Comprehensive judgment method for reliability of different in-situ stress measurement methods in the same area

Abstract

The embodiment of the present application relates to the field of in-situ stress measurement, and discloses a comprehensive determination method for the reliability of different in-situ stress measurement methods in the same area. The comprehensive determination method includes: obtaining first in-situ stress measurement data based on the first in-situ stress measurement method; obtaining second in-situ stress measurement data based on the second in-situ stress measurement method in the same area; using the Bayesian linear regression method to model and align the data of the first in-situ stress measurement data; using the Bayesian linear regression method to model and align the data of the second in-situ stress measurement data; drawing a first probability density map based on the first in-situ stress measurement data after data alignment, drawing a second probability density map based on the second in-situ stress measurement data after data alignment, and comparing the similarity between the first probability density map and the second probability density map. The comprehensive determination method uses Bayesian linear regression to realize the reliability determination of the test results obtained by using multiple different in-situ stress measurement methods in the same area.

Description

Technical Field

[0001] This application relates to the technical field of in-situ stress measurement, and more specifically, to a comprehensive determination method for the reliability of different in-situ stress measurement methods in the same area. Background Art

[0002] Understanding the in-situ stress field acting on the earth's crust and lithosphere is very important for rock mechanics, geophysical research and applications, including rock mass engineering design and construction, plate tectonics, seismicity and fault studies, and underground fluid behavior. However, in-situ stress is a quantity that is difficult to measure because there are various factors affecting the source of in-situ stress. Theoretically, in-situ stress is a property of a point in the earth's crust, and the in-situ stress state of a point is usually characterized by the magnitudes and directions of three principal stresses. Due to the special nature of in-situ stress, at least two or more in-situ stress measurement methods should be used to gradually estimate in-situ stress. Combining various measurement methods and complementing each other according to their respective attributes helps to obtain a more reliable estimate of in-situ stress. However, the results of in-situ stress measurement are usually affected by many factors, and most of these factors are variable. Moreover, due to engineering conditions, the selection and layout of measurement points have significant limitations. In addition, geological conditions are often quite complex, and the in-situ stress states at each point have strong randomness. Therefore, it is very imperfect to use traditional simple statistical methods to compare the distribution characteristics of the magnitudes and directions of in-situ stress determined by different in-situ stress measurement methods based on the measurement data of only a few points. At present, there is no good quantitative comparison and analysis method for the reliability of the measurement results of different in-situ stress measurement methods and the quality of the obtained in-situ stress data.

[0003] Only when the same area has the same geological structure background and geodynamic environment can the obtained in-situ stress data be comparable. Therefore, in view of the fact that it is possible to use multiple different in-situ stress measurement methods in the same area and it is difficult to determine the reliability, it is necessary to study the determination of the reliability of different in-situ stress measurement methods in the same area. Summary of the Invention

[0004] In view of this, the embodiments of this application are committed to providing a comprehensive determination method for the reliability of different in-situ stress measurement methods in the same area to solve the problem that it is difficult to determine the reliability of multiple different in-situ stress measurement methods used in the same area in the prior art.

[0005] In a first aspect, this specification provides a comprehensive determination method for the reliability of different in-situ stress measurement methods in the same area, including:

[0006] Obtaining first in-situ stress measurement data based on a first in-situ stress measurement method;

[0007] In the same area, obtaining second in-situ stress measurement data based on a second in-situ stress measurement method;

[0008] Model and align the first in-situ stress measurement data using the Bayesian linear regression method;

[0009] Model and align the second in-situ stress measurement data using the Bayesian linear regression method;

[0010] Draw a first probability density plot based on the first in-situ stress measurement data after data alignment, draw a second probability density plot based on the second in-situ stress measurement data after data alignment, and compare the similarity between the first probability density plot and the second probability density plot;

[0011] Wherein, the second in-situ stress measurement method includes at least one in-situ stress measurement method other than the first in-situ stress measurement method.

[0012] According to the first aspect, in a possible implementation manner, the comprehensive determination method further includes:

[0013] Assume that the first in-situ stress measurement data and the second in-situ stress measurement data come from the same distribution;

[0014] Calculate the first cumulative distribution function of the first in-situ stress measurement data and the second cumulative distribution function of the second in-situ stress measurement data;

[0015] Calculate the Kolmogorov-Smirnov statistic, and the Kolmogorov-Smirnov statistic represents the maximum vertical distance between the first cumulative distribution function and the second cumulative distribution function;

[0016] According to the selected significance level, look up the corresponding Kolmogorov-Smirnov critical value;

[0017] Judge whether the first in-situ stress measurement data and the second in-situ stress measurement data follow the same distribution.

[0018] According to the first aspect, in a possible implementation manner, the comprehensive determination method further includes:

[0019] Based on the first in-situ stress measurement data, establish a principal stress distribution model along a fixed direction using a mathematical method;

[0020] Judge the degree of fit between the second in-situ stress measurement data and the principal stress distribution model. If the degree of fit is high, the reliability of the first in-situ stress measurement method is high, otherwise the reliability is low;

[0021] Wherein, the principal stress distribution model represents the corresponding relationship between the principal stress along a fixed direction in the first in-situ stress measurement data and the depth, and the degree of fit is characterized by the corresponding parameters between the second in-situ stress measurement data and the principal stress distribution model.

[0022] According to the first aspect, in a possible implementation, the degree of fit is characterized by the correlation coefficient between the second in-situ stress measurement data and the principal stress distribution model. If the correlation coefficient is higher than the fit threshold, the degree of fit is high.

