Volcanic rock reservoir porosity quantitative calculation method based on recorded and measured data

Abstract

The invention provides a volcanic rock reservoir porosity quantitative calculation method based on recorded and measured data, and belongs to the field of oil and gas field exploration. Comprising the following steps: establishing a porosity calculation model based on analysis and test data, element logging data or logging while drilling data, processing actual data by utilizing the established porosity calculation model, comparing a calculation result with core analysis and test data, and carrying out error analysis; when only element logging data exist, an element logging multiple regression method is selected to be adopted, and a porosity calculation model of multi-parameter element logging is established; and when only element logging and logging while drilling information exist, a recording and measuring coupling method is selected, and a recording and measuring multi-parameter coupling porosity calculation model is established. According to the volcanic rock reservoir porosity quantitative calculation method, common data of most wells are comprehensively considered, the volcanic rock reservoir porosity quantitative calculation method is established on the basis of sensitivity analysis according to the situation that logging-while-drilling information exists and the situation that logging-while-drilling information does not exist, the manual intervention rate and the professional degree need are reduced, and the troubles of high operability and the like are reduced.

Description

Technical Field

[0001] This invention relates to the field of oil and gas field exploration, and in particular to a method for quantitatively calculating the porosity of volcanic rock reservoirs based on recorded data. Background Technology

[0002] Mesozoic buried hill volcanic reservoirs are characterized by complex rock compositions and diverse lithologies, making rapid in-situ assessment of physical properties challenging, particularly the rock skeleton density. Influenced by the lithological complexity, the volcanic rock skeleton density varies significantly with rock composition compared to the constant skeleton density of conventional sandstone and mudstone. Existing assessment methods rely on well logging data, and conventional single-logging petrophysical volumetric models cannot effectively achieve rapid quantitative assessment of porosity. While methods for assessing formation porosity based on engineering parameters (work index, mechanical specific energy) have been reported, these methods are subject to numerous limitations, such as variations in formation lithology and the influence of reservoir space type on the response characteristics of engineering parameters. These factors can alter the baseline values ​​of engineering parameters, leading to inaccurate porosity calculations. Therefore, there is an urgent need for a method that enables rapid quantitative in-situ assessment of porosity in both scenarios, with and without well logging data. Summary of the Invention

[0003] In view of this, the present invention aims to propose a quantitative calculation method for porosity of volcanic rock reservoirs based on logging data, which can comprehensively utilize logging data and well logging data to achieve rapid quantitative evaluation of porosity in the field.

[0004] To achieve the above objectives, the technical solution of the present invention is implemented as follows: a quantitative calculation method for porosity of volcanic rock reservoirs based on logging data, comprising: establishing a porosity calculation model based on analytical data, elemental logging data, or logging-while-drilling data; processing actual data using the established porosity calculation model; comparing the calculation results with core analytical data; and conducting error analysis.

[0005] In the absence of logging-while-drilling data, proceed as follows:

[0006] Based on the analysis of laboratory data and elemental logging data, a correlation analysis was conducted between core analysis porosity and elemental logging data to determine the sensitive element logging parameters related to porosity. Then, combined with the significance analysis of P-values, multicollinearity among the sensitive element logging parameters was eliminated.

[0007] Based on the optimized sensitivity parameters, a multi-parameter element logging porosity calculation model was established using multiple linear regression. The established multi-parameter element logging porosity calculation model was used to process actual data. The calculation results were compared with core analysis and testing data, and error analysis was carried out.

[0008] If logging-while-drilling data is available, proceed as follows:

[0009] Rock skeleton density was calculated using multiple regression.

[0010] The logging density is calibrated based on the rock density from core analysis;

[0011] Based on the volumetric model of rock physics theory, the density difference between the rock skeleton density and the rock density is calculated.

[0012] By using linear regression, the relationship between porosity and density difference was established, and a multi-parameter coupled porosity calculation model was determined. The determined multi-parameter coupled porosity calculation model was used to process actual well data, and the calculation results were compared with core analysis and test data, and error analysis was carried out.

