Prediction method for isothermal absolute adsorption curve of methane in shale

Abstract

The invention relates to a method for predicting an isothermal absolute adsorption curve of methane in shale. The method comprises the following steps: collecting methane isothermal adsorption parameters of a research area and data of influence factors of the methane isothermal adsorption parameters; determining the promotion or weakening effect of different factors on the methane isothermal adsorption constant; determining the correlation degree of each influence factor and the methane isothermal adsorption parameter by using a Pearson correlation analysis method; determining the contribution degree of each influence factor and the methane isothermal adsorption parameter by using a random forest feature importance evaluation method; calculating a comprehensive correlation degree to obtain main control factors influencing methane isothermal adsorption parameters; taking the control factor as an input parameter, taking a methane isothermal adsorption parameter as an output, and carrying out machine learning by adopting an XGboost machine learning algorithm and a random forest machine learning algorithm; and calculating by adopting a formula to obtain the absolute adsorption quantity of methane. The methane isothermal adsorption capacity of shale can be rapidly and accurately evaluated, and the experiment time and the experiment cost can be reduced.

Description

Technical Field

[0001] The invention relates to the technical field of shale methane content prediction, and in particular to a method for predicting an isothermal absolute adsorption curve of methane in shale. Background Art

[0002] The methane content in shale is a key indicator for evaluating shale reservoirs. Currently, most methods for determining the adsorption capacity of shale are to quantitatively determine it using a methane isothermal adsorption model after a methane isothermal adsorption experiment. However, the experimental method is single, time-consuming and costly, which is not conducive to the rapid evaluation of the methane adsorption capacity of shale. Summary of the invention

[0003] In order to solve the above technical problems, the present invention provides a method for predicting the isothermal absolute adsorption curve of methane in shale.

[0004] The present invention is achieved through the following technical solutions:

[0005] A method for predicting an isothermal absolute adsorption curve of methane in shale comprises the following steps:

[0006] The Langmuir volume (V L ) and Langmuir pressure (P L ) and the data on the factors affecting the methane isotherm adsorption parameters to construct a data set; the influencing factors include shale mineral composition, organic carbon content, organic matter maturity, specific surface area, pore volume, water content and temperature;

[0007] Analyze the correlation degree of each influencing factor on the methane isotherm adsorption parameters, and determine the promoting or weakening effect of different factors on the methane isotherm adsorption constant;

[0008] The Pearson correlation analysis method was used to calculate the correlation coefficient of each influencing factor on the methane isotherm adsorption parameters, and to determine the degree of correlation between each influencing factor and the methane isotherm adsorption parameters.

[0009] The importance of each influencing factor was analyzed using the random forest feature importance evaluation method to determine the contribution of each influencing factor to the methane isotherm adsorption parameters;

[0010] The calculation results of Pearson correlation analysis and random forest feature importance evaluation method were combined to calculate the comprehensive correlation between each influencing factor and methane isotherm adsorption parameters, and the main controlling factors affecting methane isotherm adsorption parameters were obtained;

[0011] The control factors are used as input parameters and the methane isotherm adsorption parameters are used as outputs. The XGboost machine learning algorithm and the random forest machine learning algorithm are used for machine learning to obtain two prediction models. The main control factors of the test set are brought into the prediction model for calculation, and the model with the smallest root mean square error is selected to obtain the methane isotherm adsorption parameters, which include the Langmuir volume and the Langmuir pressure.

[0012] Substituting the methane isotherm adsorption parameters into the following formula, the absolute adsorption amount V of methane is calculated:

[0013] V=V ads =PV L / (P L +P)

[0014] In the above formula, V L is the Langmuir volume, cm3 / g; p L is the Langmuir pressure, MPa; p is the pressure, MPa.

[0015] Optionally, the mineral composition includes quartz content, clay mineral content, feldspar content, carbonate content, and pyrite content.

[0016] Preferably, the main controlling factors include total organic carbon content, temperature, quartz content, carbonate content, pyrite content, feldspar content, and clay mineral content.

[0017] Among them, the following formula is used to calculate the comprehensive correlation between each influencing factor and the methane isotherm adsorption parameters:

[0018]

[0019] In the above formula, γ i is the correlation degree obtained by random forest feature importance analysis, F i is the weight obtained by Pearson, η i is the comprehensive correlation of each influencing factor.