[0023] According to the first aspect, in a possible implementation, the comprehensive determination method further includes:

[0024] At the same or approximate depth level, calculate the principal stress difference rate f between the principal stress in the first in-situ stress measurement data and the principal stress in the same direction in the second in-situ stress measurement data according to the following formula. If the principal stress difference rate is not greater than the principal stress difference threshold, the reliability of the first in-situ stress measurement method is high; otherwise, the reliability is low.

[0025]

[0026] where, σ i is the maximum horizontal principal stress, minimum horizontal principal stress or vertical principal stress in the first in-situ stress measurement data, and σ j is the corresponding maximum horizontal principal stress, minimum horizontal principal stress or vertical principal stress in the second in-situ stress measurement data.

[0027] According to the first aspect, in a possible implementation, the comprehensive determination method further includes:

[0028] Determine the first stress structure of the first in-situ stress measurement data;

[0029] Determine the second stress structure of the second in-situ stress measurement data;

[0030] Compare whether the first dominant stress structure of the first in-situ stress measurement data and the second dominant stress structure of the second in-situ stress measurement data are consistent. If they are consistent, the reliability of the first in-situ stress measurement method is high; otherwise, the reliability is low.

[0031] where, the first dominant stress structure and the second dominant stress structure respectively represent the stress structures reflected by the measurement data with the most conforming to a certain stress structure in the first stress structure and the second stress structure.

[0032] According to the first aspect, in a possible implementation, the comprehensive determination method further includes:

[0033] Calculate the first ratio of the average horizontal principal stress to the vertical principal stress of the first in-situ stress measurement data;

[0034] Calculate the second ratio of the average horizontal principal stress to the vertical principal stress of the second in-situ stress measurement data;

[0035] Compare whether the variation laws of the first ratio and the second ratio with depth are consistent. If they are consistent, the reliability of the first in-situ stress measurement method is high; otherwise, it is low.

[0036] According to the first aspect, in a possible implementation, the comprehensive determination method further includes:

[0037] Perform non-linear fitting on the first ratio and the second ratio with depth respectively to generate a first fitting curve and a second fitting curve;

[0038] Compare the fitting curves of the first ratio and the second ratio with the Brown-Hoek inner and outer envelopes. If the first fitting curve and the second fitting curve are inside the Brown-Hoek inner and outer envelopes, the reliability of the first in-situ stress measurement method is high; otherwise, it is low.

[0039] According to the first aspect, in a possible implementation, the comprehensive determination method further includes:

[0040] Draw a first rose diagram based on the direction of the maximum horizontal principal stress of the first in-situ stress measurement data;

[0041] Draw a second rose diagram based on the direction of the maximum horizontal principal stress of the second in-situ stress measurement data;

[0042] If the first dominant direction of the first rose diagram and the second dominant direction of the first rose diagram are consistent, the reliability of the first in-situ stress measurement method is high; otherwise, it is low;

[0043] Wherein, the first dominant direction and the second dominant direction respectively represent the main acting directions of the maximum horizontal principal stress of the first rose diagram and the maximum horizontal principal stress of the second rose diagram.

[0044] According to the first aspect, in a possible implementation, the comprehensive determination method further includes: If the included angle between the first dominant direction and the second dominant direction is less than the included angle threshold, the reliability of the first in-situ stress measurement method is high; otherwise, it is low.

[0045] In a second aspect, this specification provides a comprehensive determination system for the reliability of different in-situ stress measurement methods in the same area, including:

[0046] A first in-situ stress acquisition module, configured to obtain first in-situ stress measurement data based on a first in-situ stress measurement method;

[0047] A second in-situ stress acquisition module, configured to obtain second in-situ stress measurement data in the same area based on a second in-situ stress measurement method;

[0048] A reliability judgment module, configured to:

[0049] Model and align the first in-situ stress measurement data using the Bayesian linear regression method;

[0050] Model and align the second in-situ stress measurement data using the Bayesian linear regression method;

[0051] Draw the first probability density plot based on the first in-situ stress measurement data after data alignment, draw the second probability density plot based on the second in-situ stress measurement data after data alignment, and compare the similarity between the first probability density plot and the second probability density plot;

[0052] Wherein, the second in-situ stress measurement method includes at least one in-situ stress measurement method other than the first in-situ stress measurement method.

[0053] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0054] In multiple embodiments provided in this specification, the comprehensive determination method obtains the first in-situ stress measurement data based on the first in-situ stress measurement method; in the same area, obtains the second in-situ stress measurement data based on the second in-situ stress measurement method; models and aligns the first in-situ stress measurement data using the Bayesian linear regression method; models and aligns the second in-situ stress measurement data using the Bayesian linear regression method; draws the first probability density plot based on the first in-situ stress measurement data after data alignment, draws the second probability density plot based on the second in-situ stress measurement data after data alignment, and compares the similarity between the first probability density plot and the second probability density plot. In this way, the comprehensive determination method realizes the comprehensive determination of the reliability of different in-situ stress measurement methods in the same area, which helps to further improve the measurement accuracy and reliability of the in-situ stress measurement method. Brief Description of the Drawings

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0056] Figure 1 It is a schematic flowchart of the comprehensive determination method for the reliability of different in-situ stress measurement methods in the same area provided for at least one embodiment of this specification;

[0057] Figure 2 It is a schematic diagram of the improved Kolmogorov-Smirnov statistical test process for different in-situ stress measurement methods provided for at least one embodiment of this specification;

