Composite mineralization end effect geophysical in-situ characterization method for rock physical property constraint

Abstract

The invention provides a rock physical property constrained composite mineralization end effect geophysical in-situ characterization method, and relates to the technical field of metal mineral geophysical prospecting, the method comprises the following steps: firstly, testing pressure and temperature conditions at different depths on a rock sample to obtain physical property influence rules of temperature and pressure on different rock types; then utilizing Bayesian classification to take rock (ore) physical property parameters as training and testing samples to carry out probability statistics under a variable temperature and pressure condition, establishing a deep lithology identification equation, forming an accurate rock physical property correlation function of a key geologic body related to mineralization, and finally, based on a Gramian constraint method, calculating the rock physical property correlation function of the key geologic body related to mineralization. Fusion of multiple geophysical data is achieved under the constraint of rock physical property prior information, a three-dimensional comprehensive geophysical model under the composite mineralization effect is obtained, and then conditions, processes and laws during accumulation of mineralization elements or compounds are represented. According to the method, the defects in the prior art can be overcome, and the geophysical in-situ characterization technology is realized.

Description

Technical Field

[0001] The invention relates to the technical field of geophysical exploration of metal minerals, and in particular to a geophysical in-situ characterization method for composite mineralization end-effects constrained by rock physical properties. Background Art

[0002] Based on the mineralization system, the basic issues of ore deposit research can be summarized as: the material source of mineralization "source", the transportation and migration process "transport", and the concentration and accumulation of ore-forming elements or compounds "gathering". The terminal effect of mineralization, corresponding to the "gathering" in mineralization, refers to the study of the conditions, processes and laws of the accumulation of ore-forming elements or compounds, and is also the essential issue of deep mineral exploration. Magmatism is an important way for metals from deep in the earth to migrate to the surface. It has the characteristics of multi-stage and multi-style. When different stages and (or) different styles of magma intrusion interact with each other, the mineralization will be complex. Under the superposition of multi-stage and multi-type magma-ore formation, large or even super-large deposits are easily formed. Therefore, choosing a suitable method to accurately characterize the terminal effect of composite mineralization will be beneficial to obtain structural information, understand the aggregation mechanism of ore-forming elements, provide guidance for the prediction of large and super-large deposits, and promote a new round of strategic mineral exploration in my country.

[0003] The main means of characterizing the terminal effect of composite mineralization are geochemical testing and geophysical exploration. Geochemical testing methods test the isotopes, temperature, pressure and composition of rocks (ores), fluid inclusions, etc., which can quantitatively obtain the age, temperature and pressure of magma-ore formation, and the source of ore-forming materials, and provide time and attribute information for exploring the terminal effect of composite mineralization. However, geochemical research cannot effectively reflect the fine spatial structure information of composite magma-ore formation, and cannot reveal the macroscopic main controlling factors and mineralization laws of composite mineralization. Compared with geochemical testing, geophysical exploration can obtain the fine structure of ore-forming elements based on the characteristics of physical property differences, and provide spatial and physical property information for exploring the terminal effect of composite mineralization. However, there are still challenges in using geophysical methods to study the terminal effect of mineralization. The key problem is that the resolution of a single geophysical method is limited and there is non-uniqueness. Therefore, how to establish a unified geophysical joint inversion framework and realize the coupling of multiple geophysical data is an urgent problem to be solved.

[0004] The key to achieving geophysical joint inversion lies in the constraint method. At present, structural constraints and rock property constraints are commonly used in geophysical joint inversion. The correlation between geometric shapes or different rock property parameters is used to promote the coupling of multiple geophysical data. Compared with structural constraints, rock property constraints have a stronger coupling effect, but there are still the following problems: (1) Rock property constraints require higher prior information and require the establishment of accurate rock property correlation functions to achieve multi-parameter coupling; (2) The physical property parameters used in the rock property constraint method are derived from surface rock samples and cannot represent the true physical properties of the underground medium, making the inversion results may not truly reflect the underground structural properties. Studies on the end effects of mineralization have found that, especially in composite mineralization areas, mineralization will cause multiple changes in the structure and properties of the underground medium. The geophysical results reflect the spatial structural characteristics of the physical state of the underground medium in a short period of time and cannot fully characterize the composite mineralization end effects. Therefore, it is very necessary to design a geophysical in-situ characterization method for composite mineralization end effects constrained by rock properties under variable temperature and pressure conditions. Summary of the invention

[0005] In order to overcome the deficiencies of the prior art, the object of the present invention is to provide a geophysical in-situ characterization method for composite mineralization end-effects constrained by rock physical properties.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] The present invention provides a geophysical in-situ characterization method for composite mineralization end-effect constrained by rock physical properties, comprising:

[0008] Step 1: Based on the rock physical property comprehensive testing system, the temperature and pressure at different depths are simulated to measure the resistivity, permeability and porosity of rock samples of key geological units in the composite mineralization area;

[0009] Step 2: Determine the linear relationship between temperature, pressure and the resistivity, permeability and porosity of rock samples, and use a multivariate regression model to establish resistivity, permeability and porosity models at different temperatures and pressures;

