Rock failure short-term and temporary early warning method and device based on acoustic emission clustering analysis

Abstract

The invention provides a rock failure short-term and temporary early warning method and device based on acoustic emission clustering analysis, and relates to the technical field of rock mechanics and geotechnical engineering. The method comprises the following steps: acquiring acoustic emission characteristic parameters in a rock deformation and failure process; carrying out clustering analysis on the acoustic emission characteristic parameters by adopting a Kmeans + + algorithm to obtain clustering labels corresponding to the acoustic emission characteristic parameters; according to the clustering labels, a random forest algorithm is adopted, and importance scores of the acoustic emission characteristic parameters are calculated; constructing an early warning index set according to the acoustic emission characteristic parameters; an initial CNN-LSTM model is constructed; according to the early warning index set, establishing a sample data set of the rock failure conventional signal and the precursor signal; the initial CNN-LSTM model is trained according to the sample data set, and a trained CNN-LSTM model is obtained; acquiring a real-time acoustic emission signal; and inputting a real-time acoustic emission signal into the trained CNN-LSTM model, and carrying out early warning by identifying the category of the acoustic emission signal. According to the invention, early warning can be carried out in the rock complex damage process.

Description

Technical Field

[0001] The present invention relates to the technical field of rock mechanics and geotechnical engineering, and in particular to a short-term early warning method and device for rock failure based on acoustic emission cluster analysis. Background Art

[0002] Mineral resources play an irreplaceable role in people's lives and social development. As mining progresses, engineering activities disrupt the original stress equilibrium of the rock mass, leading to stress concentrations that trigger the initiation and propagation of rock cracks. This can also lead to rock spalling, tunnel spalling, and roof collapse, creating ground pressure hazards. Especially in deep, high-stress environments, rockbursts in hard rock pose a serious threat to underground production safety. Therefore, research on precursor signals of rock failure and early warning technologies is crucial for ensuring safe mine production.

[0003] Acoustic emission (AE) technology has been widely used to study rock failure. Before rock failure, AE signals increase significantly. AE ring counts, energy, and event rates all begin to rise rapidly during the crack propagation phase. During this phase, AE signals reflect the localized release of strain energy stored within the rock, but cannot accurately identify warning points near macroscopic failure. Furthermore, using an abnormal increase in a single indicator as a warning is susceptible to interference from other abnormal AE signals in field environments. AE signal processing occurs only after rock failure, making it difficult to discern the overall trend of AE signals during actual failure. Consequently, there is a lack of effective nonparametric statistical methods and machine learning algorithms to analyze AE signal sequences and extract discriminative features. Existing research has proposed using relevant mathematical analysis theories and methods to assign weights to early warning indicators to achieve a comprehensive, multi-indicator early warning of rock failure. However, the weighting process involves a degree of subjectivity, resulting in low accuracy in early warning results. Existing methods use the changing trends of AE parameters to identify near-failure rock. However, before failure, the peak values of AE parameters are unclear, and the magnitude of data growth is difficult to measure, making accurate real-time identification impossible. Rock is a heterogeneous material, and its failure is inherently complex. Even for rocks of the same lithology, the macroscopic and microscopic manifestations of the failure process can differ significantly. Therefore, existing models struggle to accurately describe the physical mechanisms underlying complex rock failure processes and provide timely and effective early warnings. Summary of the Invention

[0004] To address the technical problems of existing technologies, such as the difficulty in accurately describing the physical mechanisms of complex rock failure processes and providing timely and effective early warnings, as well as the problem that existing technologies are easily interfered with by abnormal signals, resulting in low accuracy of early warning results, the present invention provides a method and device for short-term early warning of rock failure based on acoustic emission cluster analysis. The technical solution is as follows:

[0005] In one aspect, a short-term early warning method for rock failure based on acoustic emission cluster analysis is provided. The method is implemented by a short-term early warning device for rock failure based on acoustic emission cluster analysis. The method includes:

[0006] S1. Obtaining characteristic parameters of acoustic emission during rock deformation and failure;

[0007] S2. Performing cluster analysis on the acoustic emission characteristic parameters using the Kmeans++ algorithm to obtain cluster labels corresponding to the acoustic emission characteristic parameters;

[0008] S3. Calculate the importance scores of the acoustic emission characteristic parameters using a random forest algorithm based on the cluster labels; extract the acoustic emission characteristic parameters with an importance score greater than 0.1 and construct a set of early warning indicators;

[0009] S4. Build the initial CNN-LSTM dual-branch feature fusion model;

[0010] S5. Establish a sample data set of conventional signals and precursor signals of rock failure based on the set of early warning indicators; train an initial CNN-LSTM dual-branch feature fusion model based on the sample data set to obtain a trained CNN-LSTM dual-branch feature fusion model;

[0011] S6. Acquire real-time acoustic emission signals; input the real-time acoustic emission signals into the trained CNN-LSTM dual-branch feature fusion model for recognition to obtain recognition results, and use the acoustic emission signals with a recognition result of 1 as early warning signals of rock damage.