[0058] Figure 3(a) - Figure 3(c)Probability density plots of the maximum horizontal principal stress, minimum horizontal principal stress, and vertical principal stress measured in a mining area using the stress relief method and hydraulic fracturing method provided for at least one embodiment of this specification;

[0059] Figure 4 Schematic diagram comparing the p-values of the Kolmogorov-Smirnov test for the three principal stresses provided for at least one embodiment of this specification with the selected significance level;

[0060] Figure 5 Trend of change of the three principal stresses with depth and their fitting equations (R 2 is the correlation coefficient of the fitting equation) provided for at least one embodiment of this specification;

[0061] Figure 6 K calculated from the in-situ stress data obtained by the stress relief method and hydraulic fracturing method provided for at least one embodiment of this specification av value and its comparison with the Brown-Hoek inner and outer envelopes;

[0062] Figure 7(a) - Figure 7(b) Rose diagrams of the maximum horizontal principal stress direction obtained by the stress relief method and rose diagrams of the maximum horizontal principal stress direction obtained by the hydraulic fracturing method provided for at least one embodiment of this specification respectively;

[0063] Figure 8 Schematic diagram of the structure of a comprehensive determination system 80 for the reliability of different in-situ stress measurement methods in the same area provided for at least one embodiment of this specification. Specific Embodiments

[0064] Unless otherwise defined, technical terms or scientific terms used in the embodiments of this specification should have the ordinary meaning understood by those of ordinary skill in the field to which this specification belongs. The "first", "second" and similar terms used in the embodiments of this specification do not denote any order, quantity or importance, but are only used to avoid confusion of components.

[0065] Unless otherwise required by the context, throughout the specification, "a plurality" means "at least two", and "including" is interpreted in an open, inclusive sense, that is, "including, but not limited to". In the description of the specification, the terms "one embodiment", "some embodiments", "exemplary embodiments", "examples", "specific examples" or "some examples" etc. are intended to indicate that specific features, structures, materials or characteristics related to the embodiment or example are included in at least one embodiment or example of this specification. The schematic representations of the above terms do not necessarily refer to the same embodiment or example.

[0066] The technical solutions in the embodiments of the present specification will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present specification. Obviously, the described embodiments are only a part of the embodiments of the present specification, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present specification without creative efforts shall fall within the protection scope of the present specification.

[0067] Exemplary method

[0068] In-situ stress measurement can be applied to multiple fields, including rock mass engineering design and construction, plate tectonics, seismicity and fault studies, and underground fluid behavior, etc. Therefore, it is very important to determine the reliability of in-situ stress measurement methods.

[0069] In order to rigorously and objectively determine whether different types of in-situ stress measurement data in the same area are consistent and reliable, and to solve the problems that traditional simple statistical methods cannot systematically quantitatively compare and the rigor and objectivity of comparison results are poor, a comprehensive determination method for the reliability of different in-situ stress measurement methods in the same area is provided in at least one embodiment of the present specification. As Figure 1 shown, the comprehensive determination method may specifically include the following steps.

[0070] Step S110: Obtain first in-situ stress measurement data based on the first in-situ stress measurement method.

[0071] The first in-situ stress measurement method can be any one of the existing in-situ stress measurement methods, and generally, an in-situ stress measurement method suitable for the location to be measured is selected.

[0072] Step S120: In the same area, obtain second in-situ stress measurement data based on the second in-situ stress measurement method.

[0073] Similarly, based on the second in-situ stress measurement method, second in-situ stress measurement data can be obtained. The second in-situ stress measurement method includes at least one in-situ stress measurement method other than the first in-situ stress measurement method.

[0074] Step S130: Use the Bayesian linear regression method to model and align the first in-situ stress measurement data.

[0075] Although different in-situ stress measurement methods are used to measure in-situ stress in the same area (with the same geological structure background), they generally do not measure at the same depth. Therefore, strictly speaking, it is impossible to directly compare and evaluate their measurement results. Therefore, in order to more intuitively verify whether the stress data measured by different in-situ stress measurement methods follow the same distribution, it is first necessary to align different types of in-situ stress data. Considering the inevitable errors and other uncertainties that occur during in-situ stress measurement, a new Bayesian linear regression method is proposed for modeling and data alignment. This new method is more flexible and comprehensive in statistical modeling problems, and innovatively provides a more stable evaluation in dealing with small sample problems by introducing prior information.

[0076] Step S140: Use the Bayesian linear regression method to model and align the second in-situ stress measurement data.

[0077] Similarly, the second in-situ stress measurement data can be modeled and aligned.

[0078] Step S150: Draw a first probability density plot based on the first in-situ stress measurement data after data alignment, draw a second probability density plot based on the second in-situ stress measurement data after data alignment, and compare the similarity between the first probability density plot and the second probability density plot.

[0079] After data completion, the probability density plot is used to determine whether the in-situ stress data obtained by different in-situ stress measurement methods follow the same distribution, thereby verifying the consistency and effectiveness of the data obtained by different types of in-situ stress measurement methods.

[0080] Through the above steps, by using the Bayesian linear regression method to judge whether different types of in-situ stress data follow the same distribution, the reliability of different in-situ stress measurement methods in the same area is determined, which helps to further improve the measurement accuracy and reliability of in-situ stress measurement methods.

