Method for determining characteristic parameters of different mineral elements of complex rock

Abstract

The invention discloses a method for determining characteristic parameters of different mineral elements of a complex rock, and the method comprises the steps: building an integrated application method of element logging data and element logging data, and verifying the reliability of integrated element data in combination with core element data; based on rock core mineral data, classifying according to lithology and counting mineral distribution characteristics of different rock cores, and determining complex multi-mineral volume models of different complex lithology; with reference to a conventional logging mineral inversion response equation, element response equations of different lithological characters are established according to the complex multi-mineral volume model of each lithological character, and characteristic values of different mineral elements of each complex lithological character are determined by using an intelligent regression algorithm; and calculating the volume content of each component of different complex lithology by combining with comprehensive logging data, and verifying the reliability of characteristic values of different mineral elements. According to the method, the overall calculation speed is increased, and the problem that the mineral component element characteristic values are difficult to determine is solved.

Description

Technical Field

[0001] The invention relates to the field of mineral inversion, and in particular to a method for determining characteristic parameters of different mineral elements in complex rocks. Background Art

[0002] Mineral inversion is an indispensable part of reservoir exploration. Different mineral contents have a great influence on the reservoir space. Accurate inversion of mineral contents of reservoirs with different lithologies is of great significance to reservoir exploration. The commonly used method for calculating the characteristic values ​​of mineral elements is to statistically analyze the rock mineral data and element data of the study area, and use the intelligent regression algorithm to find the corresponding relationship between elements and minerals, which is the characteristic value of the corresponding element of the mineral. In general, this method has great limitations. First, the types of minerals counted are limited, mostly conventional minerals, such as quartz, calcite, dolomite, potassium feldspar, plagioclase, etc., and the types of complex lithological minerals are more varied, which cannot meet the inversion requirements; secondly, the scope of application is limited. This method is only for the mineral data of the study area, and the element characteristic values ​​finally calculated cannot be directly used in other areas; finally, there are great requirements for the volume of the calculation data set. If the data set is too small, the characteristic values ​​finally obtained are not reliable, and the characteristic values ​​of the same elements of the same minerals are prone to appear, and there are orders of magnitude differences in different lithologies; however, when the data set is too large, it will generate a huge amount of calculation and reduce efficiency. Summary of the invention

[0003] The purpose of the present invention is to provide a method for determining characteristic parameters of different mineral elements in complex rocks to solve the technical problems mentioned in the background technology.

[0004] A method for determining characteristic parameters of different mineral elements in complex rocks, the method comprising the following steps: S1. Establish an integrated application method for element logging and element logging data, and verify the reliability of the integrated element data by combining core element data; S2. Based on the core mineral data, the complex multi-mineral volume model of different complex lithologies is determined by classifying according to lithology and statistically analyzing the mineral distribution characteristics of different cores; S3. With reference to the conventional well logging mineral inversion response equation, the element response equations of different lithologies are established according to the complex multi-mineral volume model of each lithology, and the characteristic values ​​of different mineral elements of each complex lithology are determined by using the intelligent regression algorithm; S4. Combine comprehensive logging data to calculate the volume content of each component of different complex lithologies and verify the reliability of characteristic values ​​of different mineral elements.

[0005] As a further technical solution of the present invention, in step S1, an integrated application method of element logging and element logging data is established, and the reliability of the integrated element data is verified by combining the core element data, including the following steps: S11, element logging data integration method; The element logging data includes common elements and some trace elements. The data provided by the core major element analysis, namely the core element data, focuses only on 10 common element oxides, showing their respective relative percentage contents, including sodium oxide ( ) and magnesium oxide ( )wait; The principle of the element logging integration method is: select 10 elements corresponding to the core oxides in the element logging data, namely sodium (Na), magnesium (Mg), aluminum (Al), silicon (Si), phosphorus (P), potassium (K), calcium (Ca), titanium (Ti), manganese (Mn) and iron (Fe) for integration; first use the logging content of each element to reversely calculate the content of each oxide, then calculate the total content of oxygen corresponding to these ten elements, and finally integrate the ten elements with oxygen. The calculation steps are as follows: First, the molar mass fraction of each non-oxygen element in different oxides is calculated based on the molar mass of the element: Formula 1: ; In the formula, is the molar mass fraction of element E in the corresponding oxide; is the total molar mass of element E in the oxide chemical formula, in units of ; is the molar mass of the oxide corresponding to element E, in units of ; Then calculate the molar mass fraction of oxygen in different oxides , the formula is the same as above; Next, the above two mass fractions are used to calculate the total content of oxygen elements corresponding to each element: Formula 2: ; In the formula, is the total content of oxygen corresponding to each element, in %; is the logging data corresponding to the E element; is the molar mass fraction of oxygen element in different oxides, unit is %; Finally, the eleven elements are integrated and the calculation formula is as follows: Formula 3: ; In the formula, is the relative percentage content of E element in mud logging, unit is %; is the logging data corresponding to the E element; It is the sum of the logging data of 11 elements: Na, Mg, Al, Si, P, K, Ca, Ti, Mn, Fe and O; S12, obtaining core element data; First, the content of each element is calculated using the core XRF data. Then, the total content of oxygen corresponding to the ten elements is calculated using the content of each element. Finally, the ten elements and oxygen are integrated together. First, find the content of different elements: Formula 4: ; In the formula, is the core percentage content of element E, unit is %; is the relative content of the oxide of the core corresponding to the E element, in units of %; The subsequent calculation process is the same as the element logging data integration method. First, the content of each oxide is calculated using the core percentage content of each element, and then the total content of oxygen corresponding to these ten elements is calculated. Finally, the ten elements and oxygen are integrated together. S13. Evaluate the reliability of integrated element logging data; The logging element data were integrated and compared with the core element data: the element data of Na, Mg, Al, Si, P, K, Ca, Ti, Mn, Fe, etc. after the element logging integration were used to make a cross plot with the core element data, where different colors represent different elements, and the black dotted line is the 45° trend line. It can be seen from the figure that the results of the integrated logging of Na, Mg, Al, Si, P, K, Ca, Ti, Mn, Fe, etc. Percent content of core They are basically evenly distributed around the trend line, proving that the integrated element logging data is reliable. The logging element data are integrated according to Formula 3.