[0012] Optionally, the characteristic parameters of acoustic emission during the rock deformation and failure process include: acoustic emission event rate, number of events, ring count, cumulative ring count, energy, cumulative energy, rise time, duration, average rise time, average duration and average frequency.

[0013] Optionally, the step S2 uses a Kmeans++ algorithm to perform cluster analysis on the acoustic emission characteristic parameters to obtain cluster labels corresponding to the acoustic emission characteristic parameters, including:

[0014] S21, performing standardization processing on the acoustic emission characteristic parameters to obtain standardized acoustic emission characteristic parameters;

[0015] The normalization process is expressed by the following formula (1):

[0016] (1)

[0017] Among them, x i Indicates the original value; X min Represents the minimum value of the characteristic parameter X; Xmax Indicates the maximum value of the feature parameter X; min indicates the lower bound of the scaled data range; max indicates the upper bound of the scaled data range; where min is -1 and max is 1;

[0018] S22, calculating the sum of squares of the acoustic emission parameter clustering corresponding to different K values, and determining the optimal number of clusters;

[0019] The process of determining the optimal number of clusters is expressed by the following formula (2):

[0020] (2)

[0021] Where SSE represents the sum of squared errors; K represents the number of acoustic emission parameter classifications; C i represents the data point set of the nth category; x represents a data point in the set, representing the characteristic parameters of acoustic emission every 1 second, μ i Represents the set C i the center of

[0022] S23. Using the optimal number of clusters as the target category number of the Kmeans++ algorithm, cluster analysis is performed on the standardized acoustic emission characteristic parameters to obtain cluster labels corresponding to the acoustic emission characteristic parameters.

[0023] Optionally, the step S3 calculates the importance score of the acoustic emission characteristic parameter using a random forest algorithm according to the clustering label, including:

[0024] S31. Calculate the importance contribution of the acoustic emission characteristic parameters corresponding to each node in each decision tree using a random forest algorithm based on the acoustic emission characteristic parameters;

[0025] S32. Calculate the importance score of the acoustic emission characteristic parameter corresponding to the entire decision tree based on the importance contribution of the acoustic emission characteristic parameter of each node;

[0026] S33. Calculate the final importance scores of the acoustic emission characteristic parameters corresponding to all trees in the random forest based on the importance scores of the acoustic emission characteristic parameters corresponding to the entire decision tree.

[0027] Optionally, the process of calculating the importance contribution of the acoustic emission characteristic parameter corresponding to each node in each decision tree is expressed by the following formula (3):

[0028] (3)

[0029] in, Represents the importance contribution of acoustic emission feature parameter j in decision tree node n; I(parent) represents the impurity of the parent node; I(left child) represents the impurity of the left child node; I(right child) represents the impurity of the right child node;

[0030] The process of calculating the importance score of the acoustic emission characteristic parameters corresponding to the entire decision tree is expressed by the following formula (4):

[0031] (4)

[0032] in, represents the importance score of the acoustic emission feature parameter j in the entire decision tree; N represents the node set of the decision tree T;

[0033] The process of calculating the final importance scores of the acoustic emission characteristic parameters corresponding to all trees in the random forest is expressed by the following formula (5):

[0034] (5)

[0035] Among them, all trees in the random forest are represented as {T1, T2, ..., T m}; represents the average importance score of all trees in the random forest to the acoustic emission feature parameter j; m represents the number of decision trees in the random forest, which is gradually increased during the calculation process until the feature parameter importance ranking remains basically unchanged.

[0036] Optionally, the initial CNN-LSTM dual-branch feature fusion model of S4 includes: a CNN convolutional neural network and an LSTM bidirectional long short-term memory network, a feature splicing layer, a fully connected layer and an output layer;

[0037] The CNN convolutional neural network includes a convolutional layer, a batch normalization layer, a maximum pooling layer, and a flattening layer.

[0038] The convolutional layer has 64 filters, a convolution kernel size of 3, and an activation function of ReLU; the maximum pooling layer has a pooling window size of 2; and the LSTM bidirectional long short-term memory network consists of one bidirectional LSTM layer containing 64 units.

[0039] Wherein, the CNN convolutional neural network is used to extract local features of the feature matrix;

[0040] The LSTM bidirectional long short-term memory network is used to extract the global temporal features of the feature matrix.