[0081] Specifically, for a given data set Q = {X, Y}, where X = (x1, x2,..., x N ) T , Y = (y1, y2,..., y N ) T , respectively representing the data obtained by two in-situ stress measurement methods. The basic model of linear regression can be expressed as:

[0082] y = f(x) + ε = ω T x + ε (1)

[0083] In the formula: ε ~ N(0, σ 2 ).

[0084] In this Bayesian linear regression, ω and y are regarded as unknown random variables, and then inference and prediction are carried out step by step. In the inference stage, the distribution that the parameter ω follows is derived based on the Bayesian formula; in the prediction stage, based on the distribution of the parameter ω obtained by inference, the target distribution y is predicted.

[0085] In the inference stage, according to the Bayesian formula, we have:

[0086]

[0087] According to Equation (2), we can obtain:

[0088] P(ω|X,Y) ∝ P(Y|ω,X)P(ω) (3)

[0089] Among them, the likelihood part is:

[0090]

[0091] The prior part is assumed to be:

[0092] P(ω) = N(0, ∑ p ) (5)

[0093] According to the self-conjugacy of the Gaussian distribution, the posterior distribution can be obtained as:

[0094] P(ω|X,Y) = N(μ ω , ∑ ω ) (6)

[0095] After substituting and using the method of completing the square, we can obtain:

[0096]

[0097] In the formula:

[0098] In the prediction stage, the linear regression model is:

[0099] y* = f(x*) + ε (8)

[0100] According to the properties of the Gaussian distribution, we can obtain:

[0101] y * ~N(x *T μ ω , x *T ∑ ω x * + σ 2 I) (9)

[0102] After data alignment, a probability density plot is used to determine whether the in-situ stress data obtained by different in-situ stress measurement methods (such as the stress relief method) follow the same distribution, thereby verifying the consistency and validity of different types of in-situ stress data. The probability density function is a function that describes the probability distribution of a random variable. For a continuous random variable, the value of the probability density function at a certain point does not represent the probability, but the probability density. The probability density function must satisfy the non-negativity and normalization conditions.

[0103] For a continuous random variable E, its probability density function is p(e), where e ∈ (-∞, +∞). For any real number e, we have:

[0104]

[0105] According to the above method, first, the stress data obtained by one of the in-situ stress measurement methods (such as the stress relief method) is modeled, and the depth information in the stress data obtained based on other in-situ stress measurement methods (such as the hydraulic fracturing method, the acoustic emission method, etc.) is predicted. At the same time, the stress data measured by the hydraulic fracturing method, the acoustic emission method, etc. is modeled, and the depth information in the measurement data based on the stress relief method is predicted. After completing the above steps, data alignment is achieved. Then, probability density plots are respectively drawn for the stress data measured by different in-situ stress measurement methods based on the true values and predicted values corresponding to the depth information in one of the in-situ stress measurement methods and the true values and predicted values corresponding to the depth information in other in-situ stress measurement methods. By comparing and analyzing the shapes of the probability density plots of the principal stresses obtained by different in-situ stress measurement methods and the expected values of the corresponding data, it is judged whether different types of in-situ stress data follow the same distribution.

[0106] To further ensure and verify the distribution of different types of in-situ stress data and the correctness of the obtained probability density plots, an improved Kolmogorov-Smirnov statistical test method is proposed and used to compare the similarity of the cumulative distributions of different types of in-situ stress data sets. The proposed test process is as Figure 2 shown. The specific test process is as follows:

[0107] ① Establish the null hypothesis: Assume that different types of in-situ stress data come from the same distribution.

[0108] ② Calculate the cumulative distribution function: For each category of data, calculate the cumulative distribution function value for each data point. For the sorted data points, calculate the cumulative frequency for each point. The cumulative distribution function describes the probability that a real-valued random variable is not greater than a specific value. Assuming X is the random variable, the cumulative distribution function is F(x) = P(X ≤ x), which is the integral of the probability density function and can completely describe the probability distribution of a real random variable X. Sorting ensures that the cumulative probability of data points can be calculated sequentially. For each additional data point, the probability increases cumulatively. If the data is not sorted, the cumulative probability of each value cannot be accurately calculated because the cumulative probability should increase monotonically as the value increases.

[0109] ③ Calculate the Kolmogorov-Smirnov statistic (D), which is the maximum vertical distance between any two cumulative distribution functions:

[0110] D = max|F1(x) - F2(x)| (11)

[0111] where F1(x) and F2(x) are the cumulative distribution functions of the two sets of data respectively.

[0112] ④ Find the critical value: According to the selected significance level (usually 0.05 or 0.01), find the corresponding Kolmogorov-Smirnov critical value. The significance level is a measure used to determine the criterion for rejecting the null hypothesis in hypothesis testing and can be considered as the maximum tolerance probability value when the actual null hypothesis holds but is wrongly rejected. Generally, 0.05 or 0.01 is taken. Statistical books usually provide a table of critical values for the Kolmogorov-Smirnov test, and the critical value can be found based on the significance level and the sample size.

[0113] ⑤ Judge the result: Use a self-written MATLAB program to perform the Kolmogorov-Smirnov test on the aligned different types of in-situ stress data. If the calculated Kolmogorov-Smirnov statistic (D) is greater than the critical value, reject the null hypothesis and consider that the different types of in-situ stress data do not come from the same distribution. If the calculated Kolmogorov-Smirnov statistic (D) is less than or equal to the critical value, accept the null hypothesis and consider that the different types of in-situ stress data come from the same distribution. Or, convert the maximum difference between the distributions of different types of in-situ stress data into a p-value according to Equation (12), and then compare the p-value with the selected significance level. If the p-value is less than the selected significance level, reject the null hypothesis and consider that these data do not follow the same distribution; otherwise, vice versa. A common method to calculate the p-value is based on the Kolmogorov-Smirnov statistic and the sample size, and Equation (12) is an approximate calculation expression.

