of the area with a gas content below the preset lower limit is 0.75-1.25 m3 / t.

3. A system for predicting gas content in coal-bearing formations through well logging and inversion, characterized in that, include: The model building unit is used to predict the gas content of coal seams by using sonic transit time, deep lateral resistivity and natural gamma logging values, and to establish a gas content logging prediction model through data regression analysis. It includes: using borehole sonic transit time, deep lateral resistivity and natural gamma logging values ​​as variables AC, RD and GR, and using the preset test gas content as the target value, and obtaining the coefficients a, b, c of the three variables and the constant d through data regression analysis, and establishing a gas content logging prediction model, expressed as formula (1): （1） The model validation unit validates the gas content logging prediction model using the measured gas content of the coal seam at the sampling points. This includes: validating the gas content logging prediction model using the measured coal seam gas content; substituting the sonic transit time, deep lateral resistivity, and natural gamma logging values ​​into the gas content logging prediction model to obtain the corresponding predicted gas content value; dividing the absolute value of the difference between the predicted and measured gas content values ​​by the measured gas content value to obtain the prediction error; if the prediction error is large, the model is reconstructed; if the prediction error is within the allowable range, the gas content logging prediction model passes validation. The inversion unit uses seismic and borehole data to invert the target exploration section, including: using borehole logging data to create a synthetic record, then establishing a low-frequency model, and directly inverting the target exploration section through inversion software to obtain data volumes of sonic transit time, deep lateral resistivity and natural gamma logging values. The extraction unit is used to extract the attribute values ​​of the target layer plane by extracting the above data volume, including: interpreting the target layer and difference value in the seismic data volume, and extracting the amplitude attribute value of the target layer from the three data volumes of sonic transit time, deep lateral resistivity and natural gamma logging value respectively, that is, the plane attribute values ​​of sonic transit time, deep lateral resistivity and natural gamma logging value corresponding to the target layer respectively. The result output unit is used to substitute the attribute values ​​of the target layer plane into the verified gas content logging prediction model to obtain the gas content plane variation characteristics of the target exploration layer; wherein, the gas content plane variation characteristics of the target exploration layer include the distribution range of high gas content areas and the distribution range of areas with gas content below the preset lower limit.

4. The system according to claim 3, characterized in that: The gas content range of the high gas content area is 1.8-2.5 m3 / t, and the gas content range of the area with a gas content below the preset lower limit is 0.75-1.25 m3 / t.

Images