[0041] Optionally, the step S5 of training the initial CNN-LSTM dual-branch feature fusion model according to the sample data set to obtain a trained CNN-LSTM dual-branch feature fusion model includes:

[0042] S51, inputting the sample data set into the initial CNN-LSTM dual-branch feature fusion model, performing parallel processing through the CNN convolutional neural network and the LSTM bidirectional long short-term memory network, performing local feature extraction through the convolution layer to obtain local feature representation; performing temporal feature processing through the LSTM bidirectional long short-term memory network to obtain temporal feature representation;

[0043] S52, inputting the local features into a batch normalization layer for processing to obtain a batch-normalized feature representation; inputting the batch-normalized feature representation into a maximum pooling layer, processing it through a flattening layer, and obtaining a one-dimensional vector feature representation;

[0044] S53, inputting the one-dimensional vector feature representation and the time series feature representation into the feature concatenation layer for feature concatenation to obtain a comprehensive feature representation;

[0045] S54. Input the comprehensive feature representation into the fully connected layer, perform nonlinear transformation on the comprehensive feature representation through the ReLU activation function to obtain the final feature representation; input the final feature representation into the output layer, use the Sigmoid activation function to map the final feature into a binary classification probability; according to the binary classification probability, use the binary cross entropy loss function for training to obtain the trained CNN-LSTM dual-branch feature fusion model.

[0046] On the other hand, a short-term early warning device for rock failure based on acoustic emission cluster analysis is provided. The device is applied to a short-term early warning method for rock failure based on acoustic emission cluster analysis. The device includes:

[0047] An acquisition unit, used to obtain characteristic parameters of acoustic emission during rock deformation and failure;

[0048] A cluster analysis unit, configured to perform cluster analysis on the acoustic emission characteristic parameters using a Kmeans++ algorithm to obtain cluster labels corresponding to the acoustic emission characteristic parameters;

[0049] A first construction unit is configured to calculate the importance scores of the acoustic emission characteristic parameters using a random forest algorithm based on the cluster labels; extract the acoustic emission characteristic parameters with an importance score greater than 0.1 to construct a warning indicator set;

[0050] The second construction unit is used to build the initial CNN-LSTM dual-branch feature fusion model;

[0051] A training unit is used to establish a sample data set of conventional signals and precursor signals of rock failure based on the early warning indicator set; train an initial CNN-LSTM dual-branch feature fusion model based on the sample data set to obtain a trained CNN-LSTM dual-branch feature fusion model;

[0052] The early warning unit is used to obtain real-time acoustic emission signals; the real-time acoustic emission signals are input into the trained CNN-LSTM dual-branch feature fusion model for recognition to obtain the recognition result, and the acoustic emission signal with the recognition result of 1 is used as the early warning signal of rock failure.

[0053] Optionally, the characteristic parameters of acoustic emission during the rock deformation and failure process include: acoustic emission event rate, number of events, ring count, cumulative ring count, energy, cumulative energy, rise time, duration, average rise time, average duration and average frequency.

[0054] Optionally, the cluster analysis unit is used to:

[0055] performing standardization processing on the acoustic emission characteristic parameters to obtain standardized acoustic emission characteristic parameters;

[0056] The normalization process is expressed by the following formula (1):

[0057] (1)

[0058] Among them, x i Indicates the original value; X min Represents the minimum value of the characteristic parameter X; X max Indicates the maximum value of the feature parameter X; min indicates the lower bound of the scaled data range; max indicates the upper bound of the scaled data range; where min is -1 and max is 1;

[0059] Calculate the sum of square errors of acoustic emission parameter clustering corresponding to different K values and determine the optimal number of clusters;

[0060] The process of determining the optimal number of clusters is expressed by the following formula (2):

[0061] (2)

[0062] Where SSE represents the sum of squared errors; K represents the number of acoustic emission parameter classifications; C i represents the data point set of the nth category; x represents a data point in the set, representing the characteristic parameters of acoustic emission every 1 second, μ i Represents the set C i the center of

[0063] The optimal number of clusters is used as the target category number of the Kmeans++ algorithm, and cluster analysis is performed on the standardized acoustic emission characteristic parameters to obtain the cluster labels corresponding to the acoustic emission characteristic parameters.

[0064] Optionally, the first building unit is used to:

[0065] According to the acoustic emission characteristic parameters, the random forest algorithm is used to calculate the importance contribution of the acoustic emission characteristic parameters corresponding to each node in each decision tree;

[0066] According to the importance contribution of the acoustic emission characteristic parameters of each node, the importance score of the acoustic emission characteristic parameters corresponding to the entire decision tree is calculated;

[0067] According to the importance score of the acoustic emission feature parameters corresponding to the entire decision tree, the final importance score of the acoustic emission feature parameters corresponding to all trees in the random forest is calculated.