[0020] Compared with the prior art, this application has at least the following beneficial effects:

[0021] This application establishes a machine learning model of the XGboost algorithm and the random forest algorithm based on a large amount of methane isotherm adsorption data, which can quickly evaluate the methane isotherm adsorption capacity of shale; and subject to the constraints of the Langmuir model, it has a relatively accurate prediction result for the absolute isotherm adsorption capacity of methane, which can reduce experimental time and experimental costs when judging the methane isotherm adsorption capacity of shale. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 The thermal diagram of the correlation coefficients of the Langmuir parameters in the embodiment;

[0024] Figure 2 The various factors in the embodiment have an influence on V L Contribution diagram of

[0025] Figure 3 In the embodiment, various factors have an effect on P L Contribution diagram of

[0026] Figure 4 In the embodiment V L Comprehensive contribution graph;

[0027] Figure 5 In the embodiment L Comprehensive contribution graph;

[0028] Figure 6 Comparison of predicted value and true value in the embodiment Figure 1 ;

[0029] Figure 7 Comparison of predicted value and true value in the embodiment Figure 2 . DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0031] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments of the present invention can be combined with each other. It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same and similar parts between the various embodiments can be referred to each other.

[0032] This embodiment discloses a method for predicting an isothermal absolute adsorption curve of methane in shale, comprising the following steps:

[0033] S1. Collect geological data of the study area, including mineral composition (quartz content, clay mineral content, carbonate content, feldspar content, pyrite content), organic carbon content, organic matter maturity, specific surface area, pore volume, water content, temperature, and methane isotherm adsorption parameters (V L , P L ) and other relevant information, and construct relevant data sets;

[0034] In the Langmuir model, the methane isotherm adsorption parameters include the Langmuir volume and the Langmuir pressure. The Langmuir volume is referred to as V in the following. L , Langmuir pressure is referred to as P L .

[0035] S2, using mathematical statistics method to analyze methane isotherm adsorption parameters (V L , P L ) and the correlation between shale mineral composition, organic carbon content, organic matter maturity, specific surface area, pore volume, water content, temperature and other parameters; it can preliminarily determine whether each factor is related to the methane isotherm adsorption parameter (V L , P L ) exists.

[0036] S3, due to the different dimensions and large numerical differences of different factors, the data of each influencing factor are normalized to eliminate the influence of the dimensions between different factors. The normalization formula is shown in formula (1):

[0037]

[0038] In the above formula, x(i)', dimensionless, is the normalized sequence; x(i) is the original sequence; is the minimum value in the sequence; is the maximum value in the sequence.

[0039] S4, the Pearson correlation analysis criterion was introduced to calculate the quartz content, clay mineral content, feldspar content, carbonate content, pyrite content, total organic carbon content, organic matter maturity, specific surface area, pore volume, water content, temperature and other parameters for methane isotherm adsorption parameter (V L , P L )’s correlation coefficient.

[0040] The calculation results of the correlation coefficient are as follows: Figure 1 As shown, each influencing factor is related to V L The correlation degree from large to small is: total organic carbon content > carbonate content > quartz content > temperature > pyrite content > pore volume > feldspar content > specific surface area > water content > clay mineral content > organic matter maturity; each factor and P LThe degree of correlation from large to small is: temperature > carbonate content > quartz content > feldspar content > organic matter maturity > clay mineral content > total organic carbon content > water content > pyrite content > specific surface area > pore volume.

[0041] At the same time, the random forest feature importance evaluation method is used to evaluate the feature importance and analyze the importance of each influencing factor. The results are as follows Figure 2 , Figure 3 As shown in the figure, each factor has an impact on V L The contribution from large to small is: total organic carbon content > quartz content > temperature > clay mineral content > feldspar content > carbonate content > pyrite content > pore volume > specific surface area > water content > organic matter maturity;

[0042] The influencing factors and P L The contribution from large to small is as follows: quartz content > temperature > total organic carbon content > feldspar content > carbonate content > pyrite content > clay mineral content > specific surface area > pore volume > organic matter maturity > water content.

[0043] S5, based on the calculation results of Pearson correlation analysis and random forest feature importance evaluation method, the relationship between each influencing factor and methane isotherm adsorption parameter (V L , P L )’s comprehensive correlation:

[0044]

[0045] In the above formula, γ i is the correlation degree obtained by random forest feature importance analysis, F i is the weight obtained by Pearson, η i is the comprehensive correlation of each influencing factor.