[0010] Step 3: Measure the permeability, porosity and rock density of different rock types indoors, and establish rock density models under different permeabilities and porosities;

[0011] Step 4: Substitute the permeability and porosity models under different temperatures and pressures into the rock density models under different permeabilities and porosities to obtain the rock density models under different temperatures and pressures;

[0012] Step 5: Based on a conventional magnetic susceptibility meter, obtain the magnetic susceptibility of the rock sample;

[0013] Step 6: Use samples with known lithology as training sets, use Bayesian classification to perform probability statistics on the obtained data, and determine the lithology of unknown samples;

[0014] Step 7: Based on the rock type and the physical property data after standardization, the deep lithology is discriminated by discriminant analysis, and a deep lithology identification equation considering temperature and pressure conditions is established;

[0015] Step 8: Based on the linear regression model, the resistivity, density and magnetic susceptibility data under variable temperature and pressure conditions are processed to obtain the prior information of the physical properties of resistivity and density, resistivity and magnetic susceptibility;

[0016] Step 9: Perform 3D inversion using electromagnetic data to obtain a resistivity model;

[0017] Step 10: Using the resistivity model as a common reference model, based on prior information on physical properties, it is converted into an initial model for joint inversion of gravity and magnetic methods;

[0018] Step 11: Integrate the electromagnetic inversion results into the wide range constraints as a reference model, and carry out the minimization of the objective function of the three-dimensional joint inversion of gravity and magnetic methods considering Gramian constraints.

[0019] Preferably, in step 2, the linear relationship between temperature, pressure and the resistivity, permeability and porosity of the rock sample is determined, and a multivariate regression model is used to establish a resistivity, permeability and porosity model at different temperatures and pressures, specifically:

[0020] According to the different rock types, the influence of temperature T and pressure p at different depths on rock resistivity R, permeability k and porosity φ is statistically analyzed. For the case of linear relationship, the multivariate linear regression model is used to establish the relationship between resistivity, permeability and porosity of different rock types and temperature and pressure, that is: ; ; ; In the case of nonlinear relationships, a nonlinear regression model is used, which is expressed as ; ; ; In the formula, a i 、b i 、c i d i 、e i and f i is the coefficient, i=1, 2, 3,…, n.

[0021] Preferably, in step 3, the permeability, porosity and rock density of different rock types in the room are measured, and a rock density model under different permeabilities and porosities is established, specifically:

[0022] The permeability and porosity of different rock types at room temperature and pressure are measured, and the rock density ρ is measured using a density meter in an indoor environment to determine the linear relationship between the permeability and porosity of different rock types at room temperature and pressure and the density. A multivariate linear regression model or a nonlinear regression model is used to establish a relationship model between the density and permeability and porosity of different rock types, that is, ; or .

[0023] Preferably, in step 4, the permeability and porosity models at different temperatures and pressures are introduced into the rock density models at different permeabilities and porosities to obtain the rock density models at different temperatures and pressures, specifically:

[0024] The permeability and porosity models under different temperatures and pressures are introduced into the rock density models under different permeability and porosity to obtain the rock density models under different temperatures and pressures, that is, ; or, .

[0025] Preferably, in step 6, samples of known lithology are used as training sets, and Bayesian classification is used to perform probability statistics on the obtained data to determine the lithology of unknown samples, specifically:

[0026] According to the obtained resistivity, density and magnetic susceptibility data of different rock types under different temperature and pressure conditions, the lithology identification of deep rock mass is carried out. Bayesian classification is used to take the rock physical property parameters under variable temperature and pressure conditions as training and test samples for probability statistics. Among them, the rock physical property parameters include density, resistivity and magnetic susceptibility. The rock type is represented as c event and the rock physical property parameter is represented as x event. The rock type c and the corresponding attribute x are two events occurring in the same space. It is assumed that the complete rock type c is composed of multiple rock types c1, c2, ..., c N If x occurs, then a type in c must occur. According to the probability, we can get the known categories c of the sample set. j The prior probability and various conditional probabilities For unknown samples, the Bayesian formula can calculate the probability of the type to which the sample to be tested belongs, that is, the posterior probability, which is:

[0027] Given the rock property parameter x, the probability of a certain rock type is ; In the formula, x represents the rock property, c j (j=1, 2,···, N) are N different rock types, P(x / c j ) is assumed to be rock type c j The probability of observing x when j ) indicates that before observing x, a specific type c j The prior probability of P(x) is the probability of observing x from all samples, that is, ; For a given point x, P(x) is determined by Indicates that the unknown sample x belongs to c j The probability density of the class, then P(x / c j ) can be expressed as the probability density , where a represents a very small interval near x, and the posterior probability is .

