Rock damage monitoring and early warning method, system and device based on multi-index fusion and storage medium

Abstract

The invention discloses a rock damage monitoring and early warning method, system and device based on multi-index fusion and a storage medium, and the method comprises the steps: carrying out the simulation loading test of a rock test piece collected in a target monitoring environment, collecting the test data of a plurality of indexes of the rock test piece at each preset moment, and obtaining a multi-index data set, the indexes comprise an elastic energy ratio, a dissipated energy ratio and acoustic emission energy; the multi-index data set is preprocessed; constructing a rock damage early warning model based on a sharp point mutation theory and the preprocessed multi-index data set; and performing rock damage monitoring and early warning on the target monitoring environment based on the rock damage early warning model. According to the method, comprehensive, accurate and stable monitoring and early warning of the rock damage state are realized by comprehensively utilizing a plurality of characteristics such as the acoustic emission energy, the elastic energy ratio and the dissipated energy ratio.

Description

Technical Field

[0001] The present application relates to the field of rock monitoring technology, and in particular to a rock damage monitoring and early warning method, system, device and storage medium based on multi-indicator fusion. Background Art

[0002] As a major component of nature, rocks have complex and changeable mechanical properties. When subjected to external loads, they will undergo a whole process from elastic deformation to final destruction. Rock fracture is not only a complex energy conversion process, but also involves a variety of physical phenomena such as stress, strain, microcrack initiation and propagation. In the engineering field, especially in underground mining, tunneling, geological disaster prevention and control, oil and gas extraction and other scenarios, rock stability is crucial. Rock instability and destruction may lead to serious safety accidents, such as ground collapse, tunnel collapse or slope landslide. Therefore, real-time monitoring and early warning of the evolution of rock damage are particularly important.

[0003] At present, the monitoring methods of rock damage status mainly include stress-strain monitoring method, acoustic emission monitoring method and energy index detection method.

[0004] The stress-strain monitoring method determines the degree of damage by measuring the changes in stress and strain of rocks during the stress process. The stress-strain relationship can reflect the elastic deformation, plastic deformation and final destruction process of rocks. However, the changes in stress and strain are localized and cannot capture the early initiation of cracks inside rocks and the microscopic energy conversion process. This makes the prediction of the stress-strain monitoring method often lag behind the actual damage process, resulting in untimely warnings.

[0005] Acoustic Emission (AE) technology evaluates the damage state of rock by monitoring the high-frequency microseismic signals generated during the crack expansion and destruction process inside the rock. Acoustic emission signals can reflect the dynamic process of micro crack initiation and expansion, and have high sensitivity. It is currently a commonly used rock damage monitoring method. However, relying solely on the intensity or cumulative energy of the acoustic emission signal for judgment is easily affected by noise, and accuracy is difficult to guarantee in a high noise background environment.

[0006] In the study of rock mechanics, energy accumulation and dissipation are considered to be important characteristics of rock failure. By monitoring the elastic energy and dissipated energy of rock during loading, the degree of rock damage can be determined. Although this method can reflect the overall energy change of rock, the singleness of its indicators makes it insufficient for early identification of rock damage.

[0007] In rock damage monitoring, traditional methods often rely on a single indicator (such as stress, acoustic emission or energy indicator) for judgment. However, the rock failure process is a complex phenomenon with multiple dimensions and scales. A single indicator is difficult to fully reflect the overall picture of rock damage. The damage inside the rock is the result of crack initiation, expansion and interaction. A single indicator is difficult to capture the damage evolution characteristics at different stages. For example, stress and strain monitoring mainly reflects local deformation information, while rock damage usually starts from microscopic cracks, which will expand to destruction before stress or strain monitoring reaches a critical state. Acoustic emission signals are easily interfered by environmental noise, resulting in false alarms or missed alarms, especially in complex engineering background environments, where accuracy cannot be guaranteed. Therefore, the limitations of the single indicator method in rock damage early warning are very obvious, and there is an urgent need to integrate multiple indicators to achieve more comprehensive, accurate and stable rock damage state monitoring and early warning. Summary of the invention

[0008] In order to solve the above technical problems, the present application provides a rock damage monitoring and early warning method, system, device and storage medium based on multi-indicator fusion. By comprehensively utilizing multiple characteristics such as acoustic emission energy, elastic energy ratio and dissipated energy ratio, comprehensive, accurate and stable monitoring and early warning of rock damage status are achieved.

[0009] The first objective of this application is to provide a rock damage monitoring and early warning method based on multi-indicator fusion.

[0010] The above-mentioned application objective 1 of the present application is achieved through the following technical solutions:

[0011] A rock damage monitoring and early warning method based on multi-index fusion, the method comprising the following steps:

[0012] S1, performing a simulated loading test on a rock specimen collected in a target monitoring environment, and collecting test data of multiple indicators of the rock specimen at each preset time to obtain a multi-indicator data set, wherein the indicators include elastic energy proportion, dissipated energy proportion and acoustic emission energy;

[0013] S2, preprocessing the multi-indicator data set;

[0014] S3, constructing a rock damage early warning model based on cusp catastrophe theory and preprocessed multi-index data sets;

[0015] S4, performing rock damage monitoring and early warning on the target monitoring environment based on the rock damage early warning model.

[0016] Preferably, in step S1, collecting test data of multiple indicators of the rock specimen at various preset moments to obtain a multi-indicator data set includes:

[0017] S11, collecting the axial stress, axial strain, elastic modulus and acoustic emission energy of the rock specimen at each preset time during the simulated loading test;

[0018] S12, based on the collected axial stress, axial strain and elastic modulus of the rock specimen at each preset time, calculate the elastic energy proportion and dissipated energy proportion of the rock specimen at each preset time according to a preset indicator data calculation model;

[0019] S13, fusing the elastic energy proportion, dissipated energy proportion and acoustic emission energy at each preset moment to obtain the multi-index data set.