[0006] As a further technical solution of the present invention, in step S2, based on the core mineral data, the complex multi-mineral volume model of different complex lithologies is determined by classifying according to lithology and counting the mineral distribution characteristics of different cores, including the following steps: S21. Count the mineral components of different lithologies; Statistical analysis of mineral compositions under major lithology categories, and analysis of mineral compositions and contents of various secondary lithologies under major lithology categories, and further classification and statistics of secondary lithology minerals; S22. Establish complex multi-mineral volume models with different lithologies; When establishing a complex multi-mineral volume model, the secondary lithological minerals are distinguished into major minerals and minor minerals according to their average content. Major minerals refer to those with a higher content in the rock (average content higher than 10%), including the main minerals that constitute the rock and some occasional minerals with relatively high content; minor minerals refer to those with a lower content in the overall rock (average content higher than 5%). Based on this, the mineral types corresponding to the secondary lithology are screened, with the main minerals with high content being given priority, followed by the low-content secondary minerals that contribute significantly to the logging response. A complex multi-mineral volume model is constructed based on the final mineral composition.

[0007] As a further technical solution of the present invention, in step S3, referring to the conventional well logging mineral inversion response equation, according to the complex component volume model of each lithology, the element response equation of different lithologies is established, and the characteristic values ​​of different mineral elements of each complex lithology are determined by using an intelligent regression algorithm, which includes the following steps: S31, establish the response equation of the element; Referring to the method for establishing the natural gamma response equation in Formula 5, the response equations of each element are established, as shown in Formula 6. They are incorporated into the inversion equation group during the final inversion and participate in the mineral inversion calculation. The response equations are as follows: Formula 5: , ; Formula 6: ; In the formula, Representative Components Indicates The natural gamma value of the components; is the weight percentage content of each element after integration; For the The element characteristic values ​​of the components; For the The weight percentage content of the components; S32. Calculate the element characteristic value by using intelligent regression algorithm in combination with chemical formula; In terms of elemental characteristic values, since the chemical compositions of different minerals are different, there are significant differences in their elemental compositions. Therefore, it is necessary to determine the elemental composition of mineral components in different models based on complex multi-mineral volume models of different lithologies in the study area. Since the chemical formulas of some minerals are fixed, the molar mass fraction of the corresponding elements can be directly calculated with the help of the chemical formula to obtain the elemental characteristic values ​​corresponding to the mineral. After the elemental characteristic values ​​of all minerals with determined chemical formulas are calculated, the intelligent regression algorithm is used to solve the elemental characteristic values ​​of the remaining minerals. The intelligent regression algorithm performs mathematical analysis based on the relationship between independent variables and dependent variables. The present invention uses the elemental logging data of different minerals as dependent variables and the weight content of the corresponding minerals as predictive variables to perform intelligent regression algorithms. Next, the regression equation between elements and minerals is established according to different mineral components.

[0008] Taking the Si element characteristic parameter values ​​of different minerals as an example, the calculation formula of the intelligent regression algorithm model can be obtained: Formula 7: ; The characteristic value of Si element in quartz is known and can be directly substituted into the above formula: is the value of silicon in the integrated element logging data; Represents the rock mineral model The weight percentage content of the mineral components; For the The regression coefficient of the mineral component is also the target we want to obtain, that is, the characteristic value of the element response equation. At the same time, since the element logging measurement is based on the rock skeleton, the element characteristic values ​​of the pore fluid are all 0; While regressing to obtain the characteristic values ​​of different mineral elements, the regression correlation coefficient can also be calculated. The higher the correlation coefficient, the better the relationship between the element and the mineral and the more effective the feature is. The lower the correlation coefficient, the worse the relationship between the mineral and the element, and the characteristic value is a redundant feature.