[0013] Furthermore, in the absence of logging-while-drilling data, Pearson correlation analysis was used to conduct a correlation analysis between porosity and logging data to obtain the correlation coefficient R between porosity and logging data. Based on the correlation coefficient R, sensitivity parameters were determined. Let X be the independent variable and Y be the dependent variable. The steps for calculating the correlation between the two include:

[0014] First, calculate the covariance between variables X and Y:

[0015] Cov(X,Y)=E[(XE(X))(YE(Y))] (1)

[0016] In the formula: Cov(X,Y) is the covariance of X and Y, and E(X) and E(Y) are the expected values ​​of variables X and Y, respectively. The expected values ​​of variables X, E(X) and E(Y), are calculated as follows:

[0017] The distribution law of discrete random variable X is as follows:

[0018] P{X=x k}=p k k = 1, 2, ..., n (2)

[0019] If the following series converges absolutely:

[0020]

[0021] Then the expected value E(X) of variable X is:

[0022]

[0023] The calculation method for E(y) is the same as that for E(X);

[0024] Next, calculate the standard deviations of variables X and Y:

[0025] Calculate the variance Var(X) of variable X:

[0026] Var(X)=E[(XE(X)) 2 (5)

[0027] Calculate the variance Var(Y) of variable Y:

[0028]

[0029] The standard deviations of variables X and Y are respectively:

[0030]

[0031] Finally, calculate the Pearson coefficient R:

[0032]

[0033] The correlation coefficient R ranges from -1 to R and from 1 to 0. When the Pearson coefficient R > 0, it indicates that variables X and Y are positively correlated; when R < 0, it indicates that variables X and Y are negatively correlated; when R = 0, it indicates that variables X and Y are not correlated. The closer |R| is to 1, the better the correlation between variables X and Y; the closer |R| is to 0, the worse the correlation between variables X and Y.

[0034] Furthermore, in the absence of logging-while-drilling data, a porosity calculation model based on multi-parameter element logging is established using multiple linear regression. This refers to using the multiple linear regression method to fit the sensitivity parameters with the core porosity to obtain a porosity calculation model based on multi-parameter element logging.

[0035] A multi-parameter element logging porosity calculation model for Mesozoic buried hill volcanic reservoirs was constructed using the multiple linear regression method.

[0036] Let x be the k independent variables that affect the dependent variable y. i Establish a k-variable linear regression model:

[0037] y = a0 + a1x1 + a2x2 + ... + a k x k +e (10)

[0038]

[0039] The residual between the estimated value and the observed value is:

[0040]

[0041] Where: a0, a1, ..., a k All are regression coefficients; y is an estimated value; Here, represents the observed value; e represents the residual; to obtain the regression coefficients, the least squares method is used to minimize the sum of squared residuals:

[0042]

[0043] For the n sets of data obtained from observation (Y) k X nk G is a0, a1, ..., a k Since it is a non-negative quadratic function, it must have a local minimum, i.e., a0, a1, ..., a k Must be satisfied:

[0044]

[0045] Wherein, minG is the minimum residual sum of squares;

[0046] These are m+1 undetermined coefficients a0, a1, ..., a k The simultaneous equations (14) and (15) can be obtained from equation (14):

[0047]

[0048] in:

[0049]

[0050] Substituting a0 into equations (15) and (16) respectively, we get:

[0051]

[0052] in:

[0053]

[0054] In the formula, m represents the number of observation samples;

[0055] The undetermined coefficients a0, a1, ..., a can be solved using equation (19). k Calculate a0 from equation (18), substitute it back into equation (19), and you will get the multiple linear regression equation (10).

[0056] Furthermore, with logging-while-drilling data available, sensitivity parameters are optimized using a correlation heatmap matrix, and p-value significance analysis is performed. Then, a rock skeleton density (DM) calculation model is established using multiple linear regression.