[0068] Optionally, the process of calculating the importance contribution of the acoustic emission characteristic parameter corresponding to each node in each decision tree is expressed by the following formula (3):

[0069] (3)

[0070] in, Represents the importance contribution of acoustic emission feature parameter j in decision tree node n; I(parent) represents the impurity of the parent node; I(left child) represents the impurity of the left child node; I(right child) represents the impurity of the right child node;

[0071] The process of calculating the importance score of the acoustic emission characteristic parameters corresponding to the entire decision tree is expressed by the following formula (4):

[0072] (4)

[0073] in, represents the importance score of the acoustic emission feature parameter j in the entire decision tree; N represents the node set of the decision tree T;

[0074] The process of calculating the final importance scores of the acoustic emission characteristic parameters corresponding to all trees in the random forest is expressed by the following formula (5):

[0075] (5)

[0076] Among them, all trees in the random forest are represented as {T1, T2, ..., T m}; represents the average importance score of all trees in the random forest to the acoustic emission feature parameter j; m represents the number of decision trees in the random forest, which is gradually increased during the calculation process until the feature parameter importance ranking remains basically unchanged.

[0077] Optionally, the initial CNN-LSTM dual-branch feature fusion model includes: a CNN convolutional neural network and an LSTM bidirectional long short-term memory network, a feature splicing layer, a fully connected layer and an output layer;

[0078] The CNN convolutional neural network includes a convolutional layer, a batch normalization layer, a maximum pooling layer, and a flattening layer.

[0079] The convolutional layer has 64 filters, a convolution kernel size of 3, and an activation function of ReLU; the maximum pooling layer has a pooling window size of 2; and the LSTM bidirectional long short-term memory network consists of one bidirectional LSTM layer containing 64 units.

[0080] Wherein, the CNN convolutional neural network is used to extract local features of the feature matrix;

[0081] The LSTM bidirectional long short-term memory network is used to extract the global temporal features of the feature matrix.

[0082] Optionally, the training unit is used to:

[0083] The sample data set is input into the initial CNN-LSTM dual-branch feature fusion model, and is processed in parallel by the CNN convolutional neural network and the LSTM bidirectional long short-term memory network. Local features are extracted through the convolution layer to obtain local feature representation; and time series feature processing is performed through the LSTM bidirectional long short-term memory network to obtain time series feature representation;

[0084] The local features are input into the batch normalization layer for processing to obtain the batch normalized feature representation; the batch normalized feature representation is input into the maximum pooling layer and processed through the flattening layer to obtain the feature representation of a one-dimensional vector;

[0085] The one-dimensional vector feature representation and the time series feature representation are input into the feature concatenation layer for feature concatenation to obtain a comprehensive feature representation;

[0086] The comprehensive feature representation is input into the fully connected layer, and the comprehensive feature representation is nonlinearly transformed through the ReLU activation function to obtain the final feature representation; the final feature representation is input into the output layer, and the Sigmoid activation function is used to map the final feature into a binary classification probability; based on the binary classification probability, the binary cross entropy loss function is used for training to obtain the trained CNN-LSTM dual-branch feature fusion model.

[0087] On the other hand, a short-term warning device for rock failure based on acoustic emission cluster analysis is provided. The short-term warning device for rock failure based on acoustic emission cluster analysis includes: a processor; a memory, wherein computer-readable instructions are stored on the memory. When the computer-readable instructions are executed by the processor, any one of the above-mentioned short-term warning methods for rock failure based on acoustic emission cluster analysis is implemented.

[0088] On the other hand, a computer-readable storage medium is provided, wherein the storage medium stores at least one instruction, and the at least one instruction is loaded and executed by a processor to implement any of the above-mentioned methods for short-term early warning of rock failure based on acoustic emission cluster analysis.

[0089] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0090] The embodiment of the present invention first obtains acoustic emission characteristic parameters during rock deformation and failure; uses the Kmeans++ algorithm to perform cluster analysis on the acoustic emission characteristic parameters to obtain cluster labels corresponding to the acoustic emission characteristic parameters; uses the random forest algorithm to calculate the importance scores of the acoustic emission characteristic parameters based on the cluster labels; extracts acoustic emission characteristic parameters with importance scores greater than 0.1 to construct an early warning indicator set; secondly, constructs an initial CNN-LSTM dual-branch feature fusion model; establishes a sample data set of conventional signals and precursor signals of rock failure based on the early warning indicator set; trains the initial CNN-LSTM dual-branch feature fusion model based on the sample data set to obtain a trained CNN-LSTM dual-branch feature fusion model; finally, obtains a real-time acoustic emission signal; inputs the real-time acoustic emission signal into the trained CNN-LSTM dual-branch feature fusion model for recognition to obtain a recognition result, and uses the acoustic emission signal with a recognition result of 1 as a warning signal of rock failure.