[0028] Preferably, in step 7, the rock type and the physical property data after the normalization are used to discriminate the deep lithology by using the discriminant analysis method, and a deep lithology identification equation considering the temperature and pressure conditions is established, specifically:

[0029] After probabilistic statistics of rock physical parameters under variable temperature and pressure conditions, the lithology, density, magnetic susceptibility and resistivity of all samples were determined, and data standardization was carried out. The deep lithology was discriminated based on the discriminant analysis method. According to the rock type and the corresponding physical parameters, the discriminant function was obtained, which is in the form of ; In the formula, y is the discriminant value, is the standardized deep rock physical parameter, is the discriminant coefficient of each physical property parameter. By substituting the physical property parameter value into the above formula, the discriminant function value corresponding to each type of rock can be obtained, and the type corresponding to the maximum function value is the rock type, thereby establishing a deep lithology identification equation considering temperature and pressure conditions, systematically summarizing the geophysical response characteristics of key geological bodies, and providing constraints for the joint inversion of multiple geophysical data.

[0030] Preferably, in step 8, based on a linear regression model, the resistivity, density and magnetic susceptibility data under variable temperature and pressure conditions are processed to obtain the physical property prior information of resistivity and density, resistivity and magnetic susceptibility, specifically, Based on the obtained resistivity, density and magnetic susceptibility of all rock samples, a linear regression model is used to form the physical property prior information of resistivity and density, resistivity and magnetic susceptibility, that is, ; ; In the formula, are the coefficients of each parameter.

[0031] Preferably, in step 9, three-dimensional inversion is performed using electromagnetic data to obtain a resistivity model, specifically, The three-dimensional inversion is performed using electromagnetic data to obtain the resistivity model. The objective function of the resistivity model is: ; In the formula, is the model parameter vector, is the weighting coefficient of gravity, magnetic and electromagnetic data, α is the regularization factor, express , is the observed data covariance matrix, Stabilizing functionals for the model , is the model weight matrix, is a physical property constraint term. If Gramian constraint is selected, it can be expressed as: ; In the formula, i is the weight coefficient of the i-th Gramian constraint term, i=1, 2, 3, is the Gramian constraint coupling term, is the weight coefficient of the i-th penalty function term, i=1, 2, 3, and is also constant during the inversion process. It is a constraint item of penalty function. When it exceeds the limit range, it acts as a constraint.

[0032] Preferably, in step 10, the resistivity model is used as a common reference model and converted into an initial model for joint inversion of gravity and magnetic methods based on prior information of physical properties, specifically, The resistivity model is used as a common reference model to integrate the prior information of physical properties into the joint inversion of geophysics. The inversion results of electromagnetic data are converted into a common reference model for the joint inversion of gravity and magnetic data. ; ; ; In the formula, and represent the a priori relationships among density, magnetic susceptibility and resistivity, respectively.

[0033] Preferably, in step 11, the electromagnetic inversion results are integrated into the wide range constraints as a reference model, and the three-dimensional joint inversion objective function minimization of the gravity and magnetic methods considering the Gramian constraints is carried out respectively, specifically:

[0034] The electromagnetic inversion results are integrated into the wide range constraints as a reference model, and the three-dimensional joint inversion objective function minimization of the gravity and magnetic methods considering the Gramian constraints is carried out separately, that is, ; .

[0035] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0036] The present invention provides a geophysical in-situ characterization method for composite mineralization end-effect constrained by rock physical properties, the method comprising: simulating temperature and pressure at different depths based on a comprehensive rock physical property testing system, measuring resistivity, permeability and porosity of rock samples of key geological units in the composite mineralization area, judging the linear relationship between temperature, pressure and resistivity, permeability and porosity of rock samples, using a multivariate regression model to establish resistivity, permeability and porosity models at different temperatures and pressures, measuring permeability, porosity and rock density of different rock types indoors, establishing rock density models at different permeabilities and porosities, bringing the permeability and porosity models at different temperatures and pressures into rock density models at different permeabilities and porosities, obtaining rock density models at different temperatures and pressures, and obtaining rock sample density models based on a conventional magnetic susceptibility meter. , using known lithology samples as training sets, using Bayesian classification to perform probability statistics on the obtained data, determining the lithology of unknown samples, using discriminant analysis to discriminate deep lithology for rock types and physical property data after standardization, establishing a deep lithology identification equation considering temperature and pressure conditions, based on a linear regression model, processing resistivity, density and magnetic susceptibility data under variable temperature and pressure conditions, obtaining physical property prior information of resistivity and density, resistivity and magnetic susceptibility, using electromagnetic data for three-dimensional inversion, obtaining a resistivity model, using the resistivity model as a common reference model, based on physical property prior information, converting it into an initial model for joint inversion of gravity and magnetic methods, integrating the electromagnetic inversion results as a reference model into a wide range of constraints, and carrying out three-dimensional joint inversion objective function minimization of gravity and magnetic methods considering Gramian constraints. The present invention has the following advantages:

[0037] 1. The present invention applies the test system to simulate the variable temperature and pressure conditions at different depths to carry out rock physical property testing, which can effectively restore the temperature and pressure environment at the real depth, extract the real physical property characteristics of the underground medium, and provide accurate constraints for geophysical information fusion;

[0038] 2. The present invention has formed a set of "real rock physical property constraints - deep lithology identification - multi-source geophysical coupling - composite mineralization law" technology. In terms of lithology identification and geophysical information fusion, the Bayesian classification and discriminant analysis method is used to avoid the influence of subjective judgment on the lithology of unknown samples, and the joint inversion scheme considering Gramian constraints is used to integrate multiple geophysical information to a great extent;