[0009] As a further technical solution of the present invention, in step S4, the volume content of each component of different complex lithologies is calculated in combination with the comprehensive logging data, and the reliability of the characteristic values ​​of different mineral elements is verified, which includes the following steps: S41. Calculate the volume content of each lithology component in the well by combining comprehensive logging data; The mineral inversion is performed using comprehensive logging data to obtain the volume content of underground mineral components. The volume fraction of each mineral component is compared with the core XRD mineral data. However, the core XRD mineral data is the mass fraction without adding pores, so the calculation results need to be converted. The conversion method is as follows: Formula 8: ; In the formula, is the mass fraction of a certain mineral; For the The volume fraction of the components; is the density of a mineral in the framework; S42. Verify the reliability of characteristic values ​​of different mineral elements; The mineral content obtained by parameter inversion is basically consistent with the mineral content determined by core XRD. Using these two data to make an error table, the overall average absolute error is less than 5%, and the content of each mineral is consistent with the actual rock characteristics, which proves that the obtained mineral element characteristic values ​​are reliable.

[0010] Explanation of terms involved in the present invention: Mineral inversion: Mineral inversion is a technical method to infer the type, content, distribution and other information of minerals in the underground or other research areas through the analysis and processing of geophysical and geochemical data; Element logging: A method of analyzing drill cuttings in oil exploration and development to obtain data on the content of major rock-forming elements in rocks.

[0011] Molar mass: Molar mass is a physical unit. The mass of a substance per unit amount of substance is called molar mass. Molar mass fraction: refers to the ratio of the amount of a component in a mixture to the total amount of the mixture; XRF: X-ray fluorescence spectrometry is a method of determining the type and relative content of elements in a sample by measuring the energy and intensity of characteristic X-ray fluorescence produced by the transition of inner-shell electrons of atoms after being irradiated by high-energy X-rays. Volume model: According to the composition of the rock and the differences in its physical and chemical properties, the rock volume is divided into several corresponding parts, and then the contribution of each part to the measurement data is studied, and the measurement result is regarded as the sum of the contributions of these parts; Response equation: The equation consisting of the actual measured value and the sum of the contribution values ​​of each component in the volume model, that is, the actual measured value is equal to the sum of the contribution values ​​of each component in the volume model; Characteristic parameter: the contribution value of the volume unit of each component in the response equation; Intelligent regression algorithm: It is a regression prediction method that uses machine learning and other methods to automatically mine data patterns and predict or estimate dependent variables by the optimal combination of multiple independent variables; Mean absolute error: It is a statistical indicator that measures the average difference between the predicted value and the actual value by taking the absolute value of the difference between the predicted value and the actual value and averaging it.

[0012] Beneficial effects achieved by the present invention: More mineral types are counted, and the corresponding element characteristic values ​​of both conventional and unconventional minerals can be calculated to meet the inversion needs of multi-mineral complex lithology; It has a wider range of applications, and the element characteristic values ​​calculated from fixed chemical formulas are more accurate; Combined with the intelligent regression algorithm, the workload of calculating element characteristic values ​​is reduced, the overall calculation speed is improved, and the problem of difficulty in determining the characteristic values ​​of mineral components is solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a comparison chart of the integrated logging element data and the core element data in the present invention.

[0014] Figure 2 This is a picture of granite gneiss in the embodiment of the present invention.

[0015] Figure 3 This is a statistical diagram of the mineral distribution of granite gneiss in the study area in the embodiment of the present invention.

[0016] Figure 4 This is a picture of granite according to an embodiment of the present invention.

[0017] Figure 5 This is a statistical diagram of granite mineral distribution in the study area of ​​an embodiment of the present invention.

[0018] Figure 6 This is a complex multi-mineral volume model of gneiss according to an embodiment of the present invention.

[0019] Figure 7 This is a complex multi-mineral volume model of granite according to an embodiment of the present invention.

[0020] Figure 8 This is a comprehensive bar chart of mineral inversion of Well XX according to an embodiment of the present invention.

[0021] Fig. 9 is a flow chart of an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The technical solution of the present invention is described in detail below in conjunction with the specific drawings.