[0020] Preferably, in step S12, the indicator data calculation model is as follows:

[0021]

[0022] Among them, η is the elastic energy ratio, λ is the dissipated energy ratio, U is the total input energy, and U e is the elastic strain energy, U d is the dissipated energy;

[0023] The calculation formula of the total input energy U is:

[0024]

[0025] Among them, ε is the axial strain and σ is the axial stress;

[0026]

[0027] Among them, E u is the tangent elastic modulus, E u ≈E0, E0 is the elastic modulus;

[0028] Elastic strain energy U e The approximate calculation is:

[0029]

[0030] Then the dissipated energy U d The calculation formula is:

[0031]

[0032] Preferably, in step S2, the preprocessing of the multi-index data set includes:

[0033] S21, normalizing each test data of elastic energy ratio, dissipated energy ratio and acoustic emission energy in the multi-index data set by using a deviation normalization method;

[0034] S22, the entropy weight method is used to weight the normalized elastic energy proportion, dissipated energy proportion and acoustic emission energy.

[0035] Preferably, in step S21, the calculation formula of the deviation standardization method is as follows:

[0036]

[0037] Among them, x' ij is the normalized index value, x ij is the original indicator value in the multi-indicator data set, indicating the value of the jth data of the ith indicator, where i=1,2,3, j=1,2,3…n, and n is a positive integer indicating the number of data of the indicator corresponding to the multi-indicator data set.

[0038] Preferably, in step S22, the entropy weight method weighting process is as follows:

[0039] S221, normalize the normalized index value to obtain a standardized matrix, and the calculation formula is as follows:

[0040]

[0041] Among them, P ij is the standardized matrix;

[0042] S222, calculate the entropy value of the i-th indicator, the formula is as follows:

[0043]

[0044] Among them, e i is the information entropy of the ith indicator, k is a constant;

[0045] S223, calculate the information utility value of each indicator, the calculation formula is as follows:

[0046] d i =1-e i

[0047] Among them, d i is the information utility value of each indicator;

[0048] S224, according to the information utility value of each indicator, determine the weight coefficient of each indicator, and the calculation formula is as follows:

[0049]

[0050] Among them, w i is the weight of the i-th indicator;

[0051] S225, based on the calculated weights of each index, weight the normalized elastic energy ratio, dissipated energy ratio, and acoustic emission energy to obtain the index values ​​of the preprocessed multi-index data set, and the calculation formula is as follows:

[0052] X i =w i x i

[0053] Among them, X i is the weighted result of the i-th indicator, x i is the ith indicator.

[0054] Preferably, in step S3, constructing a rock damage early warning model based on cusp catastrophe theory and the preprocessed multi-index data set includes:

[0055] S31, constructing the discriminant function in the rock damage early warning model based on the cusp catastrophe theory;

[0056] S32, constructing the expression of the control factor in the discriminant function of the rock damage early warning model based on the preprocessed multi-index data set;

[0057] S33, using the maximum likelihood estimation method to calculate the linear transformation coefficient and the constant term in the expression of the control factor in the discriminant function;

[0058] S34, constructing the rock damage early warning model based on the calculated linear transformation coefficient and constant term in the expression of the control factor in the discriminant function, the expression of the control factor in the discriminant function and the discriminant function.

[0059] The second objective of this application is to provide a rock damage monitoring and early warning system based on multi-indicator fusion.

[0060] The second application objective of the present application is achieved through the following technical solutions:

[0061] A rock damage monitoring and early warning system based on multi-index fusion, the system comprising:

[0062] An indicator data set construction module is used to perform a simulated loading test on a rock specimen collected in a target monitoring environment, and collect test data of multiple indicators of the rock specimen at each preset time to obtain a multi-indicator data set, wherein the indicators include elastic energy proportion, dissipated energy proportion and acoustic emission energy;

[0063] A data preprocessing module, used for preprocessing the multi-indicator data set;

[0064] Model building module, used to build rock damage early warning model based on cusp catastrophe theory and preprocessed multi-indicator data set;

[0065] The rock damage monitoring and early warning module is used to perform rock damage monitoring and early warning on the target monitoring environment based on the rock damage early warning model.

[0066] Preferably, collecting test data of multiple indicators of the rock specimen at various preset moments to obtain a multi-indicator data set includes:

[0067] Collect the axial stress, axial strain, elastic modulus and acoustic emission energy of the rock specimen at each preset time during the simulated loading test;

[0068] Based on the collected axial stress, axial strain and elastic modulus of the rock specimen at each preset time, the elastic energy proportion and dissipated energy proportion of the rock specimen at each preset time are calculated according to a preset index data calculation model;

[0069] The elastic energy proportion, dissipated energy proportion and acoustic emission energy at each preset moment are fused to obtain the multi-index data set.

[0070] Preferably, the indicator data calculation model is as follows:

[0071]

[0072] Among them, η is the elastic energy ratio, λ is the dissipated energy ratio, U is the total input energy, and U e is the elastic strain energy, U d is the dissipated energy;

[0073] The calculation formula of the total input energy U is:

[0074]

[0075] Among them, ε is the axial strain and σ is the axial stress;

[0076]

[0077] Among them, E u is the tangent elastic modulus, E u ≈E0, E0 is the elastic modulus;

[0078] Elastic strain energy U e The approximate calculation is:

[0079]

[0080] Then the dissipated energy U d The calculation formula is:

[0081]

[0082] Preferably, the preprocessing of the multi-indicator data set includes:

[0083] The deviation normalization method is used to normalize the elastic energy ratio, dissipated energy ratio and acoustic emission energy test data in the multi-index data set;

[0084] The entropy weight method is used to weight the normalized elastic energy proportion, dissipated energy proportion and acoustic emission energy.

[0085] Preferably, the calculation formula of the deviation standardization method is as follows:

[0086]

[0087] Among them, x' ij is the normalized index value, x ij is the original indicator value in the multi-indicator data set, indicating the value of the jth data of the ith indicator, where i=1,2,3, j=1,2,3…n, and n is a positive integer indicating the number of data of the indicator corresponding to the multi-indicator data set.