[0039] 3. The present invention studies the composite mineralization area from two aspects: deep rock properties and structural composition of inversion results, which can comprehensively characterize the enrichment conditions and laws of mineralizing materials and further determine the enrichment location of the ore body. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0041] Figure 1 A flow chart of a method provided by an embodiment of the present invention;

[0042] Figure 2 A flowchart of a method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] The purpose of the present invention is to provide a geophysical in-situ characterization method for composite mineralization end-effects constrained by rock physical properties. Compared with traditional methods, this method breaks through the bottleneck that the parameters constrained by rock physical properties cannot represent the real physical properties of underground media and have strict requirements on prior information. It avoids the defects of the traditional single characterization system of geochemistry or integrated geophysics of composite mineralization end-effects, can obtain the underground structure and medium properties of composite mineralization areas, and accurately characterize the composite mineralization end-effects.

[0045] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0046] Figure 1 A flow chart of a method provided by an embodiment of the present invention, Figure 2 A flowchart of a method provided by an embodiment of the present invention, such as Figure 1 and Figure 2 As shown, the present invention provides a geophysical in-situ characterization method for composite mineralization end-effect constrained by rock physical properties, comprising:

[0047] Step 1: Based on the rock physical property comprehensive testing system, the temperature and pressure at different depths are simulated to measure the resistivity, permeability and porosity of rock samples of key geological units in the composite mineralization area;

[0048] Step 2: Determine the linear relationship between temperature, pressure and the resistivity, permeability and porosity of rock samples, and use a multivariate regression model to establish resistivity, permeability and porosity models at different temperatures and pressures;

[0049] Step 3: Measure the permeability, porosity and rock density of different rock types indoors, and establish rock density models under different permeabilities and porosities;

[0050] Step 4: Substitute the permeability and porosity models under different temperatures and pressures into the rock density models under different permeabilities and porosities to obtain the rock density models under different temperatures and pressures;

[0051] Step 5: Based on a conventional magnetic susceptibility meter, obtain the magnetic susceptibility of the rock sample;

[0052] Step 6: Use samples with known lithology as training sets, use Bayesian classification to perform probability statistics on the obtained data, and determine the lithology of unknown samples;

[0053] Step 7: Based on the rock type and the physical property data after standardization, the deep lithology is discriminated by discriminant analysis, and a deep lithology identification equation considering temperature and pressure conditions is established;

[0054] Step 8: Based on the linear regression model, the resistivity, density and magnetic susceptibility data under variable temperature and pressure conditions are processed to obtain the prior information of the physical properties of resistivity and density, resistivity and magnetic susceptibility;

[0055] Step 9: Perform 3D inversion using electromagnetic data to obtain a resistivity model;

[0056] Step 10: Using the resistivity model as a common reference model, based on prior information on physical properties, it is converted into an initial model for joint inversion of gravity and magnetic methods;

[0057] Step 11: Integrate the electromagnetic inversion results into the wide range constraints as a reference model, and carry out the minimization of the objective function of the three-dimensional joint inversion of gravity and magnetic methods considering Gramian constraints.

[0058] In step 1, based on the rock physical property comprehensive testing system to simulate the temperature and pressure at different depths, the resistivity, permeability and porosity of the rock samples of the key geological units in the composite mineralization area are measured, specifically:

[0059] The comprehensive rock physical property testing system BenchLab7000 produced by NER Company of the United States (the maximum parameters can be set to 10000 psi confining pressure and 200 ℃) was used to measure the physical properties of rock samples in the key geological units of the composite mineralization area. The temperature and pressure at different depths were calculated according to the pressure and temperature gradient, and the corresponding temperature and pressure parameters were set to obtain the rock resistivity, permeability and porosity at different pressures and temperatures.

[0060] In step 2, the linear relationship between temperature, pressure and the resistivity, permeability and porosity of the rock sample is determined, and a multivariate regression model is used to establish the resistivity, permeability and porosity models at different temperatures and pressures, specifically:

[0061] According to different rock types, the influence of temperature T and pressure p at different depths on rock resistivity R, permeability k and porosity φ is statistically analyzed. For the case of linear relationship, the multivariate linear regression model is used to establish the relationship between resistivity, permeability and porosity of different rock types and temperature and pressure, that is: ; ; ; In the case of nonlinear relationships, a nonlinear regression model is used, which is expressed as: ; ; ; In the formula, a i 、b i 、c i d i 、e i and f i is the coefficient, i=1, 2, 3. For the relationship between resistivity, temperature and pressure, the constant term a1 ranges from 10 to 10 4 Ω·m, for the linear regression model, the coefficient b1 of temperature T is usually in the range of 0.001~0.01Ω·m / ℃, and the coefficient c1 of pressure p is usually in the range of -0.1~0.1 Ω·m / GPa; for the nonlinear regression model, the coefficient b1 of the first term of temperature T is usually in the range of 0.001~0.01 Ω·m / ℃, and the coefficient c1 of the second term of temperature T is usually in the range of 10 -6 ~10 -4 Ω·m / ℃2 The first-order coefficient d1 of pressure p is usually in the range of -0.1~0.1 Ω·m / GPa, and the second-order coefficient e1 of pressure p is usually in the range of -0.01~0.01 Ω·m / GPa 2 The temperature-pressure cross-term coefficient f1 is usually in the range of -0.001~0.001Ω·m / (℃·GPa). For the relationship between permeability, temperature and pressure, the constant term a2 is in the range of 10 -20 ~10 -12 m 2 For the linear regression model, the coefficient b2 of temperature T is usually in the range of 0.001~0.01 m 2 / ℃, the coefficient c2 of pressure p is usually in the range of -0.1~0.1 m 2 / GPa; For nonlinear regression models, the first-order coefficient b2 of temperature T is usually in the range of 0.001~0.01 m 2 / ℃, the quadratic coefficient c2 of temperature T is usually in the range of 10 -6 ~10 -4 m 2 / ℃ 2 The first-order coefficient d2 of pressure p is usually in the range of -0.1~0.1 m 2 / GPa, the quadratic coefficient e2 of pressure p is usually in the range of -0.01~0.01 m 2 / GPa 2 The temperature-pressure cross-term coefficient f2 is usually in the range of -0.001~0.001 m 2 / (℃·GPa). For the relationship between porosity, temperature and pressure, the constant term a3 ranges from 0.01 to 0.4. For the linear regression model, the coefficient b3 of temperature T is usually in the range of 10 -5 ~10 -3 ℃ -1 , the coefficient c3 of pressure p is usually in the range of -10 -4 ~10 -4 MPa -1 ; For nonlinear regression models, the coefficient b3 of the temperature T is usually in the range of 10 -5 ~10 -3 ℃ -1 The quadratic coefficient c3 of temperature T is usually in the range of 10 -8 ~10 -6 ℃ -2 The first-order coefficient d3 of pressure p is usually in the range of -10 -4 ~10 -4 MPa -1 The quadratic coefficient e3 of pressure p is usually in the range of -10 -7 ~10 -7MPa -2 The temperature-pressure cross-term coefficient f3 is usually in the range of -10 -6 ~10 -6 (℃·MPa) -1 .

[0062] In step 3, the permeability, porosity and rock density of different rock types in the room are measured, and the rock density model under different permeabilities and porosities is established, specifically: Measure the permeability and porosity of different rock types at room temperature and pressure, and use a density meter to measure the rock density ρ in an indoor environment to determine the linear relationship between the permeability and porosity of different rock types at room temperature and pressure and density, and use a multiple linear regression model or a nonlinear regression model to establish a relationship model between the density and permeability and porosity of different rock types, that is, ; or ; Where a4, b4, c4, d4, e4 and f4 are coefficients. The constants a4 and a5 are in the range of 2.0~3.0 g / cm 3 For the linear regression model, the coefficient b4 of the permeability k is usually in the range of -10 -4 ~10 -4 g / cm 3 ·m 2 , the coefficient c4 of porosity φ is usually in the range of -0.1~0.1; for nonlinear regression models, the coefficient b5 of the first-order term of permeability k is usually in the range of -10 -4 ~10 -4 g / cm 3 ·m 2 , the coefficient c5 of the quadratic term of permeability k is usually in the range of -10 -8 ~10 -8 g / cm 3 ·m 4 The porosity φ first-order coefficient d5 is usually in the range of -0.1~0.1, the porosity φ second-order coefficient e5 is usually in the range of -0.01~0.01, and the permeability-porosity cross-term coefficient f5 is usually in the range of -10 -5 ~10 -5 g / cm 3 ·m 2 .

[0063] In step 4, the permeability and porosity models at different temperatures and pressures are introduced into the rock density models at different permeabilities and porosities to obtain the rock density models at different temperatures and pressures, specifically: The permeability and porosity models under different temperatures and pressures are introduced into the rock density models under different permeabilities and porosities to obtain the rock density models under different temperature (T) and pressure (p), that is, ; or .

[0064] In step 5, the magnetic susceptibility of the rock sample is obtained based on a conventional magnetic susceptibility meter, specifically, Since magnetic susceptibility is only affected by extreme temperature conditions, a conventional magnetic susceptibility meter was used in an indoor environment to measure the magnetic susceptibility (χ) of different rock types.

[0065] In step 6, samples with known lithology are used as training sets, and the obtained data are subjected to probability statistics using Bayesian classification to determine the lithology of unknown samples, specifically:

[0066] Based on the resistivity, density and magnetic susceptibility data of different rock types under different temperature and pressure conditions, the deep rock mass lithology identification is carried out to determine the sample lithology that cannot accurately identify the rock type in the sample, and provide a reference for the geological interpretation of the subsequent geophysical inversion results. Bayesian classification is used to take the rock physical parameters under variable temperature and pressure conditions as training and test samples for probability statistics. Among them, the rock physical parameters include density, resistivity, and magnetic susceptibility. The rock type is represented as c event, and the rock physical parameter is represented as x event. The rock type c and the corresponding attribute x are two events occurring in the same space. It is assumed that the complete rock type c is composed of multiple rock types c1, c2, ..., c N If x occurs, then a type in c must occur. According to the probability, we can get the known categories c of the sample set. j The prior probability and various conditional probabilities For unknown samples, the Bayesian formula can calculate the probability of the type to which the sample to be tested belongs, that is, the posterior probability, which is:

[0067] The probability of a certain rock type given a rock property parameter x is: ; In the formula, x represents the rock property, c j (j=1, 2,···, N) are N different rock types, P(x / c j ) is assumed to be rock type c j The probability of observing x when j ) indicates that before observing x, a specific type c j The prior probability of P(x) represents the probability of observing x from all samples, that is: ; For a given point x, P(x) is determined by Indicates that the unknown sample x belongs to c j The probability density of the class, then P(x / c j ) can be expressed as the probability density , where a represents a very small interval near x, the posterior probability is: .

[0068] In step 7, the rock type and the physical property data after normalization are used to discriminate the deep lithology by using the discriminant analysis method, and a deep lithology identification equation considering temperature and pressure conditions is established, specifically, After conducting probability statistics on the physical parameters of rocks (ores) under variable temperature and pressure conditions, the lithology, density, magnetic susceptibility and resistivity of all samples were determined, and data standardization was carried out. The deep lithology was discriminated based on the discriminant analysis method. According to the rock type and the corresponding physical parameters (density, magnetic susceptibility and resistivity), the discriminant function was obtained, which is in the form of: ; In the formula, y is the discriminant value, is the standardized deep rock physical parameter, is the discrimination coefficient of each physical property parameter, where a0 is a constant term with a value range of -10~10; the density coefficient a1 has a value range of -10~10 (g / cm 3 ) -1 ; The magnetic susceptibility coefficient a2 ranges from -10 3 ~10 3 (SI) -1 ; The resistivity coefficient a3 ranges from -10 3 ~10 -3 (Ω·m) -1 By substituting the physical property parameter values ​​into the above formula, the discriminant function values ​​corresponding to each type of rock can be obtained, and the type corresponding to the maximum function value is the rock type, thereby establishing a deep lithology identification equation that takes temperature and pressure conditions into consideration, systematically summarizing the geophysical response characteristics of key geological bodies, and providing constraints for the joint inversion of multiple geophysical data.

[0069] In step 8, based on the linear regression model, the resistivity, density and magnetic susceptibility data under variable temperature and pressure conditions are processed to obtain the physical property prior information of resistivity and density, resistivity and magnetic susceptibility, specifically:

[0070] Based on the resistivity, density and magnetic susceptibility of all rock samples obtained, a linear regression model is used to form the physical property prior information of resistivity and density, resistivity and magnetic susceptibility, namely: ; ; In the formula, are the coefficients of each parameter. For rock density ( ) and resistivity (R), the constant term h1 ranges from 2.0 to 3.0 g / cm 3 , resistivity coefficient The value range is -10 -3 ~10 -3 g / cm 3 / (Ω·m); for rock magnetic susceptibility ( ) and resistivity (R), the constant term h2 ranges from 10 -5 ~10 -1 SI, resistivity coefficient The value range is -10 -6 ~10 -6 SI / (Ω·m).

[0071] In step 9, the electromagnetic data is used for three-dimensional inversion to obtain the resistivity model, specifically:

[0072] Using electromagnetic data (EM), the objective function of the joint inversion considering Gramian constraints is minimized. Using electromagnetic data for three-dimensional inversion, the resistivity model is obtained. The objective function of the resistivity model is: ; In the formula, is the model parameter vector, is the weighting coefficient of gravity, magnetic and electromagnetic data, α is the regularization factor, express , is the observed data covariance matrix, Stabilizing functionals for the model , is the model weight matrix, is a physical property constraint term. If Gramian constraint is selected, it can be expressed as ; In the formula, i is the weight coefficient of the i-th Gramian constraint term (usually in the range of 10 -3 Up to 10 3 The value depends on the complexity of the inversion problem and the noise level of the data. However, due to the influence of the complex underground structure and mining work in the mineralization area, the weight coefficient ranges from 10 to 10 3 )、i=1,2,3, is the Gramian constraint coupling term, is the weight coefficient of the i-th penalty function term (usually in the range of 10 -3 Up to 10 -2 , in the inversion process, it is adjusted according to the changes in fitting residuals and model complexity), i=1, 2, 3, and is also fixed in the inversion process, It is a constraint item of penalty function. When it exceeds the limit range, it acts as a constraint.

[0073] In step 10, the resistivity model is used as a common reference model and converted into an initial model for joint inversion of gravity and magnetic methods based on prior information of physical properties, specifically:

[0074] The resistivity model is used as a common reference model to integrate the prior information of physical properties into the joint geophysical inversion. The inversion results of electromagnetic data are converted through the model and used as the common reference model for the joint inversion of gravity and magnetic data. ; ; ; In the formula, and represent the a priori relationships among density, magnetic susceptibility and resistivity, respectively.