[0023] See also Fig. 9 This embodiment provides a method for determining characteristic parameters of different mineral elements in complex rocks, the method comprising the following steps: S1. Establish an integrated application method for element logging and element logging data, and verify the reliability of the integrated element data by combining core element data; S2. Based on the core mineral data, the complex multi-mineral volume model of different complex lithologies is determined by classifying according to lithology and statistically analyzing the mineral distribution characteristics of different cores; S3. With reference to the conventional well logging mineral inversion response equation, the element response equations of different lithologies are established according to the complex multi-mineral volume model of each lithology, and the characteristic values ​​of different mineral elements of each complex lithology are determined by using the intelligent regression algorithm; S4. Combine comprehensive logging data to calculate the volume content of each component of different complex lithologies and verify the reliability of characteristic values ​​of different mineral elements.

[0024] In this embodiment, in step S1, an integrated application method of element logging and element logging data is established, and the reliability of the integrated element data is verified by combining the core element data, including the following steps: S11, element logging data integration method; Element logging data includes common elements and some trace elements, generally involving details of 17 elements, including common elements such as sodium (Na) and magnesium (Mg), as well as some trace elements such as nickel (Nr) and vanadium (Vr). However, the data provided by the major element analysis of the core, i.e., the core element data, only focuses on 10 common element oxides, showing their respective relative percentage contents, including sodium oxide ( ) and magnesium oxide ( ) etc.; It can be seen that there is a clear difference between the element logging data and the core element data in the scope of element information. In view of this, before using the element logging data to participate in related calculations, it is necessary to first carry out integrated correction operations on it to ensure that the data can match the core element data and meet the requirements of subsequent analysis and calculation; The principle of the element logging integration method is: select 10 elements corresponding to the core oxides in the element logging data, namely sodium (Na), magnesium (Mg), aluminum (Al), silicon (Si), phosphorus (P), potassium (K), calcium (Ca), titanium (Ti), manganese (Mn) and iron (Fe) for integration; first use the logging content of each element to reversely calculate the content of each oxide, then calculate the total content of oxygen corresponding to these ten elements, and finally integrate the ten elements with oxygen. The calculation steps are as follows: First, the molar mass fraction of each non-oxygen element in different oxides is calculated based on the molar mass of the element: Formula 1: ; In the formula, is the molar mass fraction of element E in the corresponding oxide; is the total molar mass of element E in the oxide chemical formula, in units of ; is the molar mass of the oxide corresponding to element E, in units of ; Then calculate the molar mass fraction of oxygen in different oxides , the formula is the same as above; Next, the above two mass fractions are used to calculate the total content of oxygen elements corresponding to each element: Formula 2: ; In the formula, is the total content of oxygen corresponding to each element, in %; is the logging data corresponding to the E element; is the molar mass fraction of oxygen element in different oxides, unit is %; Finally, the eleven elements are integrated and the calculation formula is as follows: Formula 3: ; In the formula, is the relative percentage content of E element in mud logging, unit is %; is the logging data corresponding to the E element (if it is the oxygen element, it is the calculated data); It is the sum of the logging data of 11 elements: Na, Mg, Al, Si, P, K, Ca, Ti, Mn, Fe and O (the oxygen element is the calculated data); S12, obtaining core element data; The method of obtaining core element analysis data is roughly the same as the method of integrating element logging data. First, the content of each element is calculated using the core XRF data. Then, the total content of oxygen corresponding to these ten elements is calculated using the content of each element. Finally, the ten elements and oxygen are integrated together. First, find the content of different elements: Formula 4: ; In the formula, is the core percentage content of element E, unit is %; is the relative content of the oxide of the core corresponding to the E element, in units of %; The subsequent calculation process is the same as the element logging data integration method. First, the content of each oxide is calculated using the core percentage content of each element, and then the total content of oxygen corresponding to these ten elements is calculated. Finally, the ten elements and oxygen are integrated together. S13. Evaluate the reliability of integrated element logging data; The logging element data were integrated and compared with the core element data. The comparative analysis results are as follows: Figure 1 :The cross-plot of the element data of Na, Mg, Al, Si, P, K, Ca, Ti, Mn, Fe and other elements after element logging integration and the core element data is made. Different colors represent different elements, and the black dotted line is the 45° trend line. It can be seen from the figure that the results of Na, Mg, Al, Si, P, K, Ca, Ti, Mn, Fe and other elements after logging integration Percent content of core They are basically evenly distributed around the trend line, proving that the integrated element logging data is reliable. The logging element data are integrated according to Formula 3.

[0025] In this embodiment, in step S2, based on the core mineral data, the cores are classified according to lithology and the mineral distribution characteristics of different cores are counted to determine the multi-mineral model and complex component volume model of different complex lithologies, including the following steps: S21. Count the mineral components of different lithologies; Statistical analysis of mineral compositions under major lithology categories, and analysis of mineral compositions and contents of various secondary lithologies under major lithology categories, and further classification and statistics of secondary lithology minerals; Taking the core mineral data of a certain area as an example, it was found through sorting and statistics that even under the same lithology category, the mineral composition of different types of lithology is still very different. Therefore, it is necessary to judge the main minerals of different lithologies based on the combination characteristics of rock-forming minerals in the study area, and then establish a complex multi-mineral volume model for complex lithologies.