[0088] Preferably, the entropy weight method weighting process is as follows:

[0089] The normalized index values ​​are standardized to obtain a standardized matrix. The calculation formula is as follows:

[0090]

[0091] Among them, P ij is the standardized matrix;

[0092] Calculate the entropy value of the i-th indicator, the formula is as follows:

[0093]

[0094] Among them, e i is the information entropy of the ith indicator, k is a constant;

[0095] Calculate the information utility value of each indicator using the following formula:

[0096] d i =1-e i

[0097] Among them, d i is the information utility value of each indicator;

[0098] According to the information utility value of each indicator, the weight coefficient of each indicator is determined, and the calculation formula is as follows:

[0099]

[0100] Among them, w i is the weight of the i-th indicator;

[0101] Based on the calculated weights of each index, the normalized elastic energy ratio, dissipated energy ratio and acoustic emission energy are weighted to obtain the index values ​​of the preprocessed multi-index data set. The calculation formula is as follows:

[0102] X i =w i x i

[0103] Among them, X i is the weighted result of the i-th indicator, x i is the ith indicator.

[0104] Preferably, the construction of a rock damage early warning model based on cusp catastrophe theory and preprocessed multi-index data set includes:

[0105] Construct the discriminant function in the rock damage early warning model based on the cusp catastrophe theory;

[0106] Based on the preprocessed multi-index data set, the expression of the control factor in the discriminant function of the rock damage early warning model is constructed;

[0107] The maximum likelihood estimation method is used to calculate the linear transformation coefficient and the constant term in the expression of the control factor in the discriminant function;

[0108] The rock damage early warning model is constructed based on the calculated linear transformation coefficient and constant term in the expression of the control factor in the discriminant function, the expression of the control factor in the discriminant function and the discriminant function.

[0109] The third objective of the present application is to provide a rock damage monitoring and early warning device based on multi-indicator fusion.

[0110] The third application objective of this application is achieved through the following technical solutions:

[0111] A rock damage monitoring and early warning device based on multi-index fusion, comprising:

[0112] A memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the rock damage monitoring and early warning method based on multi-indicator fusion as described in any one of the first objectives of the present application are implemented.

[0113] The fourth object of this application is to provide a computer-readable storage medium.

[0114] The fourth application objective of this application is achieved through the following technical solutions:

[0115] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a rock damage monitoring and early warning method based on multi-indicator fusion as described in any one of the first objectives of the present application.

[0116] In summary, the present application discloses a rock damage monitoring and early warning method, system, device and storage medium based on multi-index fusion, which conducts simulated loading tests on rock specimens collected in a target monitoring environment, and collects test data of multiple indicators of the rock specimens at various preset moments, to obtain a multi-index data set, wherein the indicators include elastic energy proportion, dissipated energy proportion and acoustic emission energy; preprocesses the multi-index data set; constructs a rock damage early warning model based on cusp mutation theory and the preprocessed multi-index data set; and monitors and warns rock damage in the target monitoring environment based on the rock damage early warning model. The present application realizes comprehensive, accurate and stable monitoring and early warning of rock damage status by comprehensively utilizing multiple features such as acoustic emission energy, elastic energy proportion and dissipated energy proportion, effectively overcomes the limitations of traditional single-indicator damage early warning methods, enhances the early recognition capability of rock internal damage status, and improves the accuracy and robustness of the early warning model in complex engineering environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0117] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0118] Figure 1 This is a flow chart of a rock damage monitoring and early warning method based on multi-index fusion in one embodiment of the present application;

[0119] Figure 2(a) is a graph showing the evolution characteristics of the sandstone mutation index of the complete specimen at different test stages when using the traditional single index method for rock damage early warning;

[0120] Figure 2(b) is a graph showing the evolution characteristic curve of sandstone mutation index when the hole inclination is 0° at different test stages when using the traditional single index method for rock damage early warning;

[0121] Figure 2(c) is a graph showing the evolution characteristic curve of sandstone mutation index when the hole inclination angle is 30° at different test stages when using the traditional single index method for rock damage early warning;

[0122] Figure 2(d) is a graph showing the evolution characteristic curve of the sandstone mutation index when the hole inclination angle is 60° at different test stages when using the traditional single index method for rock damage early warning;

[0123] Figure 2(e) is a graph showing the evolution characteristic curve of the sandstone mutation index when the hole inclination angle is 90° at different test stages when using the traditional single index method for rock damage early warning;

[0124] FIG3( a ) is a graph showing the evolution characteristics of sandstone mutation indicators of a complete specimen at different test stages when performing rock damage early warning using a multi-indicator fusion method according to an embodiment of the present application;

[0125] FIG3( b ) is a graph showing the evolution characteristics of sandstone mutation indicators at different test stages when the hole inclination angle is 0° when performing rock damage early warning using the multi-indicator fusion method of an embodiment of the present application;

[0126] FIG3( c ) is a graph showing the evolution characteristics of sandstone mutation indicators at different test stages when the hole inclination angle is 30° when performing rock damage early warning using the multi-indicator fusion method of an embodiment of the present application;

[0127] FIG3( d ) is a graph showing the evolution characteristics of sandstone mutation indicators at different test stages when the hole inclination angle is 60° when performing rock damage early warning using the multi-indicator fusion method of an embodiment of the present application;

[0128] FIG3( e ) is a graph showing the evolution characteristics of sandstone mutation indicators at different test stages when the hole inclination angle is 90° when performing rock damage early warning using the multi-indicator fusion method of an embodiment of the present application;

[0129] Figure 4 This is a structural schematic diagram of a rock damage monitoring and early warning system based on multi-index fusion in one embodiment of the present application;

[0130] Figure 5 This is a structural schematic diagram of a rock damage monitoring and early warning device based on multi-indicator fusion in one embodiment of the present application. DETAILED DESCRIPTION

[0131] In order to enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below. Obviously, the described embodiments are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0132] In the embodiments provided in the present application, it should be understood that the disclosed methods and systems can be implemented in other ways. The system embodiments described below are merely schematic. For example, the division of units and modules is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or modules can be combined, or can be integrated into another system, or some features can be ignored, or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0133] In addition, all functional units in the embodiments of the present application may be integrated into one processor, or each unit may be a separate device, or two or more units may be integrated into one device; each functional unit in the embodiments of the present application may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0134] A person skilled in the art can understand that all or part of the steps of the following method embodiments can be completed by program instructions and related hardware. The aforementioned program instructions can be stored in a computer-readable storage medium. When the program instructions are executed, the steps of the following method embodiments are executed; and the aforementioned storage medium includes: a mobile storage device, a read-only memory (ROM), a magnetic disk or an optical disk, and other media that can store program codes.