[0057] DM = w1x1 + w2x2 + ... + w n x n +w (22)

[0058] In the formula, w1, w2, w n w represents the fitting coefficients, x1, x2, x... nThe preferred sensitivity parameter is represented by DM, which represents the rock skeleton density.

[0059] Furthermore, with logging-while-drilling data available, a linear regression model was established using core analysis of rock density and logging density:

[0060] ρ c =aρ b +b (23)

[0061] In the formula, ρ c This represents the value after well logging density calibration, a and b represent fitting coefficients, and ρ b Indicates logging density;

[0062] Then, the established linear regression model was used to calibrate the logging density.

[0063] Furthermore, with logging-while-drilling data available, based on the volumetric model of rock physics theory, and utilizing core analysis data on porosity, rock skeleton density, and rock density, a relationship is established between porosity and the difference between rock skeleton density and rock density. A porosity calculation model is then established through linear regression.

[0064] POR=aΔρ+b (24)

[0065] In the formula, POR represents the porosity calculated by multi-parameter coupling of recording and measurement, a and b represent fitting coefficients, and Δρ represents the difference between the rock skeleton density and the rock density.

[0066] Compared with existing technologies, the quantitative calculation method for volcanic rock reservoir porosity based on recorded data described in this invention has the following advantages:

[0067] (1) The method described in this invention couples multiple logging parameters to quickly and accurately perform quantitative calculation of porosity of fluids in volcanic rock reservoirs;

[0068] (2) This invention takes into account the data (logging and logging) common to most wells. Based on sensitivity analysis, it establishes a quantitative calculation method for porosity of volcanic rock reservoirs for two situations: when there is logging while drilling data and when there is no logging while drilling data. This reduces the need for manual intervention and professional expertise, and addresses the challenges of high operability.

[0069] (3) Based on logging and logging-while-drilling data, this invention allows for flexible selection of calculation methods depending on the field data acquisition situation. When only elemental logging data is available, the elemental logging multivariate regression method is selected to establish a multi-parameter elemental logging porosity calculation model. When only elemental logging and logging-while-drilling data are available, the logging-logging coupling method is selected to establish a logging-logging multi-parameter coupled porosity calculation model.

[0070] (4) In view of the characteristics that engineering parameters are easily affected by factors such as lithology and reservoir space, this invention starts with core analysis and testing data. First, based on the correlation analysis between elemental logging data and porosity, a multi-parameter elemental logging porosity calculation model is constructed. When there is drilling density logging data, a logging-coupled porosity calculation model is constructed based on the rock volume model.

[0071] (5) The present invention is simple to operate, easy to promote and apply, and suitable for rapid on-site quantitative calculation of porosity. Attached Figure Description

[0072] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0073] Figure 1 For technology roadmap;

[0074] Figure 2 The results of the sensitivity parameter analysis of the heatmap matrix of porosity and elemental logging data based on Person correlation analysis;

[0075] Figure 3 A comparison chart showing the porosity calculated using a multiple linear regression model for core analysis and elemental logging.

[0076] Figure 4 Cross plot showing the correlation between porosity and density difference (rock skeleton density and rock density);

[0077] Figure 5 The results of sensitivity parameter analysis of the heatmap matrix of rock skeleton and elemental logging data based on Person correlation analysis;

[0078] Figure 6 A comparison chart of the analyzed and calculated skeleton density of the core sample;

[0079] Figure 7 This is a cross plot showing the correlation between logging density and core analysis density.

[0080] Figure 8 This is a comparison chart of density after well logging calibration and density analysis from core samples.