[0026] First, taking metamorphic rocks as an example, it is found that its constituent minerals include clay minerals (including chlorite, kaolinite, illite and montmorillonite), quartz, potassium feldspar, plagioclase, calcite, biotite, amorphous, magnesite, muscovite, rock salt, epidote, hematite, amphibole, ilmenite, pyrite, diopside, dolomite, titanite, barite and zeolite, etc., and various secondary lithologies under the metamorphic rock category, including schist, gneiss and amphibolite, have different mineral compositions and contents, so the secondary lithology minerals are further classified and counted; Here we take gneiss as an example among the secondary lithological minerals. Figure 2 This is a slice of granite gneiss. It can be seen that the mineral composition of granite gneiss is extremely complex. Finally, its mineral distribution map is obtained ( Figure 3 ). Among them, quartz accounts for 34%, which is the most abundant mineral; feldspar accounts for 32%, followed by clay minerals, which account for 25%, also occupying a large share. Calcite, biotite, and amorphous minerals account for 3%, 2%, and 1% respectively, with relatively low contents; other minerals account for 3%, including dolomite, pyrite, common pyroxene, titanite, hornblende, ilmenite and other minerals; Next, taking igneous rocks as an example, it is found that its constituent minerals include clay minerals (including chlorite, kaolinite, illite and montmorillonite), quartz, potassium feldspar, plagioclase, calcite, dolomite, lauric zeolite, iron dolomite, anhydrite, pyrite, siderite, magnetite, hematite, amphibole, biotite, common pyroxene, diopside, rock salt, barite, ilmenite, epidote, muscovite, zeolite, spinel, amorphous and other 25 minerals. The various secondary lithologies under the igneous rock category, such as basalt, andesite, diabase, gabbro, diorite, granite, etc., have different mineral compositions and contents, so the secondary lithology minerals are further classified and counted; Here we take granite in the secondary lithology as an example. Figure 4 This is a granite slice diagram. It can be observed that the mineral composition of granite is extremely complex. Finally, its mineral distribution map is obtained ( Figure 5). Among them, feldspar accounts for 37%, quartz accounts for 27%, clay minerals account for 9%, amorphous minerals account for 6%, epidote accounts for 5%, muscovite accounts for 4%, calcite accounts for 2%, dolomite accounts for 1%, and other minerals account for 9%, including occasional minerals such as ilmenite, anhydrite, pyrite, siderite, magnetite, etc.; S22. Establish complex multi-mineral volume models with different lithologies; In order to ensure the feasibility of multi-mineral inversion and the rationality of complex multi-mineral volume models, the different minerals of each secondary lithology are reasonably deleted based on the chemical properties and logging characteristics of the above-mentioned minerals; it is found that when the mineral content is less than 5%, the contribution to the rock characteristics is low and it is difficult to invert. Therefore, when establishing a complex multi-mineral volume model, the secondary lithology minerals are distinguished into major minerals and minor minerals according to the average content. Among them, major minerals refer to those with a higher content in the rock (average content higher than 10%), including the main minerals that constitute the rock and some occasional minerals with relatively high content; minor minerals refer to those with a lower content in the overall rock (average content higher than 5%). Based on this, the mineral types corresponding to the secondary lithology are screened, with the main minerals with high content being selected first, followed by the secondary minerals with low content that contribute significantly to the logging response. A complex multi-mineral volume model is constructed based on the final mineral composition. The first is gneiss, and finally a complex multi-mineral volume model consisting of eight components, including quartz, feldspar, clay minerals, amorphous, calcite, biotite, dolomite, and pores, is obtained ( Figure 6 ); Secondly, granite, and finally a complex multi-mineral volume model consisting of 8 components including quartz, feldspar, clay minerals, epidote, muscovite, calcite, amorphous and pores was obtained ( Figure 7 ), it can be found that the mineral composition of each rock type is not the same, and each has its own proprietary minerals.