[0135] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, "multiple" and "several" mean two or more, unless otherwise clearly and specifically defined.

[0136] The present application embodiment provides a rock damage monitoring and early warning method based on multi-index fusion, such as Figure 1 As shown, the method may include the following steps:

[0137] S1, conducting a simulated loading test on a rock specimen collected in a target monitoring environment, and collecting test data of multiple indicators of the rock specimen at each preset time to obtain a multi-indicator data set, wherein the indicators include elastic energy proportion, dissipated energy proportion and acoustic emission energy;

[0138] S2, preprocessing of multi-indicator datasets;

[0139] S3, constructing a rock damage early warning model based on cusp catastrophe theory and preprocessed multi-index data sets;

[0140] S4, rock damage monitoring and early warning are performed on the target monitoring environment based on the rock damage early warning model.

[0141] In this embodiment, when rock damage monitoring and early warning are performed on the target monitoring environment based on the rock damage early warning model, the corresponding monitoring results of the rock damage status can be obtained by collecting data of multiple indicators of the rock in the target monitoring environment at each monitoring time, and inputting the monitored indicator data into the rock damage early warning model, and finally issuing a corresponding early warning signal based on the monitoring results.

[0142] In summary, the rock damage monitoring and early warning method based on multi-index fusion of this embodiment first performs a simulated loading test on the rock specimens collected in the target monitoring environment, and collects test data of multiple indicators of the rock specimens at each preset time to obtain a multi-index data set, wherein the indicators include elastic energy ratio, dissipated energy ratio and acoustic emission energy; then the multi-index data set is preprocessed; then a rock damage early warning model is constructed based on the cusp mutation theory and the preprocessed multi-index data set; finally, rock damage monitoring and early warning are performed on the target monitoring environment based on the rock damage early warning model. This application realizes comprehensive, accurate and stable monitoring and early warning of rock damage status by comprehensively utilizing multiple features such as acoustic emission energy, elastic energy ratio and dissipated energy ratio, effectively overcomes the limitations of the traditional single-indicator damage early warning method, enhances the early recognition capability of the internal damage status of the rock, and improves the accuracy and robustness of the early warning model in complex engineering environments.

[0143] In one embodiment, in step S1, test data of multiple indicators of the rock specimen at each preset time is collected to obtain a multi-indicator data set including:

[0144] S11, collecting the axial stress, axial strain, elastic modulus and acoustic emission energy of the rock specimen at each preset time during the simulated loading test;

[0145] S12, based on the collected axial stress, axial strain and elastic modulus of the rock specimen at each preset time, according to a preset index data calculation model, calculate the elastic energy proportion and dissipated energy proportion of the rock specimen at each preset time;

[0146] S13, fusing the elastic energy proportion, dissipated energy proportion and acoustic emission energy at each preset moment to obtain a multi-index data set.

[0147] In one embodiment, in step S12, the indicator data calculation model is as follows:

[0148]

[0149] Among them, η is the elastic energy ratio, λ is the dissipated energy ratio, U is the total input energy, and U e is the elastic strain energy, U d is the dissipated energy;

[0150] The calculation formula for the total input energy U is:

[0151]

[0152] Among them, ε is the axial strain and σ is the axial stress;

[0153]

[0154] Among them, E u is the tangent elastic modulus, E u ≈E0, E0 is the elastic modulus;

[0155] Elastic strain energy U e The approximate calculation is:

[0156]

[0157] Then the dissipated energy U d The calculation formula is:

[0158]

[0159] In one embodiment, in step S2, preprocessing the multi-indicator data set includes:

[0160] S21, using the deviation standardization method to normalize the elastic energy ratio, dissipated energy ratio and acoustic emission energy test data in the multi-index data set;

[0161] S22, the entropy weight method is used to weight the normalized elastic energy proportion, dissipated energy proportion and acoustic emission energy.

[0162] In one embodiment, in step S21, the calculation formula of the deviation standardization method is as follows:

[0163]

[0164] Among them, x' ij is the normalized index value, x ij is the original indicator value in the multi-indicator data set, indicating the value of the jth data of the ith indicator, where i=1,2,3, j=1,2,3…n, and n is a positive integer indicating the number of data of the indicator corresponding to the multi-indicator data set.

[0165] In one embodiment, in step S22, the entropy weight method weighting process is as follows:

[0166] S221, normalize the normalized index value to obtain a standardized matrix, and the calculation formula is as follows:

[0167]

[0168] Among them, P ij is the standardized matrix;

[0169] S222, calculate the entropy value of the i-th indicator, the formula is as follows:

[0170]

[0171] Among them, e i is the information entropy of the ith indicator, k is a constant;

[0172] S223, calculate the information utility value of each indicator, the calculation formula is as follows:

[0173] d i =1-e i

[0174] Among them, d i is the information utility value of each indicator;

[0175] S224, according to the information utility value of each indicator, determine the weight coefficient of each indicator, and the calculation formula is as follows:

[0176]

[0177] Among them, w i is the weight of the i-th indicator;

[0178] S225, based on the calculated weights of each index, weight the normalized elastic energy ratio, dissipated energy ratio, and acoustic emission energy to obtain the index values ​​of the preprocessed multi-index data set, and the calculation formula is as follows:

[0179] X i =w i x i

[0180] Among them, X i is the weighted result of the i-th indicator, x i is the ith indicator.