[0081] Figure 9 A composite graph showing the results of porosity calculations using the two methods. Detailed Implementation

[0082] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0083] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0084] like Figure 1 As shown, this invention is a method for quantitatively calculating the porosity of volcanic rock reservoirs based on recorded data, including:

[0085] Based on analytical testing data, elemental logging data, or logging-while-drilling data, a porosity calculation model is established. This model is then used to process actual data, and the calculation results are compared with core analytical testing data, followed by error analysis. This includes using elemental logging multivariate regression and logging-while-drilling coupling methods. When only elemental logging data is available, the elemental logging multivariate regression method is used to establish a multi-parameter elemental logging porosity calculation model. When only elemental logging and logging-while-drilling data are available, the logging-while-drilling coupling method is used to establish a multi-parameter coupled porosity calculation model. Details are as follows:

[0086] In the absence of logging-while-drilling data, proceed as follows:

[0087] (1) Based on the analysis and testing data and elemental logging data, conduct correlation analysis between core analysis porosity and elemental logging data, determine the sensitive element logging parameters related to porosity, and then combine the P-value significance analysis to eliminate multicollinearity among sensitive element logging parameters.

[0088] Wells in the study area with core analysis data for porosity (POR) and rock skeleton density (DM) were selected. Common elemental logging data (Na, Mg, Al, Si, P, S, Cl, K, Ca, Ti, Mn, Fe, etc.) were chosen. Based on the heatmap matrix, Pearson correlation coefficient analysis was used to determine sensitive parameters, and whether each sensitive parameter was positively or negatively correlated with porosity was determined.

[0089] Then, using p-value significance analysis, we can perform significance analysis on the sensitivity parameters to see if there is a significant correlation between the sensitivity elements. If there is no significant correlation between the sensitivity parameters, we can consider that multicollinearity has been excluded, meaning that all the determined sensitivity parameters can be used for subsequent model building. If there is a significant correlation between some sensitivity parameters, we can only retain the sensitivity parameters that are not significantly correlated with each other for subsequent model building.

[0090] (2) Based on the selected sensitivity parameters, a multi-parameter element logging porosity calculation model is established using multiple linear regression. The established multi-parameter element logging porosity calculation model is used to process the actual data. The calculation results are compared with the core analysis and testing data, and error analysis is carried out.

[0091] In step (1) where there is no logging-while-drilling data, the specific steps include: using Pearson correlation analysis to select the sensitivity parameters based on elemental logging data and core analysis of porosity, and performing significance analysis on the sensitivity parameters to eliminate multivariate collinearity and avoid overfitting.

[0092] Pearson correlation analysis was used to analyze the correlation between porosity and logging data to obtain the correlation coefficient R between them. Based on the correlation coefficient R, sensitivity parameters were determined. The independent variable X and dependent variable Y were set, and the steps for calculating their correlation included:

[0093] First, calculate the covariance between variables X and Y:

[0094] Cov(X,Y)=E[(XE(X))(YE(Y))] (1)

[0095] In the formula: Cov(X,Y) is the covariance of X and Y, and E(X) and E(Y) are the expected values ​​of variables X and Y, respectively. The expected values ​​of variables X, E(X) and E(Y), are calculated as follows:

[0096] The distribution law of discrete random variable X is as follows:

[0097] P{X=x k}=p k k = 1, 2, ..., n (2)

[0098] If the following series converges absolutely:

[0099]

[0100] Then the expected value E(X) of variable X is:

[0101]

[0102] The calculation method for E(y) is the same as that for E(X);

[0103] Next, calculate the standard deviations of variables X and Y:

[0104] Calculate the variance Var(X) of variable X:

[0105]

[0106] Calculate the variance Var(Y) of variable Y:

[0107]

[0108] The standard deviations of variables X and Y are respectively:

[0109]

[0110] Finally, calculate the Pearson coefficient R:

[0111]

[0112] The correlation coefficient R ranges from -1 to R and from 1 to 0. When the Pearson coefficient R > 0, it indicates that variables X and Y are positively correlated; when R < 0, it indicates that variables X and Y are negatively correlated; when R = 0, it indicates that variables X and Y are not correlated. The closer |R| is to 1, the better the correlation between variables X and Y; the closer |R| is to 0, the worse the correlation between variables X and Y.