[0046] The calculation results are as follows Figure 4 , 5 As shown in the figure, factors with a comprehensive correlation greater than 0.05 were selected as the main controlling factors, and the 11 influencing dimensions affecting the Langmuir parameters were reduced to 7 influencing dimensions, which reduced the amount of data calculation and increased the calculation speed. The final main controlling factors were determined to be: total organic carbon content, temperature, quartz content, carbonate rock content, pyrite content, feldspar content, and clay mineral content.

[0047] S6, total organic carbon content, temperature, quartz content, carbonate content, pyrite content, feldspar content, clay mineral content and other key parameters are used as attribute data for systematic machine learning processing. The XGboost machine learning algorithm and the random forest machine learning algorithm are used to calculate the Langmuir parameter (V L , P L) and each main control factor are machine-learned and the prediction model is output. The model with the smallest root mean square error is selected, and the main control factor data of the test set is brought into the model to calculate the Langmuir parameter (V L , P L ), and then the Langmuir model was used to calculate the absolute adsorption amount of methane.

[0048] The absolute adsorption amount V of methane is obtained using the Langmuir equation (Equation 3):

[0049] V=V ads =PV L / (P L +P) (3)

[0050] In the above formula, V L is the Langmuir volume, cm 3 / g;p L is the Langmuir pressure, MPa; P is the pressure, MPa.

[0051] Finally, it is compared with the actual adsorption amount, and the comparison results are as follows Figure 6 , Figure 7 As shown, the results show that the predicted data is close to the real data, the prediction progress is high, and the prediction results are good.

[0052] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for predicting the absolute isothermal adsorption curve of methane in shale, characterized in that: The following steps are involved: Collect data on methane isotherm adsorption parameters and influencing factors of methane isotherm adsorption parameters in the study area to construct a data set; the influencing factors include shale mineral composition, organic carbon content, organic matter maturity, specific surface area, pore volume, water content and temperature; Analyze the correlation degree of each influencing factor on the methane isotherm adsorption parameters, and determine the promoting or weakening effect of different factors on the methane isotherm adsorption constant; The Pearson correlation analysis method was used to calculate the correlation coefficient of each influencing factor on the methane isotherm adsorption parameters, and to determine the degree of correlation between each influencing factor and the methane isotherm adsorption parameters. The importance of each influencing factor was analyzed using the random forest feature importance evaluation method to determine the contribution of each influencing factor to the methane isotherm adsorption parameters; The calculation results of Pearson correlation analysis and random forest feature importance evaluation method were combined to calculate the comprehensive correlation between each influencing factor and methane isotherm adsorption parameters, and the main controlling factors affecting methane isotherm adsorption parameters were obtained; The control factors are used as input parameters and the methane isotherm adsorption parameters are used as outputs. The XGboost machine learning algorithm and the random forest machine learning algorithm are used for machine learning to obtain two prediction models. The main control factors of the test set are brought into the prediction model for calculation, and the model with the smallest root mean square error is selected to obtain the methane isotherm adsorption parameters, which include the Langmuir volume and the Langmuir pressure. Substituting the methane isotherm adsorption parameters into the following formula, the absolute adsorption amount V of methane is calculated: V=V ads =pV L / (p L +p) In the above formula, V L is the Langmuir volume, cm 3 / g;p L is the Langmuir pressure, MPa; p is the pressure, MPa.

2. The method for predicting the isothermal absolute adsorption curve of methane in shale according to claim 1, characterized in that: The mineral composition includes quartz content, clay mineral content, feldspar content, carbonate content, and pyrite content.

3. The method for predicting the isothermal absolute adsorption curve of methane in shale according to claim 2, characterized in that: The main controlling factors include total organic carbon content, temperature, quartz content, carbonate content, pyrite content, feldspar content, and clay mineral content.

4. The method for predicting the isothermal absolute adsorption curve of methane in shale according to claim 1, characterized in that: The comprehensive correlation between each influencing factor and methane isotherm adsorption parameters is calculated using the following formula: In the above formula, γ i is the correlation degree obtained by random forest feature importance analysis, F i is the weight obtained by Pearson, η i is the comprehensive correlation of each influencing factor.

5. The method for predicting the isothermal absolute adsorption curve of methane in shale according to claim 1, characterized in that: The isothermal adsorption parameters of methane include Langmuir volume and Langmuir pressure.

Images