[0181] In one embodiment, in step S3, constructing a rock damage early warning model based on the cusp catastrophe theory and the preprocessed multi-index data set includes:

[0182] S31, constructing the discriminant function in the rock damage early warning model based on the cusp catastrophe theory;

[0183] S32, constructing the expression of the control factor in the discriminant function of the rock damage early warning model based on the preprocessed multi-index data set;

[0184] S33, using the maximum likelihood estimation method to calculate the linear transformation coefficient and the constant term in the expression of the control factor in the discriminant function;

[0185] S34, constructing a rock damage early warning model based on the calculated linear transformation coefficient and constant term in the expression of the control factor in the discriminant function, the expression of the control factor in the discriminant function and the discriminant function.

[0186] Specifically, this embodiment adopts the cusp catastrophe theory, combines the description method of stochastic differential equations and probability density functions, and introduces multi-index linear combinations to construct control factors and state factors, so as to reduce the impact of random noise on the prediction results, thereby improving the robustness and adaptability of the model.

[0187] The potential function of the cusp catastrophe theory is:

[0188] g(x,μ,v)=x 4 +μx 2 +vx

[0189] Among them, μ and v are control factors; x is the state factor; g(x,μ,ν) is the potential function;

[0190] Let the first and second derivatives of the potential function be zero, eliminate x, and solve the bifurcation set equation:

[0191] 27ν 2 +8μ 3 =0

[0192] The discriminant function is:

[0193] Δ=27ν 2 +8μ 3

[0194] Among them, Δ is the discriminant function.

[0195] By introducing the stochastic differential equation, the distribution of x under the condition of fixed μ and v is described in the form of probability density function, and its formula is:

[0196] f(x|μ,v)=Ψexp(x 4 +μx 2 +vx)

[0197] Here, Ψ is a normalization constant.

[0198] The control factors μ and v are linear combinations of m-dimensional input variables (i.e., m indicators, m=3 in this embodiment), and the state factor x is a linear transformation of the output variable Y, which is expressed by the formula:

[0199]

[0200] Among them, a0, b0, c0 are constant terms; a i , b i 、c i is the linear transformation coefficient; x is the state factor, where the output variable Y is the damage variable at the current moment.

[0201] Damage variable calculation formula:

[0202]

[0203] Among them, E c is the accumulated acoustic emission energy at the current moment, E f is the cumulative acoustic emission energy when the rock is completely destroyed, that is, the reference energy, E c and E f All can be obtained through experiments.

[0204] When the maximum likelihood estimation method is used to calculate the linear transformation coefficient and constant term in the expression of the control factor in the discriminant function, the likelihood function L is:

[0205]

[0206] where θ is a linear combination of the model parameters, θ = {a0, a1, b0, b1, b2, b3, c0, c1, c2, c3}, and P(xθ) is the probability given the parameters.

[0207] Substitute the linear combination μ and v of the solved m-dimensional input variables into the discriminant function and calculate the discriminant function value. When the discriminant function Δ and the control factor μ are both less than 0, it indicates that the rock is in a critical state and strength failure may occur, and an early warning is issued.

[0208] Specifically, when the maximum likelihood estimation method is used to calculate the linear transformation coefficient and the constant term in the expression of the control factor in the discriminant function, the method uses the DFP algorithm (Davidon-Fletcher-Powell algorithm, quasi-Newton algorithm) when calculating, and the algorithm operation steps are as follows:

[0209] Step 1: Select model parameters θ = {a0, a1, b0, b1, b2, b3, c0, c1, c2, c3} and an initial positive definite matrix H0 (set to the identity matrix), set the initial number of iterations k = 0, and set the convergence value ε = 10 -4, calculate the gradient of the current point θ:

[0210]

[0211] in, is the objective function at x k The gradient at .

[0212] if Then stop the iteration and output the approximate value. The descent direction is calculated based on the current approximate Hessian matrix and the first-order derivative (gradient) of the objective function. The calculation formula is:

[0213]

[0214] Among them, p k is the search direction of the kth iteration, H k is the Hessian matrix at the kth iteration.

[0215] Step 2: Search along the direction P k Perform a line search to find the step size α k , so that the likelihood function value decreases to the maximum:

[0216] L(θ k+1 )=L(θ k +α k P k )

[0217] Step 3: Update model parameters:

[0218] θ k+1 =θ k +α k P k

[0219] Step 4: Update the Hessian matrix

[0220] First calculate the step vector S k and the gradient change vector y k :

[0221] S k =θ k+1 -θ k

[0222]

[0223] Among them, S k is the step vector, y k is the gradient change vector.

[0224] Update the approximate Hessian matrix H k+1 :

[0225]

[0226] Step 5: If for the new point θ k , the gradient modulus If it is still greater than ε, return to step 3 and use the new point θ k+1 and the matrix H k+1 Continue the iterative process. If the number of iterations reaches the preset maximum value or the modulus of the gradient is less than ε, the algorithm ends and the current point θ k+1 is considered to be the approximate minimum point of the function, i.e., θ k+1 The value of is the value of the model parameter θ.

[0227] Next, based on the test results shown in FIG. 2 and FIG. 3 , the warning effects of the traditional single-indicator damage status warning method and the multi-indicator damage status warning method of the embodiment of the present application are compared and explained.