[0075] In step 11, the electromagnetic inversion results are integrated into the wide range constraints as a reference model, and the three-dimensional joint inversion objective function minimization of the gravity and magnetic methods considering the Gramian constraints is carried out respectively, specifically:

[0076] The electromagnetic inversion results are integrated into the wide range constraints as a reference model, and the three-dimensional joint inversion objective function minimization of the gravity and magnetic methods considering Gramian constraints is carried out, namely: ; .

[0077] The in-situ geophysical characterization method for composite mineralization terminal effect constrained by rock physical properties under variable temperature and pressure conditions proposed by the present invention has generally broken through the bottleneck that the parameters constrained by rock physical properties in previous joint inversion studies cannot represent the real physical properties of underground media and have strict requirements on prior information. At the same time, it avoids the defects of the traditional single characterization system of composite mineralization terminal effect geochemistry or integrated geophysics. Finally, the present invention can obtain the underground structure and medium properties of the composite mineralization area and accurately characterize the composite mineralization terminal effect. The above are advantages that previous methods do not have.

[0078] The specific effects of the present invention are as follows: (1) The present invention uses a test system to simulate variable temperature and pressure conditions at different depths to carry out rock physical property testing, which can effectively restore the temperature and pressure environment at the real depth, extract the real physical property characteristics of the underground medium, and provide accurate constraints for geophysical information fusion; (2) The present invention has formed a set of "real rock physical property constraints-deep lithology identification-multi-source geophysical coupling-composite mineralization law" technology. In terms of lithology identification and geophysical information fusion, the Bayesian classification and discriminant analysis method is used to avoid the influence of subjective judgment on the lithology of unknown samples, and the joint inversion scheme considering Gramian constraints is used to integrate multiple geophysical information to a great extent; (3) The present invention studies the composite mineralization area from two aspects: deep rock properties and the structural composition of the inversion results, which can comprehensively characterize the enrichment conditions and laws of mineralizing materials, and then determine the enrichment location of the ore body.

[0079] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0080] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A geophysical in-situ characterization method for composite mineralization end-effect constrained by rock properties, characterized in that: include: Step 1: Based on the rock physical property comprehensive testing system, the temperature and pressure at different depths are simulated to measure the resistivity, permeability and porosity of rock samples of key geological units in the composite mineralization area; Step 2: Determine the linear relationship between temperature, pressure and the resistivity, permeability and porosity of rock samples, and use a multivariate regression model to establish resistivity, permeability and porosity models at different temperatures and pressures; Step 3: Measure the permeability, porosity and rock density of different rock types indoors, and establish rock density models under different permeabilities and porosities; Step 4: Substitute the permeability and porosity models under different temperatures and pressures into the rock density models under different permeabilities and porosities to obtain the rock density models under different temperatures and pressures; Step 5: Based on a conventional magnetic susceptibility meter, obtain the magnetic susceptibility of the rock sample; Step 6: Use samples with known lithology as training sets, use Bayesian classification to perform probability statistics on the obtained data, and determine the lithology of unknown samples; Step 7: Based on the rock type and the physical property data after standardization, the deep lithology is discriminated by discriminant analysis, and a deep lithology identification equation considering temperature and pressure conditions is established; Step 8: Based on the linear regression model, the resistivity, density and magnetic susceptibility data under variable temperature and pressure conditions are processed to obtain the prior information of the physical properties of resistivity and density, resistivity and magnetic susceptibility; Step 9: Perform 3D inversion using electromagnetic data to obtain a resistivity model; Step 10: Using the resistivity model as a common reference model, based on prior information on physical properties, it is converted into an initial model for joint inversion of gravity and magnetic methods; Step 11: Integrate the electromagnetic inversion results into the wide range constraints as a reference model, and carry out the minimization of the objective function of the three-dimensional joint inversion of gravity and magnetic methods considering Gramian constraints.

2. The method according to claim 1, characterized in that: In step 2, the linear relationship between temperature, pressure and the resistivity, permeability and porosity of the rock sample is determined, and a multivariate regression model is used to establish the resistivity, permeability and porosity models at different temperatures and pressures, specifically: According to different rock types, the influence of temperature T and pressure p at different depths on rock resistivity R, permeability k and porosity φ is statistically analyzed. For the case of linear relationship, the multivariate linear regression model is used to establish the relationship between resistivity, permeability and porosity of different rock types and temperature and pressure, that is: ； ； ； In the case of nonlinear relationships, a nonlinear regression model is used, which is expressed as: ； ； ； In the formula, a i 、b i 、c i d i 、e i and f i is the coefficient, i=1, 2, 3,…, n.

3. The method according to claim 2, characterized in that In step 3, the permeability, porosity and rock density of different rock types in the room are measured, and the rock density model under different permeabilities and porosities is established, specifically: The permeability and porosity of different rock types at room temperature and pressure are measured, and the rock density ρ is measured using a density meter in an indoor environment to determine the linear relationship between the permeability and porosity of different rock types at room temperature and pressure and the density. A multivariate linear regression model or a nonlinear regression model is used to establish a relationship model between the density and permeability and porosity of different rock types, namely: ； or, .