[0027] In this embodiment, in step S3, referring to the conventional well logging mineral inversion response equation, according to the complex multi-mineral volume model of each lithology, the element response equations of different lithologies are established, and the characteristic values ​​of different mineral elements of each complex lithology are determined by using the intelligent regression algorithm, including the following steps: S31, establish the response equation of the element; Referring to the establishment method of the response equation of conventional logging curves such as natural gamma (GR) in Formula 5, the response equations of each element are established, as shown in Formula 6. They are incorporated into the inversion equation group during the final inversion and participate in the mineral inversion calculation. The response equations are as follows: Formula 5: , ; Formula 6: ; In the formula, Representative Components Indicates The natural gamma value of the components; is the weight percentage content of each element after integration; For the The element characteristic values ​​of the components; For the The weight percentage content of the components; S32. Calculate the element characteristic value by using intelligent regression algorithm in combination with chemical formula; In terms of elemental characteristic values, since the chemical compositions of different minerals are different, their elemental compositions are significantly different. Therefore, it is necessary to determine the elemental composition of mineral components in different models based on complex multi-mineral volume models of different lithologies in the study area. In view of the fact that the chemical formulas of some minerals are fixed, the molar mass fraction of the corresponding elements can be directly calculated with the help of the chemical formula to obtain the elemental characteristic value corresponding to the mineral, which is also the elemental characteristic value corresponding to the mineral. After the elemental characteristic values ​​of all minerals with determined chemical formulas are calculated, the intelligent regression algorithm is used to solve the elemental characteristic values ​​of the remaining minerals. In this way, the number of unknown quantities in the calculation equation will be greatly reduced, and the corresponding amount of calculation will also be reduced accordingly. The intelligent regression algorithm performs mathematical analysis based on the relationship between independent variables and dependent variables. The present invention uses the element logging data of different minerals as dependent variables and the weight content of the corresponding minerals as predictive variables to perform intelligent regression algorithms. Next, we establish the regression equation between elements and minerals according to different mineral components; here we take the gneiss in the study area as an example for calculation (the same applies to granite). According to the previous results, the complex multi-mineral volume model of gneiss in this area is composed of 8 components: quartz, feldspar, clay minerals, calcite, amorphous, biotite, dolomite and pores. The minerals with fixed chemical formulas are as follows: Quartz (SiO2), in which the element characteristic value of Si can be calculated by relative atomic mass, and finally the characteristic value of Si in quartz is 0.467. Since this mineral only contains Si, an element for which the characteristic value needs to be calculated, the characteristic values ​​of other elements are all 0.

[0028] Calcite (CaCO3), the calculation principle is the same as above, the characteristic value of Ca element in calcite is 0.4, and the characteristic values ​​of other elements are all 0.

[0029] Dolomite (CaMg(CO3)2), the calculation principle is the same as above, the characteristic value of Ca element in dolomite is 0.304, the characteristic value of Mg element is 0.218, and the characteristic values ​​of other elements are all 0.

[0030] Taking the Si element characteristic parameter values ​​of different minerals as an example, the calculation formula of the intelligent regression algorithm model can be obtained: Formula 7: ; The characteristic value of Si element in quartz is known and can be directly substituted into the above formula: is the value of silicon in the integrated element logging data; Represents the rock mineral model The weight percentage content of the mineral components; For the The regression coefficient of the mineral component is also the target we want to obtain, that is, the characteristic value of the element response equation. At the same time, since the element logging measurement is based on the rock skeleton, the element characteristic values ​​of the pore fluid are all 0; While regressing to obtain the characteristic values ​​of different mineral elements, the regression correlation coefficient can also be calculated. The higher the correlation coefficient, the better the relationship between the element and the mineral and the more effective the feature is. The lower the correlation coefficient, the worse the relationship between the mineral and the element, and the redundant feature is. Finally, the characteristic value results of gneiss elements (Table 1) and granite elements (Table 2) were calculated. These tables not only contain the characteristic values ​​of several common minerals, but also the characteristic values ​​of unconventional minerals such as biotite, muscovite, amorphous, epidote, etc.

[0031] Table 1 Characteristic values ​​of elements in granite gneiss ; Table 2 Characteristic values ​​of granite elements ; In this embodiment, in step S4, the volume content of each component of different complex lithologies is calculated in combination with the comprehensive logging data, and the reliability of the characteristic values ​​of different mineral elements is verified, which includes the following steps: S41. Calculate the volume content of each lithology component in the well by combining comprehensive logging data; The actual data in the XX well are used for calculation. The data input includes conventional curves, such as natural gamma (GR), acoustic time difference (AC), compensated neutron ( )、Compensation density( ) and element data curves, such as silicon (Si) and aluminum (Al). The inverted mineral types include quartz, feldspar, clay minerals, calcite, amorphous, biotite and dolomite.

[0032] The final result obtained by this method is the volume fraction of each mineral component. In order to verify the accuracy of the result, the volume fraction of each mineral component is compared with the core XRD mineral data. However, the core XRD mineral data is the mass fraction without adding pores, so the calculation results need to be converted; The conversion method is as follows: Formula 8: ; In the formula, is the mass fraction of a certain mineral; For the The volume fraction of the components; is the density of a mineral in the framework; The final inversion result after conversion is as follows Figure 8 As shown in the figure, the red short column represents the mineral content determined by the core XRD, and the black curve is the mineral content curve in the well calculated using the above parameters; S42. Verify the reliability of characteristic values ​​of different mineral elements; It can be seen from observation that the mineral content obtained by inversion using the above parameters is basically consistent with the mineral content determined by core XRD. The error table of these two data is shown in Table 3. The overall average absolute error is less than 5%, and the content of each mineral is consistent with the actual rock characteristics, which proves that the characteristic values ​​of mineral elements calculated by this process are reliable.