[0228] During the loading process of rock, crack expansion and hole closure lead to dynamic changes in energy dissipation and storage. When the crack expansion reaches a certain critical point, the energy dissipation increases sharply, causing the system potential function to mutate, and the control factor and discriminant function become negative, indicating that the stress state and energy distribution of the system have changed significantly, and the rock specimen has been severely damaged. In addition, the microstructural changes inside the rock also affect the overall strength and stability. The changes in the control factors and discriminant functions reflect the changes in these microstructures. When these values ​​are less than 0, it means that the microcracks have expanded to a certain extent and the stability of the overall rock structure is reduced. This microstructural change is a precursor to rock damage and destruction. By monitoring the control factors and discriminant functions, the damage state of the rock can be identified in advance. In this embodiment, the window is set to 50s and the step size is set to 10s. The calculation results and the results of the traditional single indicator damage state warning method are shown in Figure 2. As can be seen from Figure 3, the warning interval is located in the unstable crack expansion stage before the peak stress, indicating that the rock specimen is more prone to instability mutations at this stage. This stage corresponds to the process of accelerated crack expansion inside the rock. Under the action of external load, a large amount of elastic energy is accumulated inside the rock, which makes the rock specimen in an unstable state. It is very easy to become unstable and suddenly destroyed under a small external disturbance, resulting in a large amount of elastic strain energy released in a short time. In addition, the rock specimen with a hole inclination angle of 60° appeared a situation where both the control factor and the discriminant function were less than 0 in the early stage of loading. This is because the local fracture of the rock sample in the early stage caused the energy storage capacity of the rock sample to decrease, the dissipated energy increased rapidly, and the micro-element inside the rock sample underwent plastic deformation, resulting in the appearance of "false instability". In addition, it can be found that the strain values ​​corresponding to the 90° mutation interval and the warning interval are larger. The main reason is that during the loading process of the rock sample with a hole inclination angle of 90°, the hole is parallel to the loading direction, the crack expansion path is more complicated, and the accumulation of energy and the gradual evolution of damage require more deformation to reach the critical condition of system failure, which reduces the brittleness of the rock during loading and enhances the ductility, so that the rock produces more strain near the damage. In conclusion, by monitoring the changes in control factors and discriminant functions, the strength failure and damage behavior of rock samples can be effectively predicted.

[0229] In this embodiment, the results are calculated for rock specimens with different hole inclination angles, as shown in Figure 3. In the constructed rock damage early warning model, the five rock specimens correspond to the complete rock specimen and the rock specimens with holes with inclination angles of 0°, 30°, 60°, and 90°, respectively. The early warning points appear at different stages of their respective peak stresses, specifically, the complete rock specimen at 0.93σ p –0.95σ p , 0° inclination rock specimen at 0.91σ p –0.94σ p , 30° inclination rock specimen at 0.90σ p –0.91σ p, 60° inclination rock specimen at 0.96σ p –0.98σ p , 90° inclination rock specimen at 0.91σ p –0.93σ p The results show that the early warning model can give timely warnings at different stages before the rock specimen reaches the peak stress, and the warning points are relatively concentrated, with an overall distribution of more than 90% of the peak stress, showing high sensitivity and accuracy. The warning point of the complete rock specimen appears at 0.93σ p –0.95σ p The range indicates that the model can identify the potential risk of damage in advance when the material is close to the ultimate bearing capacity. For the 0° inclination rock specimen, the warning points are distributed at 0.91σ p –0.94σ p , indicating that the model can still predict the failure trend relatively stably at this angle. In contrast, the warning point of the 30° inclination rock specimen is relatively early, appearing at 0.90σ p –0.91σ p , indicating that at this inclination angle, the rock specimen may fail prematurely before the peak stress is fully exerted, and the model is more sensitive to this. The warning point of the rock specimen with an inclination angle of 60° appears the latest, at 0.96σ p –0.98σ p This may be related to the complex influence of the hole on the local damage process of the material under this inclination angle, which makes the warning point delayed until the peak stress approaches the limit. The warning point of the 90° inclination rock specimen appears at 0.91σ p –0.93σ p , similar to the behavior of the 0° rock specimen, the model can also detect the damage risk earlier at this angle. From the data results, it can be seen that the comprehensive early warning model has good robustness and wide applicability, and can provide early warning near the peak stress point of different rock specimens, which makes the model perform better in rock fracture prediction.

[0230] Comparing Figure 2 with Figure 3, it can be found that the warning interval of the improved rock damage early warning model is significantly smaller than that of the single indicator model, and there is only one warning interval. The model can predict the dangerous area more accurately, reduce the possibility of false alarms and missed alarms, and thus improve the reliability of the early warning. The smaller warning interval indicates better adaptability and robustness under different environments and conditions, which is particularly important for complex actual engineering environments.

[0231] Combined with the above results, the mutation of the discriminant function and control factor of the single-indicator rock damage early warning model indicates the beginning of rock damage, which predicts rock damage earlier than the acoustic emission precursor. Therefore, the damage state early warning model that reveals the changes in energy conversion inside the rock can be used for rock damage early warning. In addition, the single-indicator rock damage early warning model only needs to monitor the deformation and stress of the rock. However, its early warning interval is large, which is prone to false alarms and missed alarms. It cannot fully reflect the damage state of the rock and may miss some important information. The rock damage early warning model that integrates multiple monitoring indicators can more comprehensively reflect the damage state of the rock. The warning interval range is significantly smaller, which reduces the possibility of false alarms and missed alarms, improves the reliability of the early warning, is more suitable for practical engineering applications, and can better adapt to changing environments.

[0232] It should be noted that the modeling process of the multi-indicator mutation model is relatively complex. It is necessary to consider the weight distribution and fusion method of multiple indicators, and more data support is needed to ensure the accuracy of the model.

[0233] The above-mentioned embodiment of the present application overcomes the limitation of a single indicator by integrating multi-dimensional energy indicators such as elastic energy proportion, dissipated energy proportion and acoustic emission energy, and comprehensively characterizes the whole process of rock from microcrack initiation to final destruction. The importance of each indicator is dynamically adjusted by weighting through the entropy weight method to ensure the adaptability and flexibility of the model to different working conditions, thereby significantly improving the accuracy and comprehensiveness of damage warning and reducing the occurrence of false alarms and missed alarms.