[0113] In step (2) where there is no logging-while-drilling data, a porosity calculation model for multi-parameter element logging is established using multiple linear regression. This means that the sensitivity parameters are fitted to the core porosity using the multiple linear regression method to obtain a porosity calculation model based on multi-parameter element logging.

[0114] Specifically, this includes: using the multiple linear regression method to construct a porosity calculation model for multi-parameter element logging of Mesozoic buried hill volcanic reservoirs;

[0115] Let x be the k independent variables that affect the dependent variable y. i Establish a k-variable linear regression model:

[0116] y = a0 + a1x1 + a2x2 + ... + a k x k +e (10)

[0117]

[0118] The residual between the estimated value and the observed value is:

[0119]

[0120] Where: a0, a1, ..., a k All are regression coefficients; y is an estimated value; Here, represents the observed value; e represents the residual; to obtain the regression coefficients, the least squares method is used to minimize the sum of squared residuals:

[0121]

[0122] For the n sets of data obtained from observation (Y) k X nk G is a0, a1, ..., a k Since it is a non-negative quadratic function, it must have a local minimum, i.e., a0, a1, ..., a k Must be satisfied:

[0123]

[0124] Wherein, minG is the minimum residual sum of squares;

[0125] These are m+1 undetermined coefficients a0, a1, ..., a k The simultaneous equations (14) and (15) can be obtained from equation (14):

[0126]

[0127] in:

[0128]

[0129] Substituting a0 into equations (15) and (16) respectively, we get:

[0130]

[0131] in:

[0132]

[0133] In the formula, m represents the number of observation samples;

[0134] The undetermined coefficients a0, a1, ..., a can be solved using equation (19). k Calculate a0 from equation (18), substitute it back into equation (19), and you will get the multiple linear regression equation (10).

[0135] If logging-while-drilling data is available, proceed as follows:

[0136] (1) Calculate the density of the rock skeleton using the multiple regression method;

[0137] (2) The logging density is calibrated based on the rock density from the core analysis;

[0138] (3) Based on the volume model of rock physics theory, calculate the density difference between the rock skeleton density and the rock density;

[0139] (4) By linear regression, the relationship between porosity and density difference is established, the multi-parameter coupled porosity calculation model is determined, the actual well data is processed using the determined multi-parameter coupled porosity calculation model, the calculation results are compared with the core analysis and test data, and error analysis is carried out.

[0140] In step (1) where logging-while-drilling data is available, sensitivity parameters are selected using a correlation heatmap matrix, and a p-value significance analysis is performed. Then, a rock skeleton density DM calculation model is established using multiple linear regression:

[0141] DM = w1x1 + w2x2 + ... + w n x n +w (22)

[0142] In the formula, w1, w2, w n w represents the fitting coefficients, x1, x2, x... n The preferred sensitivity parameter is represented by DM, which represents the rock skeleton density.

[0143] In step (2), where logging-while-drilling data is available, a linear regression model is established using core analysis of rock density and logging density:

[0144] ρ c =aρ b +b (23)

[0145] In the formula, ρ c This represents the value after well logging density calibration, a and b represent fitting coefficients, and ρ b This indicates the logging density.

[0146] Then, the established linear regression model was used to calibrate the logging density.

[0147] In step (4) where logging-while-drilling data is available, based on the volumetric model of rock physics theory, the relationship between porosity and the difference between rock skeleton density and rock density is established using the porosity, rock skeleton density, and rock density data from core analysis. A porosity calculation model is then established through linear regression.

[0148] POR=aΔρ+b (24)

[0149] In the formula, POR represents the porosity calculated by multi-parameter coupling of recording and measurement, a and b represent fitting coefficients, and Δρ represents the difference between the rock skeleton density and the rock density.