[0033] Table 3-XX Well Mineral Average Absolute Error Table ; First, the mineral data corresponding to each lithology in the study area were statistically analyzed. Then, the element logging data was corrected to obtain integrated element logging data, laying the data foundation. A reasonable complex multi-mineral volume model was established based on the mineral analysis results of core thin section identification. Next, the element response equation was established with reference to the conventional logging optimization response equation. The integrated element logging and logging data were used, combined with the mineral chemical formula to reduce the unknown quantities, and the element characteristic parameters of the complex mineral components were obtained using an intelligent regression algorithm.

[0034] It should be noted that, in this article, the term "comprises" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of more restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0035] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for determining characteristic parameters of different mineral elements in complex rocks, characterized in that: The method comprises the following steps: S1. Establish an integrated application method for element logging and element logging data, and verify the reliability of the integrated element data by combining core element data; S2. Based on the core mineral data, the complex multi-mineral volume model of different complex lithologies is determined by classifying according to lithology and statistically analyzing the mineral distribution characteristics of different cores; S3. With reference to the conventional well logging mineral inversion response equation, the element response equations of different lithologies are established according to the complex multi-mineral volume model of each lithology, and the characteristic values ​​of different mineral elements of each complex lithology are determined by using the intelligent regression algorithm; S4. Combine comprehensive logging data to calculate the volume content of each component of different complex lithologies and verify the reliability of characteristic values ​​of different mineral elements.

2. The method for determining characteristic parameters of different mineral elements of complex rocks according to claim 1, characterized in that: In step S1, an integrated application method of element logging and element logging data is established, and the reliability of the integrated element data is verified by combining the core element data, including the following steps: S11, element logging data integration method; Element logging data include common elements and some trace elements. The data provided by core major element analysis, i.e. core element data, focuses only on 10 common element oxides, presenting their respective relative percentage contents, including sodium oxide and magnesium oxide. The principle of the element logging integration method is: select 10 elements corresponding to the core oxides in the element logging data, namely sodium, magnesium, aluminum, silicon, phosphorus, potassium, calcium, titanium, manganese and iron, for integration; first use the logging content of each element to reversely calculate the content of each oxide, then calculate the total content of oxygen corresponding to these ten elements, and finally integrate the ten elements with oxygen. The calculation steps are as follows: First, the molar mass fraction of each non-oxygen element in different oxides is calculated based on the molar mass of the element: Formula 1: ; In the formula, is the molar mass fraction of element E in the corresponding oxide; is the total molar mass of element E in the oxide chemical formula, in units of ; is the molar mass of the oxide corresponding to element E, in units of ; Then calculate the molar mass fraction of oxygen in different oxides , the formula is the same as above; Next, we use the mass fraction to find the total content of oxygen elements corresponding to each element: Formula 2: ; In the formula, is the total content of oxygen corresponding to each element, in %; is the logging data corresponding to the E element; is the molar mass fraction of oxygen element in different oxides, unit is %; Finally, the eleven elements are integrated and the calculation formula is as follows: Formula 3: ; In the formula, is the relative percentage content of E element in mud logging, unit is %; is the logging data corresponding to the E element; It is the sum of the logging data of 11 elements: Na, Mg, Al, Si, P, K, Ca, Ti, Mn, Fe and O; S12, obtaining core element data; First, the content of each element is calculated using the core XRF data. Then, the total content of oxygen corresponding to the ten elements is calculated using the content of each element. Finally, the ten elements and oxygen are integrated together. First, find the content of different elements: Formula 4: ; In the formula, is the core percentage content of element E, unit is %; is the relative content of the oxide of the core corresponding to the E element, unit is %; The subsequent calculation process is the same as the element logging data integration method. First, the content of each oxide is calculated using the core percentage content of each element, and then the total content of oxygen corresponding to these ten elements is calculated. Finally, the ten elements and oxygen are integrated together. S13. Evaluate the reliability of integrated element logging data; The logging element data were integrated and compared with the core element data. The Na, Mg, Al, Si, P, K, Ca, Ti, Mn, and Fe element data after element logging integration were used to make a cross plot with the core element data. Different colors represent different elements, and the black dotted line is a 45° trend line. It can be seen from the figure that the results of Na, Mg, Al, Si, P, K, Ca, Ti, Mn, and Fe element logging integration Percent content of core They are basically evenly distributed around the trend line, proving that the integrated element logging data is reliable. The logging element data are integrated according to Formula 3.