[0091] The embodiment of the present invention predicts rock failure through multiple acoustic emission parameters, and can automatically identify acoustic emission precursor signals of rock failure and provide early warning. The implementation of the present invention can accurately describe the physical mechanism of complex rock failure processes and provide timely and effective early warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0093] Figure 1 This is a flow chart of a short-term early warning method for rock failure based on acoustic emission cluster analysis provided by an embodiment of the present invention;

[0094] Figure 2 1 is a schematic diagram of a rock failure stress-strain curve and acoustic emission signal clustering results under uniaxial loading provided by an embodiment of the present invention;

[0095] Figure 3 This is a schematic diagram of calculating the sum of squared errors for different cluster numbers K and finding the optimal cluster number K using the elbow method, provided by an embodiment of the present invention;

[0096] Figure 4 This is a schematic diagram of an embodiment of the present invention providing a method for calculating the contribution of each acoustic emission characteristic parameter to a signal category using a random forest algorithm to obtain an importance score of each parameter;

[0097] Figure 5 This is a schematic diagram of a process for performing cluster analysis using the Kmeans++ algorithm to obtain signal category labels, provided by an embodiment of the present invention;

[0098] Figure 6 This is a structural diagram of a CNN-LSTM dual-branch feature fusion model provided by an embodiment of the present invention;

[0099] Figure 7 This is a schematic diagram of an early warning method for identifying acoustic emission signals during the entire rock deformation and failure process using a trained CNN-LSTM dual-branch feature fusion model, provided by an embodiment of the present invention;

[0100] Figure 8 This is a block diagram of a rock failure short-term warning device based on acoustic emission cluster analysis provided by an embodiment of the present invention;

[0101] Figure 9 The present invention provides a schematic structural diagram of a short-term early warning device for rock failure based on acoustic emission cluster analysis. DETAILED DESCRIPTION

[0102] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0103] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as an "exemplary" in the present invention should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of the word "exemplary" is intended to present concepts in a concrete manner. Furthermore, in the embodiments of the present invention, "and / or" can mean both or either of the two.

[0104] In the embodiments of the present invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same. The terms "of," "corresponding," and "corresponding" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same.

[0105] In the embodiments of the present invention, sometimes a subscript such as W1 may be written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are the same.

[0106] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0107] The embodiment of the present invention provides a rock failure short-term warning method based on acoustic emission cluster analysis. The method can be implemented by a rock failure short-term warning device based on acoustic emission cluster analysis. The rock failure short-term warning device based on acoustic emission cluster analysis can be a terminal or a server. Figure 1 The flowchart of the rock failure short-term early warning method based on acoustic emission cluster analysis is shown. The processing flow of this method may include the following steps:

[0108] S1. Obtain the characteristic parameters of acoustic emission during rock deformation and failure.

[0109] In a feasible implementation method, a limestone sample with a specification of Φ50×100 mm is subjected to a single-axis compression test under a displacement control mode with a loading rate of 0.002 mm / s. An acoustic emission sensor is used to monitor the deformation and failure process in real time, and the acoustic emission signal is recorded at a sampling rate of 1 MHz to obtain the acoustic emission characteristic parameters every 1 second.

[0110] Optionally, the characteristic parameters of acoustic emission during rock deformation and failure include: acoustic emission event rate, number of events, ring count, cumulative ring count, energy, cumulative energy, rise time, duration, average rise time, average duration and average frequency.

[0111] The acquired acoustic emission characteristic parameters are constructed into a matrix, and each row of the matrix corresponds to the acoustic emission characteristic parameters of each second. The Kmeans++ algorithm is used to perform cluster analysis on the acoustic emission characteristic parameters collected during the entire limestone destruction process. Each row of the matrix returns a signal category, and the acoustic emission signal category of each second is obtained, such as Figure 2 FIG2 is a schematic diagram of a rock failure stress-strain curve under uniaxial loading and acoustic emission signal clustering results provided by an embodiment of the present invention.

[0112] Among them, in the cluster analysis process, it is necessary to determine the optimal cluster K value of cluster analysis. The initial K value interval is set to (1,10). By calculating the sum of squared errors corresponding to different K values, a curve of the sum of squared errors changing with the K value is drawn; when the sum of squared errors does not decrease significantly with the increase of K value, the corresponding K value is the optimal cluster, such as Figure 3 FIG. 1 is a schematic diagram of calculating the sum of squared errors for different cluster numbers K provided by an embodiment of the present invention and finding the optimal cluster number K using the elbow method. The K value of the limestone sample in the embodiment of the present invention is calculated to be 4.