[0150] Example 1

[0151] In the absence of logging-while-drilling data:

[0152] Wells with core analysis data for porosity (POR) and rock skeleton density (DM) were selected for the study area. Common elemental logging data (Na, Mg, Al, Si, P, S, Cl, K, Ca, Ti, Mn, Fe, etc.) were also selected. Based on the analysis data and elemental logging data, a correlation analysis between core analysis porosity and elemental logging data was conducted using heatmap matrix and Pearson correlation coefficient analysis. The analysis results are as follows: Figure 2 As shown, the sensitive parameters were determined to be Mg, Si, Cl and Fe, among which Mg, Si and Cl were positively correlated with porosity, and Fe was negatively correlated with porosity.

[0153] The logging parameters of sensitive elements related to porosity were determined, and then multivariate collinearity among these parameters was eliminated using p-value significance analysis. The statistical results of the p-value significance analysis are shown in Table 1. The analysis results indicate that there is no significant correlation among the four sensitive elements Mg, Si, Cl, and Fe, thus ruling out multivariate collinearity; therefore, all four sensitive elements can be used for subsequent model establishment.

[0154] Table 1. Statistical table of significance analysis results for four sensitivity parameters related to porosity.

[0155]

[0156] Based on the sensitivity parameters Mg, Si, Cl, and Fe determined by the optimized sensitivity parameters, a multiple linear regression model was established with the porosity (POR) from core analysis. This multiple linear regression model was then used to establish a multi-parameter elemental logging porosity calculation model for Mesozoic buried hill volcanic reservoirs.

[0157] POR=16.13Mg+0.15Si+6.52Cl-0.94Fe-0.03

[0158] Using a model to process actual data, a comparison chart of core analysis porosity and porosity calculated by a multiple linear regression model based on elemental logging is shown below. Figure 3 As shown in Table 2, the statistical analysis of the porosity calculation error by elemental logging multiple linear regression shows that it has good accuracy.

[0159] Table 2 Statistical Analysis of Porosity Error Calculation Based on Elemental Logging Multiple Linear Regression

[0160]

[0161]

[0162] With logging-while-drilling data available:

[0163] First, the rock skeleton density was calculated using a multiple regression method: similar to step (1) in the absence of logging-while-drilling data, the sensitivity parameters were optimized using a correlation heatmap matrix. Figure 5 The significance of the p-value was analyzed, and the results are shown in Table 3.

[0164] Table 3. Statistical analysis results of the four sensitivity parameters related to skeleton density.

[0165]

[0166] Then, a rock skeleton density (DM) calculation model is established using multiple linear regression:

[0167] DM=-0.0048Na-0.0012Al+0.0262ln(S)-0.0363Fe+2.8509

[0168] A comparison chart of the analyzed and calculated skeleton density of core samples, based on actual data, is shown below. Figure 6 As shown in Table 4, the error analysis results indicate that the model has good accuracy.

[0169] Table 4. Statistical Analysis of Errors in Core Analysis and Calculated Rock Skeleton

[0170] Maximum value Minimum value average value Median Absolute error, % 0.12 0.0003 0.03 0.02 Relative error, % 4.24 0.01 1.07 0.91

[0171] Then, the logging density is calibrated based on the core analysis rock density. Typically, there is a difference between logging density and core analysis density, and porosity calculation requires calibration of the logging density. A linear regression model is established using the core analysis rock density and the logging density, and then this model is used to calibrate the logging density. The cross-plot of the correlation between logging density and core analysis density is shown below. Figure 7 As shown in the figure. The comparison between the density after well logging calibration and the density analysis of the core sample is as follows. Figure 8 As shown in Table 5, the statistical analysis of errors after well logging density calibration is presented. The calibrated data are closer to the core analysis results. The linear regression model is shown in the following equation:

[0172] ρ c =0.8077ρ b +0.405

[0173] Table 5. Statistical Analysis of Errors After Well Logging Density Calibration

[0174] Maximum value Minimum value average value Median Absolute error, % 0.15 0.004 0.05 0.04 Relative error, % 6.59 0.16 2.13 1.69

[0175] Then, based on the theoretical volume model of rocks, the density difference between the rock skeleton density and the rock density is calculated;

[0176] Finally, based on the analysis and testing data, a relationship was established between porosity and the difference between rock skeleton density and rock density: based on the volumetric model of rock physics theory, the difference between rock porosity and density (rock skeleton density and rock density) shows a positive linear correlation. Figure 4 A porosity calculation model was established using linear regression:

[0177] POR = 0.3654Δρ + 0.0019

[0178] Based on well logging density calibration and processing of actual well data, porosity was calculated using a theoretical volume model. The statistical analysis of the error is shown in Table 6. The error analysis results indicate that the model has good accuracy.