[0114]

[0115] where: D is the Kolmogorov - Smirnov statistic, and n is the sample size.

[0116] To further enhance the determination of the reliability of the in - situ stress measurement method, the degree of fit between measurement methods is introduced for determination. Specifically, the degree - of - fit determination method includes:

[0117] Step S210: Based on the first in - situ stress measurement data, use a mathematical method to establish a principal stress distribution model along a fixed direction.

[0118] The mathematical method can be a regression method, the least - squares method, etc.; the fixed direction can be a preset direction, such as horizontal, vertical, etc. The established principal stress distribution model is shown in the following fitting equation (13).

[0119] σ = aH + b (13)

[0120] where: σ is the principal stress along the fixed direction, H is the depth, and a and b are undetermined coefficients.

[0121] Step S220: Judge the degree of fit between the second in - situ stress measurement data and the principal stress distribution model.

[0122] Among them, the principal stress distribution model characterizes the corresponding relationship between the principal stress along the fixed direction in the first in - situ stress measurement data and the depth, and the degree of fit is characterized by the corresponding parameters between the second in - situ stress measurement data and the principal stress distribution model.

[0123] If the degree of fit is high, the reliability of the first in - situ stress measurement method is high; otherwise, it is low. Specifically, the corresponding parameter between the second in - situ stress measurement data and the principal stress distribution model can be the correlation coefficient. If the correlation coefficient is higher than the fit threshold, the degree of fit is high. Among them, the fit threshold can be a value greater than 0 and less than 1 (such as 0.85, 0.9, 0.95, etc.), and the closer to 1, the higher the degree of fit.

[0124] Through the above steps, by determining the degree of fit between measurement methods, the determination of the reliability of different in - situ stress measurement methods in the same area is realized, which helps to further improve the measurement accuracy and reliability of the in - situ stress measurement method.

[0125] In order to compare the magnitudes of the principal stress values determined by different in-situ stress measurement methods at the same or approximate depth levels, a strict new judgment criterion (Equation (14)) is proposed, that is, when the ratio of the difference between the principal stresses determined by different in-situ stress measurement methods to the maximum principal stress among them (i.e., the principal stress difference rate) f is less than or equal to the principal stress difference threshold (e.g., 20%), it is considered that the difference in the principal stress values determined by different in-situ stress measurement methods is very small. The principal stress difference threshold can be any value less than or equal to 30% and greater than 0.

[0126]

[0127] In the formula: σ i is the magnitude of the maximum horizontal principal stress, minimum horizontal principal stress or vertical principal stress determined by a certain in-situ stress measurement method, and σ j is the magnitude of the maximum horizontal principal stress, minimum horizontal principal stress or vertical principal stress corresponding to σ i determined by another in-situ stress measurement method.

[0128] In order to more accurately determine the reliability of the in-situ stress measurement method through the stress structure, a stress structure determination method is provided in at least one embodiment of this specification, including: determining the first stress structure of the first in-situ stress measurement data; determining the second stress structure of the second in-situ stress measurement data; comparing whether the first dominant stress structure of the first in-situ stress measurement data and the second dominant stress structure of the second in-situ stress measurement data are consistent. If they are consistent, the reliability of the first in-situ stress measurement method is high, otherwise it is low; wherein, the first dominant stress structure and the second dominant stress structure respectively represent the stress structures reflected by the measurement data with the most conforming to a certain stress structure in the first stress structure and the second stress structure.

[0129] Specifically, compare the magnitudes among the maximum horizontal principal stress σ H , minimum horizontal principal stress σ h and vertical principal stress σ v obtained by each in-situ stress measurement method to determine the stress structure: when the stress structure is of the type σ H >σ h >σ v (reverse fault type), it indicates that the current stress state is conducive to the formation and movement of reverse thrust faults. When the stress structure is of the type σ H >σ v >σ h (strike-slip type), it indicates that the current stress state is conducive to the formation and movement of strike-slip faults. When the stress structure is of the type σ v >σ H >σ hIn the type (normal fault type), it indicates that the current stress state is conducive to the formation and movement of normal faults; compare and analyze whether the dominant stress structures determined by each in-situ stress measurement method are consistent. Among them, the dominant stress structure refers to the stress state in a certain area that is most conducive to the formation and movement of a certain type of fault.

[0130] Calculate the ratio K of the average horizontal principal stress to the vertical principal stress av (K av = (σ H + σ h ) / 2σ v ), K av can not only reflect the relationship between the three principal stresses in the crust, but also represent the degree of compression of the lithosphere. The higher the value of K av , the higher the corresponding degree of compression. Plot the K av values determined by each in-situ stress measurement method as a function of depth, compare whether the variation laws of the K av values obtained by different methods with depth are consistent, and non-linearly fit the K av values and depth in the form of a hyperbolic function (Equation (15)), and compare the fitting curve with the widely accepted Brown-Hoek inner and outer envelope curves (Equation (16)) in the world to visually judge whether they have a similar evolution trend.

[0131]

[0132] In the formula: c and d are undetermined coefficients.