[0234] By introducing the cusp mutation model, this application can accurately identify the critical state of rock damage and achieve early warning. The use of stochastic differential equations effectively reduces the interference of noise on monitoring data in complex environments, enhances the noise resistance and robustness of the model, and enables the early warning method to maintain high accuracy and stability in complex engineering environments.

[0235] This application can not only provide early warning based on the three energy indicators of elastic energy proportion, dissipated energy proportion and acoustic emission energy, but also has good scalability. It can introduce more monitoring indicators (such as stress, deformation, etc.) according to actual engineering needs, so as to further improve the prediction accuracy and scope of application of the rock damage early warning model.

[0236] like Figure 4 As shown, the embodiment of the present application also provides a rock damage monitoring and early warning system based on multi-indicator fusion, which may include:

[0237] The indicator data set construction module 201 is used to perform a simulated loading test on a rock specimen collected in a target monitoring environment, and collect test data of multiple indicators of the rock specimen at each preset time to obtain a multi-indicator data set, wherein the indicators include elastic energy proportion, dissipated energy proportion and acoustic emission energy;

[0238] A data preprocessing module 202 is used to preprocess the multi-indicator data set;

[0239] A model building module 203 is used to build a rock damage early warning model based on the cusp catastrophe theory and the preprocessed multi-indicator data set;

[0240] The rock damage monitoring and early warning module 204 is used to perform rock damage monitoring and early warning on the target monitoring environment based on the rock damage early warning model.

[0241] In one embodiment, test data of multiple indicators of a rock specimen at each preset time is collected to obtain a multi-indicator data set including:

[0242] Collect the axial stress, axial strain, elastic modulus and acoustic emission energy of the rock specimen at each preset time during the simulated loading test;

[0243] Based on the collected axial stress, axial strain and elastic modulus of the rock specimen at each preset time, the elastic energy proportion and dissipated energy proportion of the rock specimen at each preset time are calculated according to the preset index data calculation model;

[0244] The elastic energy proportion, dissipated energy proportion and acoustic emission energy at each preset moment are fused to obtain a multi-index data set.

[0245] In one embodiment, the indicator data calculation model is as follows:

[0246]

[0247] Among them, η is the elastic energy ratio, λ is the dissipated energy ratio, U is the total input energy, and U e is the elastic strain energy, U d is the dissipated energy;

[0248] The calculation formula for the total input energy U is:

[0249]

[0250] Among them, ε is the axial strain and σ is the axial stress;

[0251]

[0252] Among them, E u is the tangent elastic modulus, E u ≈E0, E0 is the elastic modulus;

[0253] Elastic strain energy U e The approximate calculation is:

[0254]

[0255] Then the dissipated energy U d The calculation formula is:

[0256]

[0257] In one embodiment, preprocessing the multi-indicator data set includes:

[0258] The deviation standardization method is used to normalize the test data of elastic energy proportion, dissipated energy proportion and acoustic emission energy in the multi-index data set;

[0259] The entropy weight method is used to weight the normalized elastic energy proportion, dissipated energy proportion and acoustic emission energy.

[0260] In one embodiment, the calculation formula of the deviation normalization method is as follows:

[0261]

[0262] Among them, x' ij is the normalized index value, x ij is the original indicator value in the multi-indicator data set, indicating the value of the jth data of the ith indicator, where i=1,2,3, j=1,2,3…n, and n is a positive integer indicating the number of data of the indicator corresponding to the multi-indicator data set.

[0263] In one embodiment, the entropy weight method weighting process is as follows:

[0264] The normalized index values ​​are standardized to obtain a standardized matrix. The calculation formula is as follows:

[0265]

[0266] Among them, P ij is the standardized matrix;

[0267] Calculate the entropy value of the i-th indicator, the formula is as follows:

[0268]

[0269] Among them, e i is the information entropy of the ith indicator, k is a constant;

[0270] Calculate the information utility value of each indicator using the following formula:

[0271] d i =1-e i

[0272] Among them, d i is the information utility value of each indicator;

[0273] According to the information utility value of each indicator, the weight coefficient of each indicator is determined, and the calculation formula is as follows:

[0274]

[0275] Among them, w i is the weight of the i-th indicator;

[0276] Based on the calculated weights of each index, the normalized elastic energy ratio, dissipated energy ratio and acoustic emission energy are weighted to obtain the index values ​​of the preprocessed multi-index data set. The calculation formula is as follows:

[0277] X i =w i x i

[0278] Among them, X i is the weighted result of the i-th indicator, x i is the ith indicator.

[0279] In one embodiment, constructing a rock damage early warning model based on cusp catastrophe theory and preprocessed multi-indicator data sets includes:

[0280] Construct the discriminant function in the rock damage early warning model based on the cusp catastrophe theory;

[0281] Based on the preprocessed multi-index data set, the expression of the control factor in the discriminant function of the rock damage early warning model is constructed;

[0282] The maximum likelihood estimation method is used to calculate the linear transformation coefficient and constant term in the expression of the control factor in the discriminant function;

[0283] The rock damage early warning model is constructed based on the calculated linear transformation coefficient and constant term in the expression of the control factor in the discriminant function, the expression of the control factor in the discriminant function and the discriminant function.

[0284] It should be noted that the rock damage monitoring and early warning system based on multi-indicator fusion in the above embodiment has the same working principle and technical effect as the rock damage monitoring and early warning method based on multi-indicator fusion in the above embodiment, which will not be repeated here.

[0285] like Figure 5 As shown, the embodiment of the present application also provides a rock damage monitoring and early warning device based on multi-indicator fusion, which may include:

[0286] A memory 301, a processor 302, and a computer program 303 stored in the memory 301 and executable on the processor 302, wherein the memory 301 and the processor 302 communicate with each other via a bus 304, and when the processor 302 executes the computer program 303, the steps of the rock damage monitoring and early warning method based on multi-indicator fusion as in the above-mentioned method embodiment of the present application are implemented.