[0113] S2. Use Kmeans++ algorithm to perform cluster analysis on acoustic emission characteristic parameters to obtain cluster labels corresponding to the acoustic emission characteristic parameters.

[0114] Optionally, the specific implementation process of S2 includes S21-S23:

[0115] S21, performing standardization processing on the acoustic emission characteristic parameters to obtain standardized acoustic emission characteristic parameters;

[0116] The normalization process is expressed by the following formula (1):

[0117] (1)

[0118] Among them, x i Indicates the original value; X min Represents the minimum value of the characteristic parameter X; X max Indicates the maximum value of the feature parameter X; min indicates the lower bound of the scaled data range; max indicates the upper bound of the scaled data range; where min is -1 and max is 1;

[0119] S22, calculating the sum of squares of the acoustic emission parameter clustering corresponding to different K values, and determining the optimal number of clusters;

[0120] The process of determining the optimal number of clusters is expressed by the following formula (2):

[0121] (2)

[0122] Where SSE represents the sum of squared errors; K represents the number of acoustic emission parameter classifications; C i represents the data point set of the nth category; x represents a data point in the set, representing the characteristic parameters of acoustic emission every 1 second, μ i Represents the set C i the center of

[0123] Among them, when K increases, the number of data points in each category decreases, so the distance between the data points in the set and the center of the set will decrease, resulting in a decrease in the sum of squared errors; when the sum of squared errors does not decrease significantly with the increase of K value, the corresponding K value is the optimal number of categories for cluster analysis of acoustic emission characteristic parameters.

[0124] S23. Using the optimal number of clusters as the target category number of the Kmeans++ algorithm, cluster analysis is performed on the standardized acoustic emission characteristic parameters to obtain cluster labels corresponding to the acoustic emission characteristic parameters.

[0125] Among them, such as Figure 4 This is a schematic diagram of an embodiment of the present invention, which uses a random forest algorithm to calculate the contribution of each acoustic emission characteristic parameter to a signal category and obtain the importance score of each parameter.

[0126] S3. According to the clustering labels, the random forest algorithm is used to calculate the importance scores of the acoustic emission characteristic parameters; the acoustic emission characteristic parameters with an importance score greater than 0.1 are extracted to construct a set of early warning indicators.

[0127] Optionally, S3 uses a random forest algorithm based on cluster labels to calculate the importance scores of acoustic emission characteristic parameters, including:

[0128] Among them, the random forest algorithm takes the acquired acoustic emission characteristic parameters as input and takes the cluster labels corresponding to the acquired acoustic emission characteristic parameters every 1 second as output.

[0129] S31. Calculate the importance contribution of the acoustic emission characteristic parameters corresponding to each node in each decision tree using a random forest algorithm based on the acoustic emission characteristic parameters;

[0130] The importance score of each acoustic emission feature parameter is the cumulative impurity reduction of the parameter on all nodes of all decision trees in the random forest.

[0131] Taking a decision tree T as an example, the node set of the decision tree is N; for each node n∈N, assuming that the node n is split using the acoustic emission characteristic parameter j, the importance contribution of the acoustic emission characteristic parameter corresponding to each node is obtained.

[0132] Optionally, the process of calculating the importance contribution of the acoustic emission characteristic parameter corresponding to each node in each decision tree is expressed by the following formula (3):

[0133] (3)

[0134] in, Represents the importance contribution of acoustic emission feature parameter j in decision tree node n; I(parent) represents the impurity of the parent node; I(left child) represents the impurity of the left child node; I(right child) represents the impurity of the right child node;

[0135] S32. Calculate the importance score of the acoustic emission characteristic parameter corresponding to the entire decision tree based on the importance contribution of the acoustic emission characteristic parameter of each node;

[0136] The process of calculating the importance score of the acoustic emission characteristic parameters corresponding to the entire decision tree is expressed by the following formula (4):

[0137] (4)

[0138] in, represents the importance score of the acoustic emission feature parameter j in the entire decision tree; N represents the node set of the decision tree T;

[0139] S33. Calculate the final importance scores of the acoustic emission characteristic parameters corresponding to all trees in the random forest based on the importance scores of the acoustic emission characteristic parameters corresponding to the entire decision tree.

[0140] The process of calculating the final importance scores of the acoustic emission characteristic parameters corresponding to all trees in the random forest is expressed by the following formula (5):

[0141] (5)

[0142] Among them, all trees in the random forest are represented as {T1, T2, ..., T m}; represents the average importance score of all trees in the random forest for the acoustic emission feature parameter j; m represents the number of decision trees in the random forest, which is gradually increased during the calculation process until the feature parameter importance ranking remains basically unchanged.

[0143] S4. Build the initial CNN-LSTM dual-branch feature fusion model.