[0133]

[0134] Visualize the direction of σ H obtained by each in-situ stress measurement method by drawing its rose diagram, determine the dominant direction of σ H , and visually compare whether the dominant directions of σ H are consistent; in particular, the included angles of the average σ H directions obtained by different methods can be compared. For the first time, a discrimination criterion for the difference in in-situ stress directions (Equation (17)) is proposed, that is, if the included angle between the directions of the average maximum horizontal principal stress determined by different measurement methods is less than or equal to the included angle threshold (for example, 15°), it is considered that the dominant directions of σ H obtained by different methods are basically consistent. Among them, the rose diagram represents the dominant direction (main acting direction) of the principal stress, and the dominant direction refers to the dominant direction of the maximum horizontal principal stress in a certain area. The included angle threshold can be any value not less than 0 and not greater than 15°.

[0135]

[0136] In the formula: The average maximum horizontal principal stress direction determined by a certain in-situ stress measurement method The average maximum horizontal principal stress direction determined by another in-situ stress measurement method

[0137] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0138] First, according to the Bayesian linear regression method, the present invention models and aligns the in-situ stress measurement data to intuitively verify the reliability of the in-situ stress measurement method

[0139] Second, based on the improved Kolmogorov-Smirnov statistical test method, the present invention further ensures and verifies the reliability and accuracy of the in-situ stress measurement method

[0140] Third, the method for determining the reliability of the in-situ stress measurement method provided by the present invention realizes the determination of the reliability of different in-situ stress measurement methods in the same area by judging the degree of fit between the second in-situ stress measurement data and the principal stress distribution model

[0141] Fourth, the present invention compares the magnitudes and directions of the in-situ stresses obtained by different in-situ stress measurement methods through means such as the principal stress difference, stress structure, and rose diagram, and the comparison results are intuitive and accurate

[0142] Fifth, the present invention determines the reliability of the in-situ stress measurement method through various means, providing a basis for further improving the measurement accuracy of the in-situ stress measurement method

[0143] For further explanation of the present application, the following takes two in-situ stress measurement methods as examples for illustration. It can be understood that the comprehensive determination of the reliability of more than three in-situ stress measurement methods can also be carried out

[0144] The hollow inclusion stress relief method and the hydraulic fracturing method are respectively used to carry out in-situ stress measurement work in a certain underground metal mine in China, and a large number of in-situ stress measured data of two types are obtained

[0145] First, the proposed new Bayesian linear regression method is used to model and align the in-situ stress data obtained by the stress relief method and the hydraulic fracturing method, and the probability density diagrams of the maximum horizontal principal stress, the minimum horizontal principal stress, and the vertical principal stress are drawn (as Figure 3(a) - Figure 3(c) shown). It can be observed that the shapes of the probability density diagrams of the three principal stresses measured by the stress relief method and the hydraulic fracturing method in the same mining area are very similar, and the expected values of the data are basically the same, that is, the curves almost overlap, indicating that the in-situ stress data measured by the stress relief method and the hydraulic fracturing method in the same mining area basically follow the same distribution law

[0146] Secondly, the Kolmogorov-Smirnov test was performed on the aligned data using a self-developed MATLAB program. The results showed that all the in-situ stress data measured by the stress relief method and the hydraulic fracturing method accepted the null hypothesis, that is, the two types of data followed the same distribution. In addition, the p-values of the Kolmogorov-Smirnov test for the maximum horizontal principal stress, the minimum horizontal principal stress, and the vertical principal stress were 0.1800, 0.9713, and 0.8514 respectively (as Figure 4 shown), all of which were greater than the selected significance level of 0.01 (the same result was obtained for the selected significance level of 0.05). It was considered that the in-situ stress data measured by the stress relief method and the hydraulic fracturing method in this mining area followed the same distribution law.

[0147] Furthermore, based on the in-situ stress measurement results obtained by the stress relief method and the hydraulic fracturing method, a principal stress distribution model along the depth direction was established using the regression method (as Figure 5 shown), and the correlation coefficients of the distribution models of the three principal stresses of σ H 、σ h 、σ v obtained were 0.9406, 0.9521, and 0.9932 respectively, indicating that the three principal stresses obtained by the stress relief method and the hydraulic fracturing method fitted very well along the depth direction. Among them, the correlation coefficient was obtained by linear regression fitting. Here, the σ H 、σ h 、σ v data of the hydraulic fracturing method and the stress relief method were put together for fitting respectively.

[0148] Furthermore, the magnitudes of the principal stresses determined by the stress relief method and the hydraulic fracturing method at the same or approximate depth levels were compared. For example, the magnitudes of σ H 、σ h 、σ v measured by the stress relief method at a depth of 510 m were 24.55 MPa, 16.35 MPa, and 14.49 MPa respectively, and the magnitudes of σ H 、σ h 、σ v measured by the hydraulic fracturing method at a depth of 509 m were 25.02 MPa, 18.43 MPa, and 13.48 MPa respectively. According to Equation (2), the calculated f values were 1.88%, 11.29%, and 6.97% respectively, all of which were less than 20% (the principal stress difference threshold was taken as 20%); the magnitudes of σ H 、σ h 、σ v measured by the stress relief method at a depth of 600 m were 30.17 MPa, 18.83 MPa, and 16.94 MPa respectively, and the magnitudes of σ H 、σh and σ v The measured values are 30.20 MPa, 20.31 MPa and 16.39 MPa respectively. According to Equation (2), the calculated f values are 0.10%, 7.29% and 3.25% respectively, all of which are less than 20%. It can be considered that the difference in the principal stress values determined by the stress relief method and the hydraulic fracturing method is very small.