4. The method according to claim 3, characterized in that In step 4, the permeability and porosity models at different temperatures and pressures are introduced into the rock density models at different permeabilities and porosities to obtain the rock density models at different temperatures and pressures, specifically: The permeability and porosity models under different temperatures and pressures are introduced into the rock density models under different permeabilities and porosities to obtain the rock density models under different temperatures and pressures, namely: ； or, 。 5. The method according to claim 4, characterized in that In step 6, samples with known lithology are used as training sets, and the obtained data are subjected to probability statistics using Bayesian classification to determine the lithology of unknown samples, specifically: According to the obtained resistivity, density and magnetic susceptibility data of different rock types under different temperature and pressure conditions, the lithology identification of deep rock mass is carried out. Bayesian classification is used to take the rock physical property parameters under variable temperature and pressure conditions as training and test samples for probability statistics. Among them, the rock physical property parameters include density, resistivity and magnetic susceptibility. The rock type is represented as c event and the rock physical property parameter is represented as x event. The rock type c and the corresponding attribute x are two events occurring in the same space. It is assumed that the complete rock type c is composed of multiple rock types c1, c2, ..., c N If x occurs, then a type in c must occur. According to the probability, we can get the known categories c of the sample set. j The prior probability and various conditional probabilities For unknown samples, the Bayesian formula can calculate the probability of the type to which the sample to be tested belongs, that is, the posterior probability, which is: The probability of a certain rock type given a rock property parameter x is: ； In the formula, x represents the rock property, c j (j=1, 2,···, N) are N different rock types, P(x / c j ) is assumed to be rock type c j The probability of observing x when j ) indicates that before observing x, a specific type c j The prior probability of P(x) represents the probability of observing x from all samples, that is: ； For a given point x, P(x) is determined by Indicates that the unknown sample x belongs to c j The probability density of the class, then P(x / c j ) can be expressed as the probability density , where a represents a very small interval near x, the posterior probability is: 。 6. The method according to claim 5, characterized in that In step 7, the rock type and the physical property data after normalization are used to discriminate the deep lithology by using the discriminant analysis method, and a deep lithology identification equation considering temperature and pressure conditions is established, which is specifically: After probabilistic statistics of rock physical parameters under variable temperature and pressure conditions, the lithology, density, magnetic susceptibility and resistivity of all samples were determined, and data standardization was carried out. The deep lithology was discriminated based on the discriminant analysis method. According to the rock type and the corresponding physical parameters, the discriminant function was obtained, which is in the form of: ； In the formula, y is the discriminant value, is the standardized deep rock physical parameter, is the discriminant coefficient of each physical property parameter. By substituting the physical property parameter value into the above formula, the discriminant function value corresponding to each type of rock can be obtained, and the type corresponding to the maximum function value is the rock type, thereby establishing a deep lithology identification equation considering temperature and pressure conditions, systematically summarizing the geophysical response characteristics of key geological bodies, and providing constraints for the joint inversion of multiple geophysical data.

7. The method according to claim 6, characterized in that In step 8, based on the linear regression model, the resistivity, density and magnetic susceptibility data under variable temperature and pressure conditions are processed to obtain the physical property prior information of resistivity and density, resistivity and magnetic susceptibility, specifically: Based on the resistivity, density and magnetic susceptibility of all rock samples obtained, a linear regression model is used to form the physical property prior information of resistivity and density, resistivity and magnetic susceptibility, namely: ； ； In the formula, are the coefficients of each parameter.

8. The method according to claim 7, characterized in that In step 9, the electromagnetic data is used for three-dimensional inversion to obtain the resistivity model, specifically: The three-dimensional inversion is performed using electromagnetic data to obtain the resistivity model. The objective function of the resistivity model is: ； In the formula, is the model parameter vector, is the weighting coefficient of gravity, magnetic and electromagnetic data, α is the regularization factor, express , is the observed data covariance matrix, Stabilizing functionals for the model , is the model weight matrix, is a physical constraint term. If Gramian constraint is selected, it can be expressed as: ； In the formula, i is the weight coefficient of the i-th Gramian constraint term, i=1, 2, 3, is the Gramian constraint coupling term, is the weight coefficient of the i-th penalty function term, i=1, 2, 3, and is also constant during the inversion process. It is a constraint item of the penalty function. When it exceeds the limit range, it acts as a constraint.

9. The method according to claim 8, characterized in that In step 10, the resistivity model is used as a common reference model and converted into an initial model for joint inversion of gravity and magnetic methods based on prior information of physical properties, specifically: The resistivity model is used as a common reference model to integrate the prior information of physical properties into the joint geophysical inversion. The inversion results of electromagnetic data are converted through the model and used as the common reference model for the joint inversion of gravity and magnetic data. ； ； ； In the formula, and represent the a priori relationships among density, magnetic susceptibility and resistivity, respectively.

10. The method according to claim 9, characterized in that In step 11, the electromagnetic inversion results are integrated into the wide range constraints as a reference model, and the three-dimensional joint inversion objective function minimization of the gravity and magnetic methods considering the Gramian constraints is carried out respectively, specifically: The electromagnetic inversion results are integrated into the wide range constraints as a reference model, and the three-dimensional joint inversion objective function minimization of the gravity and magnetic methods considering Gramian constraints is carried out, namely: ； 。

Images