[0144] Among them, such as Figure 6 The figure shows a structural diagram of a CNN-LSTM dual-branch feature fusion model provided by an embodiment of the present invention.

[0145] Optionally, the initial CNN-LSTM dual-branch feature fusion model of S4 includes: a CNN convolutional neural network and an LSTM bidirectional long short-term memory network, a feature concatenation layer, a fully connected layer, and an output layer;

[0146] Among them, the CNN convolutional neural network includes: convolution layer, batch normalization layer, maximum pooling layer and flattening layer;

[0147] The convolutional layer has 64 filters, a convolution kernel size of 3, and an activation function of ReLU. The maximum pooling layer has a pooling window size of 2. The LSTM bidirectional long short-term memory network consists of one bidirectional LSTM layer with 64 units.

[0148] Among them, CNN convolutional neural network is used to extract local features of the feature matrix;

[0149] Among them, the LSTM bidirectional long short-term memory network is used to extract the global temporal features of the feature matrix.

[0150] Among them, the fully connected layer is a fully connected layer containing 128 neurons.

[0151] Among them, the output layer contains one neuron.

[0152] S5. Based on the early warning indicator set, a sample data set of conventional signals and precursor signals of rock failure is established; the initial CNN-LSTM dual-branch feature fusion model is trained based on the sample data set to obtain a trained CNN-LSTM dual-branch feature fusion model.

[0153] Among them, Figure 5 FIG2 is a schematic diagram of a process of performing cluster analysis using the Kmeans++ algorithm to obtain signal category labels, provided by an embodiment of the present invention.

[0154] In a feasible implementation, the process of establishing a sample data set of conventional signals and precursor signals of rock failure based on the early warning indicator set includes:

[0155] (1) The early warning indicator set is transformed into a feature matrix, and the feature matrix is re-clustered using the Kmeans++ algorithm to obtain the clustering result; the classification label corresponding to each row of the feature matrix in the clustering result is saved to form a note column;

[0156] (2) The peak stress point of the stress-strain curve is taken as the rock failure point, and a type of signal near the failure point is taken as the rock failure precursor signal, which is recorded as 1, and the rest of the signals are conventional signals, which are recorded as 0; the results are saved in the feature matrix and recorded as label columns, forming a sample data set of conventional signals and precursor signals in the rock failure process.

[0157] Among them, the established sample data set of conventional signals and precursor signals of rock failure is divided into training set, validation set and test set. The training set accounts for 0.6 of the total samples, the validation set accounts for 0.2 of the total samples, and the test set accounts for 0.2 of the total samples. The training set is used to train the CNN-LSTM two-branch feature fusion model. When the loss rate of the validation set does not decrease after 10 cumulative iterations, the model parameters with the lowest loss of the validation set are returned and the optimal model is saved. The performance of the CNN-LSTM two-branch feature fusion model on the test set is observed.

[0158] Optionally, S5 trains the initial CNN-LSTM dual-branch feature fusion model according to the sample data set, and the process of obtaining the trained CNN-LSTM dual-branch feature fusion model includes S51-S54:

[0159] S51. Input the sample data set into the initial CNN-LSTM dual-branch feature fusion model, perform parallel processing through the CNN convolutional neural network and the LSTM bidirectional long short-term memory network, extract local features through the convolution layer to obtain local feature representation; perform temporal feature processing through the LSTM bidirectional long short-term memory network to obtain temporal feature representation;

[0160] S52, inputting the local features into a batch normalization layer for processing to obtain a batch-normalized feature representation; inputting the batch-normalized feature representation into a maximum pooling layer, processing it through a flattening layer, and obtaining a one-dimensional vector feature representation;

[0161] S53, inputting the one-dimensional vector feature representation and the time series feature representation into the feature concatenation layer for feature concatenation to obtain a comprehensive feature representation;

[0162] Among them, the feature splicing layer is used to splice and fuse the features extracted by the CNN branch and the LSTM branch. This method not only preserves the independence of the two features, avoids information loss caused by early fusion, but also provides a richer joint representation for subsequent processing.

[0163] S54. Input the comprehensive feature representation into the fully connected layer, perform nonlinear transformation on the comprehensive feature representation through the ReLU activation function to obtain the final feature representation; input the final feature representation into the output layer, use the Sigmoid activation function to map the final feature into a binary classification probability; according to the binary classification probability, use the binary cross entropy loss function for training to obtain the trained CNN-LSTM dual-branch feature fusion model.

[0164] Among them, the binary classification output corresponds to the rock failure state represented by the acoustic emission signal parameters.

[0165] Among them, a Dropout layer with a ratio of 0.5 is added after the fully connected layer, and 50% of the neuron outputs are randomly discarded during the training stage to reduce the risk of overfitting during model training.