[0149] Furthermore, compare the maximum horizontal principal stress σ H , the minimum horizontal principal stress σ h and the vertical principal stress σ v obtained by each in-situ stress measurement method (Table 1). Among the stress data obtained by the stress relief method, there are 8 groups of σ H > σ h > σ v type stress structures, and 10 groups of σ H > σ v > σ h type stress structures. There are no σ v > σ H > σ h type stress structures. Among the stress data obtained by the hydraulic fracturing method, there are 10 groups of σ H > σ h > σ v type stress structures, and 13 groups of σ H > σ v > σ h type stress structures. There are no σ v > σ H > σ h type stress structures. It can be seen that the dominant stress structures determined by the stress relief method and the hydraulic fracturing method are the same, both being the σ H > σ v > σ h type stress structure.

[0150] Table 1 Stress structure statistics

[0151]

[0152] Furthermore, calculate the ratio K av of the average horizontal principal stress to the vertical principal stress. Plot the K av values determined by the stress relief method and the hydraulic fracturing method as a function of depth (as shown in Figure 6 ), and perform a non-linear fit of the K av values and the depth in the form of a hyperbolic function. It can be seen that the variation laws of the K av values obtained by the stress relief method and the hydraulic fracturing method are relatively consistent, and are within the inner and outer envelopes of Brown-Hoek, showing a similar evolution trend.

[0153] Finally, the σ obtained by stress relief and hydraulic fracturing H Direction draws a rose diagram ( Figure 7(a) - Figure 7(b) ), σ determined by stress relief method H The dominant direction is 110.9° (N69.1°W) (as shown in Figure 7(a)), and the σ determined by the hydraulic fracturing method H The dominant direction is 121.9° (N58.1°W) (as shown in Figure 7(b)). It can be seen that the angle between the average maximum horizontal principal stress direction determined by the stress relief method and the hydraulic fracturing method is 11°, which is less than 15°. It is believed that the σ obtained by these two methods is H The dominant directions of advantages are basically the same.

[0154] In summary, compared with the traditional simple statistical method, the proposed new method can compare and evaluate any different types of ground stress value and direction data to determine whether they have a consistent distribution pattern, which is not available in other current methods and is of great value for further improving the measurement accuracy of ground stress measurement methods. The comprehensive comparison method constructed above has reached a consistent conclusion, thus proving the accuracy and feasibility of the method for determining the reliability of local stress measurement methods.

[0155] Exemplary Devices

[0156] Figure 8 FIG. 8 is a schematic diagram of a system 80 for comprehensively determining the reliability of different ground stress measurement methods in the same area provided by at least one exemplary embodiment of the present application. Figure 8 As shown, the comprehensive determination system 80 for the reliability of different geostress measurement methods in the same area includes: a first geostress acquisition module 81, which is used to obtain first geostress measurement data based on the first geostress measurement method; a second geostress acquisition module 82, which is used to obtain second geostress measurement data based on the second geostress measurement method in the same area; a reliability judgment module 83, which is used to: model and align the first geostress measurement data using the Bayesian linear regression method; model and align the second geostress measurement data using the Bayesian linear regression method; draw a first probability density map based on the first geostress measurement data after data alignment, draw a second probability density map based on the second geostress measurement data after data alignment, and compare the similarity between the first probability density map and the second probability density map. Wherein, the second geostress measurement method includes at least one geostress measurement method other than the first geostress measurement method.

[0157] After the data are completed, the probability density map is used to determine whether the geostress data obtained by different geostress measurement methods obey the same distribution, thereby verifying the consistency and validity of the data obtained by different types of geostress measurement methods.

[0158] The comprehensive determination system uses the Bayesian linear regression method through the above-mentioned modules to determine whether different types of in-situ stress data follow the same distribution, realizes the determination of the reliability of different in-situ stress measurement methods in the same area, and helps to further improve the measurement accuracy and reliability of in-situ stress measurement methods.

[0159] In addition, the comprehensive determination system can also use the Kolmogorov-Smirnov test to further determine the reliability of different in-situ stress measurement methods in the same area, which will not be elaborated here.

[0160] It can be understood that the specific examples in this document are only to help those skilled in the art better understand the embodiments of this specification, rather than limiting the scope of this specification.

[0161] It can be understood that in various embodiments of this specification, the order of each step does not mean the sequence of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of this specification.

[0162] It can be understood that the various embodiments described in this specification can be implemented alone or in combination, and this specification does not limit this.

[0163] Unless otherwise specified, all technical and scientific terms used in the embodiments of this specification have the same meaning as commonly understood by those skilled in the technical field of this specification. The terms used in this specification are only for the purpose of describing specific embodiments and are not intended to limit the scope of this specification. The term "and / or" used in the embodiments of this specification and the appended claims includes any and all combinations of one or more of the related listed items. The singular forms "a", "the above" and "the" used in the embodiments of this specification and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0164] As mentioned above, this is only the specific embodiment of this specification, but the protection scope of this specification is not limited thereto. Any person skilled in the art within the technical scope disclosed in this specification can easily think of changes or substitutions, which should all be covered within the protection scope of this specification. Therefore, the protection scope of this specification should be subject to the protection scope of the claims.