[0287] Specifically, the rock damage monitoring and early warning device 3 based on multi-index fusion can be an intelligent device with memory and processor such as an industrial computer, a PC, a smart mobile terminal, or a computer component with memory and processor such as a CPU or a GPU.

[0288] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the rock damage monitoring and early warning method based on multi-indicator fusion are implemented according to any one of the first objectives of the present application.

[0289] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0290] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0291] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0292] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. The present application will not be limited to the embodiments shown herein, but should conform to the broadest scope consistent with the principles and novel features disclosed herein.

Claims

1. A rock damage monitoring and early warning method based on multi-index fusion, characterized in that: The method comprises the following steps: S1, performing a simulated loading test on a rock specimen collected in a target monitoring environment, and collecting test data of multiple indicators of the rock specimen at each preset time to obtain a multi-indicator data set, wherein the indicators include elastic energy proportion, dissipated energy proportion and acoustic emission energy; S2, preprocessing the multi-indicator data set; S3, constructing a rock damage early warning model based on cusp catastrophe theory and preprocessed multi-index data sets; S4, performing rock damage monitoring and early warning on the target monitoring environment based on the rock damage early warning model.

2. The rock damage monitoring and early warning method based on multi-index fusion according to claim 1 is characterized in that: In step S1, the test data of multiple indicators of the rock specimen at each preset time is collected to obtain a multi-indicator data set including: S11, collecting the axial stress, axial strain, elastic modulus and acoustic emission energy of the rock specimen at each preset time during the simulated loading test; S12, based on the collected axial stress, axial strain and elastic modulus of the rock specimen at each preset time, calculate the elastic energy proportion and dissipated energy proportion of the rock specimen at each preset time according to a preset indicator data calculation model; S13, fusing the elastic energy proportion, dissipated energy proportion and acoustic emission energy at each preset moment to obtain the multi-index data set.

3. The rock damage monitoring and early warning method based on multi-index fusion according to claim 2 is characterized in that: In step S12, the indicator data calculation model is as follows: Among them, η is the elastic energy ratio, λ is the dissipated energy ratio, U is the total input energy, and U e is the elastic strain energy, U d is the dissipated energy; The calculation formula of the total input energy U is: Among them, ε is the axial strain and σ is the axial stress; Among them, E u is the tangent elastic modulus, E u ≈E0, E0 is the elastic modulus; Elastic strain energy U e The approximate calculation is: Then the dissipated energy U d The calculation formula is:

4. The rock damage monitoring and early warning method based on multi-index fusion according to claim 1 is characterized in that: In step S2, the preprocessing of the multi-index data set includes: S21, normalizing each test data of elastic energy ratio, dissipated energy ratio and acoustic emission energy in the multi-index data set by using a deviation normalization method; S22, the entropy weight method is used to weight the normalized elastic energy proportion, dissipated energy proportion and acoustic emission energy.

5. The rock damage monitoring and early warning method based on multi-index fusion according to claim 4 is characterized in that: In step S21, the calculation formula of the deviation standardization method is as follows: Among them, x' ij is the normalized index value, x ij is the original indicator value in the multi-indicator data set, indicating the value of the jth data of the ith indicator, where i=1,2,3, j=1,2,3…n, and n is a positive integer indicating the number of data of the indicator corresponding to the multi-indicator data set.

6. The rock damage monitoring and early warning method based on multi-index fusion according to claim 5 is characterized in that: In step S22, the entropy weight method weighting process is as follows: S221, normalize the normalized index value to obtain a standardized matrix, and the calculation formula is as follows: Among them, P ij is the standardized matrix; S222, calculate the entropy value of the i-th indicator, the formula is as follows: Among them, e i is the information entropy of the ith indicator, k is a constant; S223, calculate the information utility value of each indicator, the calculation formula is as follows: d i =1-e i Among them, d i is the information utility value of each indicator; S224, according to the information utility value of each indicator, determine the weight coefficient of each indicator, and the calculation formula is as follows: Among them, w i is the weight of the i-th indicator; S225, based on the calculated weights of each index, weight the normalized elastic energy ratio, dissipated energy ratio, and acoustic emission energy to obtain the index values ​​of the preprocessed multi-index data set, and the calculation formula is as follows: X i =w i x i Among them, X i is the weighted result of the i-th indicator, x i is the ith indicator.

7. The rock damage monitoring and early warning method based on multi-index fusion according to claim 6 is characterized in that: In step S3, the rock damage early warning model is constructed based on the cusp catastrophe theory and the preprocessed multi-index data set, including: S31, constructing the discriminant function in the rock damage early warning model based on the cusp catastrophe theory; S32, constructing the expression of the control factor in the discriminant function of the rock damage early warning model based on the preprocessed multi-index data set; S33, using the maximum likelihood estimation method to calculate the linear transformation coefficient and the constant term in the expression of the control factor in the discriminant function; S34, constructing the rock damage early warning model based on the calculated linear transformation coefficient and constant term in the expression of the control factor in the discriminant function, the expression of the control factor in the discriminant function and the discriminant function.

8. A rock damage monitoring and early warning system based on multi-index fusion, characterized in that: include: An indicator data set construction module is used to perform a simulated loading test on a rock specimen collected in a target monitoring environment, and collect test data of multiple indicators of the rock specimen at each preset time to obtain a multi-indicator data set, wherein the indicators include elastic energy proportion, dissipated energy proportion and acoustic emission energy; A data preprocessing module, used for preprocessing the multi-indicator data set; Model building module, used to build rock damage early warning model based on cusp catastrophe theory and preprocessed multi-indicator data set; The rock damage monitoring and early warning module is used to perform rock damage monitoring and early warning on the target monitoring environment based on the rock damage early warning model.

9. A rock damage monitoring and early warning device based on multi-index fusion, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the rock damage monitoring and early warning method based on multi-index fusion as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the rock damage monitoring and early warning method based on multi-index fusion as described in any one of claims 1 to 7 are implemented.

Images