[0179] Table 6. Statistical Analysis of Porosity Calculation Error Based on Theoretical Volume Model

[0180] Maximum value Minimum value average value Median Absolute error, % 1.93 0.02 0.49 0.32 Relative error, % 14.53 0.47 5.56 4.48

[0181] The composite porosity result is calculated using two methods: one with logging-while-drilling data and the other with logging-while-drilling data. Figure 9 As shown, Mg, Fe, Al, K, Ca, and P represent the elements of magnesium, iron, aluminum, potassium, calcium, and phosphorus in the logging, respectively; DEN, ZDEN, and DENC represent the core analysis density (black), logging density (red), and calibrated density (blue), respectively; CPOR, PORC, and POR_XRF represent the core analysis porosity (black), logging-coupled calculated porosity (blue), and logging multi-parameter coupled calculated porosity (red), respectively.

[0182] from Figure 9 As can be seen, the porosity calculated by both methods is consistent with the trend of analytical data, indicating their applicability. The elemental logging multivariate regression method is suitable for quantitative evaluation of physical properties when there is no logging-while-drilling data, while the logging-while-drilling coupling method is suitable for use when there is logging-while-drilling data.

[0183] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for quantitatively calculating the porosity of volcanic rock reservoirs based on recorded data, characterized in that, include: Based on analytical test data, elemental logging data, or logging-while-drilling data, a porosity calculation model is established. The established porosity calculation model is used to process the actual data. The calculation results are compared with the core analytical test data, and error analysis is carried out. In the absence of logging-while-drilling data, proceed as follows: Based on the analysis of laboratory data and elemental logging data, a correlation analysis was conducted between core analysis porosity and elemental logging data to determine the sensitive element logging parameters related to porosity. Then, combined with the significance analysis of P-values, multicollinearity among the sensitive element logging parameters was eliminated. Based on the optimized sensitivity parameters, a multi-parameter element logging porosity calculation model was established using multiple linear regression. The established multi-parameter element logging porosity calculation model was used to process actual data. The calculation results were compared with core analysis and testing data, and error analysis was carried out. If logging-while-drilling data is available, proceed as follows: Rock skeleton density was calculated using multiple regression. The logging density is calibrated based on the rock density from core analysis; Based on the volumetric model of rock physics theory, the density difference between the rock skeleton density and the rock density is calculated. By using linear regression, the relationship between porosity and density difference was established, and a multi-parameter coupled porosity calculation model was determined. The determined multi-parameter coupled porosity calculation model was used to process actual well data, and the calculation results were compared with core analysis and test data, and error analysis was carried out.