Claims

1. A comprehensive determination method for the reliability of different in-situ stress measurement methods in the same area, characterized in that, Including: Obtaining first in-situ stress measurement data based on a first in-situ stress measurement method; In the same area, obtaining second in-situ stress measurement data based on a second in-situ stress measurement method; Using the Bayesian linear regression method to model and align the first in-situ stress measurement data; Using the Bayesian linear regression method to model and align the second in-situ stress measurement data; Drawing a first probability density plot based on the first in-situ stress measurement data after data alignment, drawing a second probability density plot based on the second in-situ stress measurement data after data alignment, and comparing the similarity between the first probability density plot and the second probability density plot to determine whether the first in-situ stress data and the second in-situ stress data follow the same distribution, thereby verifying the reliability of different in-situ stress measurement methods in the same area; Wherein, the second in-situ stress measurement method includes at least one in-situ stress measurement method other than the first in-situ stress measurement method; To further ensure and verify the correctness of the first probability density plot and the second probability density plot, it further includes: Assuming that the first in-situ stress measurement data and the second in-situ stress measurement data come from the same distribution; Calculating a first cumulative distribution function of the first in-situ stress measurement data and a second cumulative distribution function of the second in-situ stress measurement data; Calculating the Kolmogorov-Smirnov statistic, where the Kolmogorov-Smirnov statistic represents the maximum vertical distance between the first cumulative distribution function and the second cumulative distribution function; Finding the corresponding Kolmogorov-Smirnov critical value according to the selected significance level; Determining whether the first in-situ stress measurement data and the second in-situ stress measurement data follow the same distribution; It further includes: Based on the first in-situ stress measurement data, establishing a principal stress distribution model along a fixed direction using a mathematical method; Judging the degree of fit between the second in-situ stress measurement data and the principal stress distribution model. If the degree of fit is high, the reliability of the first in-situ stress measurement method is high, otherwise it is low; Wherein, the principal stress distribution model represents the corresponding relationship between the principal stress along the fixed direction in the first in-situ stress measurement data and the depth, and the degree of fit is characterized by the corresponding parameters between the second in-situ stress measurement data and the principal stress distribution model; The degree of fit is characterized by the correlation coefficient between the second in-situ stress measurement data and the principal stress distribution model. If the correlation coefficient is higher than the fit threshold, the degree of fit is high.

2. The comprehensive determination method for the reliability of the method for measuring different in-situ stresses in the same area according to claim 1, characterized in that It further includes: At the same or approximately the same depth level, calculating the principal stress difference rate f between the principal stress in the first in-situ stress measurement data and the principal stress in the same direction in the second in-situ stress measurement data according to the following formula. If the principal stress difference rate is not greater than the principal stress difference rate threshold, the reliability of the first in-situ stress measurement method is high, otherwise it is low; ； wherein, is the maximum horizontal principal stress, the minimum horizontal principal stress or the vertical principal stress in the first in-situ stress measurement data, is the corresponding maximum horizontal principal stress, the minimum horizontal principal stress or the vertical principal stress in the second in-situ stress measurement data.

3. The comprehensive determination method for the reliability of the different in-situ stress measurement methods in the same area according to claim 1, characterized in that, It further includes: Determining a first stress structure of the first in-situ stress measurement data; Determining a second stress structure of the second in-situ stress measurement data; Compare whether the first dominant stress structure of the first in-situ stress measurement data is consistent with the second dominant stress structure of the second in-situ stress measurement data. If they are consistent, the reliability of the first in-situ stress measurement method is high; otherwise, it is low. Among them, the first dominant stress structure and the second dominant stress structure respectively characterize the stress structure reflected by the measurement data with the most conforming stress structure in the first stress structure and the second stress structure.

4. The comprehensive determination method for the reliability of the method for measuring different in-situ stresses in the same area according to claim 1, characterized in that, It also includes: Calculate the first ratio of the average horizontal principal stress to the vertical principal stress of the first in-situ stress measurement data; Calculate the second ratio of the average horizontal principal stress to the vertical principal stress of the second in-situ stress measurement data; Compare whether the variation laws of the first ratio and the second ratio with depth are consistent. If they are consistent, the reliability of the first in-situ stress measurement method is high; otherwise, it is low.

5. The comprehensive determination method for the reliability of the method for measuring different in-situ stresses in the same area according to claim 4, characterized in that It also includes: Perform non-linear fitting of the first ratio and the second ratio with depth respectively to generate a first fitting curve and a second fitting curve; Compare the first fitting curve and the second fitting curve with the Brown-Hoek inner and outer envelope lines. If the first fitting curve and the second fitting curve are inside the Brown-Hoek inner and outer envelope lines, the reliability of the first in-situ stress measurement method is high; otherwise, it is low.

6. The comprehensive determination method for the reliability of the same-region different in-situ stress measurement method according to claim 1, characterized in that, It also includes: Draw a first rose diagram based on the direction of the maximum horizontal principal stress of the first in-situ stress measurement data; Draw a second rose diagram based on the direction of the maximum horizontal principal stress of the second in-situ stress measurement data; If the first dominant direction of the first rose diagram is consistent with the second dominant direction of the first rose diagram, the reliability of the first in-situ stress measurement method is high; otherwise, it is low; Among them, the first dominant direction and the second dominant direction respectively characterize the main acting directions of the maximum horizontal principal stress of the first rose diagram and the maximum horizontal principal stress of the second rose diagram.

7. The comprehensive determination method for the reliability of the different in-situ stress measurement methods in the same area according to claim 6, characterized in that, It also includes: If the included angle between the first dominant direction and the second dominant direction is less than the included angle threshold, the reliability of the first in-situ stress measurement method is high; otherwise, it is low.

Images