3. The method for determining characteristic parameters of different mineral elements of complex rocks according to claim 1, characterized in that: In step S2, based on the core mineral data, the cores are classified according to lithology and the mineral distribution characteristics of different cores are counted to determine the complex multi-mineral volume model of different complex lithologies, including the following steps: S21. Count the mineral components of different lithologies; Statistical analysis of mineral compositions under major lithology categories, and analysis of mineral compositions and contents of various secondary lithologies under major lithology categories, and further classification and statistics of secondary lithology minerals; S22. Establish complex multi-mineral volume models with different lithologies; When establishing a complex multi-mineral volume model, the secondary lithological minerals are distinguished into major minerals and minor minerals according to their average content, where major minerals refer to minerals with an average content higher than 10% in the rock; minor minerals refer to minerals with an average content higher than 5% in the overall rock. Based on this, the mineral types of secondary lithology are screened, with priority given to the main minerals with high content, followed by the low-content secondary minerals that contribute significantly to the logging response. A complex multi-mineral volume model is constructed based on the final mineral composition.

4. The method for determining characteristic parameters of different mineral elements of complex rocks according to claim 1, characterized in that: In step S3, referring to the conventional well logging mineral inversion response equation, according to the complex multi-mineral volume model of each lithology, the element response equations of different lithologies are established, and the characteristic values ​​of different mineral elements of each complex lithology are determined by using the intelligent regression algorithm, including the following steps: S31, establish the response equation of the element; Referring to the method for establishing the natural gamma response equation in Formula 5, the response equations of each element are established, as shown in Formula 6. They are incorporated into the inversion equation group during the final inversion and participate in the mineral inversion calculation. The response equations are as follows: Formula 5: , ; Formula 6: ; In the formula, Representative Components Indicates The natural gamma value of the components; is the weight percentage content of each element after integration; For the The element characteristic values ​​of the components; For the The weight percentage content of the components; S32. Calculate the element characteristic value by using intelligent regression algorithm in combination with chemical formula; In terms of elemental characteristic values, since the chemical compositions of different minerals are different, their elemental compositions vary significantly. Therefore, it is necessary to determine the elemental composition of mineral components in different models based on complex multi-mineral volume models of different lithologies in the study area. Since the chemical formulas of some minerals are fixed, the molar mass fraction of the corresponding elements can be directly calculated using the chemical formula to obtain the elemental characteristic values ​​corresponding to the mineral. After the elemental characteristic values ​​of all minerals with determined chemical formulas have been calculated, the intelligent regression algorithm is used to solve the elemental characteristic values ​​of the remaining minerals. The intelligent regression algorithm performs mathematical analysis based on the relationship between independent variables and dependent variables, and uses the elemental logging data of different minerals as dependent variables and the weight content of the corresponding minerals as the predictive variable to perform the intelligent regression algorithm. Next, the regression equation between elements and minerals is established according to different mineral components; Taking the Si element characteristic parameter values ​​of different minerals as an example, the calculation formula of the intelligent regression algorithm model can be obtained: Formula 7: ; The characteristic value of Si element in quartz is known and can be directly substituted into the above formula: is the value of silicon in the integrated element logging data; Represents the rock mineral model The weight percentage content of the mineral components; For the The regression coefficient of the mineral component is also the target we want to obtain, that is, the characteristic value of the element response equation. At the same time, since the element logging measurement is based on the rock skeleton, the element characteristic values ​​of the pore fluid are all 0; While regressing to obtain the characteristic values ​​of different mineral elements, the regression correlation coefficient can also be calculated. The higher the correlation coefficient, the better the relationship between the element and the mineral and the more effective the feature is. The lower the correlation coefficient, the worse the relationship between the mineral and the element, and the characteristic value is a redundant feature.

5. The method for determining characteristic parameters of different mineral elements of complex rocks according to claim 1, characterized in that: In step S4, the volume content of each component of different complex lithologies is calculated in combination with the comprehensive logging data, and the reliability of the characteristic values ​​of different mineral elements is verified, including the following steps: S41. Calculate the volume content of each lithology component in the well by combining comprehensive logging data; The mineral inversion is performed using comprehensive logging data to obtain the volume content of underground mineral components. The volume fraction of each mineral component is compared with the core XRD mineral data. However, the core XRD mineral data is the mass fraction without adding pores, so the calculation results need to be converted. The conversion method is as follows: Formula 8: ; In the formula, is the mass fraction of a certain mineral; For the The volume fraction of the components; is the density of a mineral in the framework; S42. Verify the reliability of characteristic values ​​of different mineral elements; The mineral content obtained by parameter inversion is basically consistent with the mineral content determined by core XRD. The overall average absolute error of the error table using these two data is less than 5%, and the content of each mineral is consistent with the actual rock characteristics, which proves that the obtained mineral element characteristic values ​​are reliable.

Images