2. The method for quantitatively calculating the porosity of volcanic rock reservoirs based on recorded data according to claim 1, characterized in that: In the absence of logging-while-drilling data, Pearson correlation analysis is used to analyze the correlation between porosity and logging data, obtaining the correlation coefficient R between the two. Based on the correlation coefficient R, sensitivity parameters are determined. Let X be the independent variable and Y be the dependent variable. The steps for calculating the correlation between the two include: First, calculate the covariance between variables X and Y: Cov(X,Y)=E[(XE(X))(YE(Y))] (1) In the formula: Cov(X,Y) is the covariance of X and Y, and E(X) and E(Y) are the expected values ​​of variables X and Y, respectively. The expected values ​​of variables X, E(X) and E(Y), are calculated as follows: The distribution law of discrete random variable X is as follows: P{X=x k }=p k k=1,2,…n (2) If the following series converges absolutely: Then the expected value E(X) of variable X is: The calculation method for E(y) is the same as that for E(X); Next, calculate the standard deviations of variables X and Y: Calculate the variance Var(X) of variable X: Calculate the variance Var(Y) of variable Y: The standard deviations of variables X and Y are respectively: Finally, calculate the Pearson coefficient R: The correlation coefficient R ranges from -1 to R and from 1 to 0. When the Pearson coefficient R > 0, it indicates that variables X and Y are positively correlated; when R < 0, it indicates that variables X and Y are negatively correlated; when R = 0, it indicates that variables X and Y are not correlated. The closer |R| is to 1, the better the correlation between variables X and Y; the closer |R| is to 0, the worse the correlation between variables X and Y.

3. The method for quantitatively calculating the porosity of volcanic rock reservoirs based on recorded data according to claim 1, characterized in that: In the absence of logging-while-drilling data, establishing a porosity calculation model for multi-parameter element logging using multiple linear regression refers to fitting the sensitivity parameters with the core porosity using the multiple linear regression method to obtain a porosity calculation model based on multi-parameter element logging. A multi-parameter element logging porosity calculation model for Mesozoic buried hill volcanic reservoirs was constructed using the multiple linear regression method. Let x be the k independent variables that affect the dependent variable y. i Establish a k-variable linear regression model: y=a0+a1x1+a2x2+…+a k x k +e (10) The residual between the estimated value and the observed value is: Where: a0, a1, ..., a k All are regression coefficients; y is an estimated value; Here, represents the observed value; e represents the residual; to obtain the regression coefficients, the least squares method is used to minimize the sum of squared residuals: For the n sets of data obtained from observation (Y) k X nk G is a0, a1, ..., a k Since it is a non-negative quadratic function, it must have a local minimum, i.e., a0, a1, ..., a k Must be satisfied: Wherein, minG is the minimum residual sum of squares; These are m+1 undetermined coefficients a0, a1, ..., a k The simultaneous equations (14) and (15) can be obtained from equation (14): in: Substituting a0 into equations (15) and (16) respectively, we get: in: In the formula, m represents the number of observation samples; The undetermined coefficients a0, a1, ..., a can be solved using equation (19). k Calculate a0 from equation (18), substitute it back into equation (19), and you will get the multiple linear regression equation (10).

4. The method for quantitatively calculating the porosity of volcanic rock reservoirs based on recorded data according to claim 1, characterized in that: With logging-while-drilling data available, sensitivity parameters are optimized using a correlation heatmap matrix, and p-value significance analysis is performed. Then, a rock skeleton density (DM) calculation model is established using multiple linear regression. DM=w1x1+w2x2+…+w n x n +w (22) In the formula, w1, w2, w n w represents the fitting coefficients, x1, x2, x... n The preferred sensitivity parameter is represented by DM, which represents the rock skeleton density.

5. The method for quantitatively calculating the porosity of volcanic rock reservoirs based on recorded data according to claim 1, characterized in that: In step 4, a linear regression model is established using core analysis of rock density and logging density: r c =ar b +b (23) In the formula, ρ c This represents the value after well logging density calibration, a and b represent fitting coefficients, and ρ b Indicates logging density; Then, the established linear regression model was used to calibrate the logging density.

6. The method for quantitatively calculating the porosity of volcanic rock reservoirs based on recorded data according to claim 1, characterized in that: With logging-while-drilling data available, based on a volumetric model derived from rock physics theory, and utilizing core analysis data on porosity, rock skeleton density, and rock density, a relationship is established between porosity and the difference between rock skeleton density and rock density. A porosity calculation model is then established through linear regression. POR=aΔρ+b (24) In the formula, POR represents the porosity calculated by multi-parameter coupling of recording and measurement, a and b represent fitting coefficients, and Δρ represents the difference between the rock skeleton density and the rock density.

Images