[0166] Among them, the fully connected layer has 128 neurons, and the activation function uses ReLU, which can perform nonlinear transformation and dimensionality reduction on the concatenated high-order features.

[0167] In a feasible implementation, a binary cross entropy loss function is used, an Adam optimizer is used for gradient descent, the number of iterations is set to 50, the learning rate is dynamically adjusted during training, and an early stopping mechanism is set to prevent overfitting. In this embodiment of the present invention, the validation set loss is used as a monitoring indicator, and when the validation set loss does not decrease during 5 consecutive iterative trainings, the learning rate adjustment is expressed by the following formula (6):

[0168] (6)

[0169] in, is the current learning rate; is the learning rate of the previous stage; f is the learning rate attenuation coefficient, which is set to 0.2 in the embodiment of the present invention.

[0170] The early stopping counter threshold is set to 10. When the cumulative validation set loss does not decrease during 10 iterations of training, the early stopping mechanism is triggered, the model stops training, and the model weight with the lowest validation set loss in the entire training process is returned.

[0171] S6. Acquire real-time acoustic emission signals; input the real-time acoustic emission signals into the trained CNN-LSTM dual-branch feature fusion model for recognition to obtain recognition results, and use the acoustic emission signals with a recognition result of 1 as early warning signals of rock damage.

[0172] In one feasible implementation, the real-time acoustic emission signal is input into a trained CNN-LSTM dual-branch feature fusion model to obtain a predicted probability. When the predicted probability is less than 0.5, the corresponding acoustic emission signal category is 0, which is a conventional signal of rock fracture, indicating that the rock has not entered the stage of aggravated damage and monitoring and analysis continue. When the predicted probability is ≥ 0.5, the corresponding acoustic emission signal category is 1, which is a precursor signal of rock damage, indicating that the rock is close to damage and an immediate warning is issued.

[0173] Among them, such as Figure 7The figure shows an early warning schematic diagram of an embodiment of the present invention, which uses a trained CNN-LSTM dual-branch feature fusion model to identify acoustic emission signals throughout the entire process of rock deformation and failure. The rock failure precursor signal calibrated according to the clustering results appears around 214 seconds, the model predicts a danger signal at 218 seconds, and the rock failure time point is 258 seconds. The model warning time lags 4 seconds behind the precursor signal but is nearly 40 seconds ahead of the rock failure time. The method provided by the present invention realizes early warning and prediction of rock failure.

[0174] The embodiment of the present invention first obtains acoustic emission characteristic parameters during rock deformation and failure; uses the Kmeans++ algorithm to perform cluster analysis on the acoustic emission characteristic parameters to obtain cluster labels corresponding to the acoustic emission characteristic parameters; uses the random forest algorithm to calculate the importance scores of the acoustic emission characteristic parameters based on the cluster labels; extracts acoustic emission characteristic parameters with importance scores greater than 0.1 to construct an early warning indicator set; secondly, constructs an initial CNN-LSTM dual-branch feature fusion model; establishes a sample data set of conventional signals and precursor signals of rock failure based on the early warning indicator set; trains the initial CNN-LSTM dual-branch feature fusion model based on the sample data set to obtain a trained CNN-LSTM dual-branch feature fusion model; finally, obtains a real-time acoustic emission signal; inputs the real-time acoustic emission signal into the trained CNN-LSTM dual-branch feature fusion model for recognition to obtain a recognition result, and uses the acoustic emission signal with a recognition result of 1 as a warning signal of rock failure.

[0175] The embodiment of the present invention predicts rock failure through multiple acoustic emission parameters, and can automatically identify acoustic emission precursor signals of rock failure and provide early warning. The implementation of the present invention can accurately describe the physical mechanism of complex rock failure processes and provide timely and effective early warning.

[0176] Figure 8 This is a block diagram of a rock failure short-term warning device based on acoustic emission cluster analysis according to an exemplary embodiment. The device is used in a rock failure short-term warning method based on acoustic emission cluster analysis. Figure 8 The device includes an acquisition unit 810, a cluster analysis unit 820, a first construction unit 830, a second construction unit 840, a training unit 850, and an early warning unit 860.

[0177] An acquisition unit 810 is used to acquire characteristic parameters of acoustic emission during the rock deformation and failure process;

[0178] A cluster analysis unit 820 is configured to perform cluster analysis on the acoustic emission characteristic parameters using a Kmeans++ algorithm to obtain cluster labels corresponding to the acoustic emission characteristic parameters;

[0179] A first construction unit 830 is configured to calculate the importance scores of the acoustic emission characteristic parameters using a random forest algorithm based on the cluster labels; extract the acoustic emission characteristic parameters with importance scores greater than 0.1 to construct a warning indicator set;

[0180] The second construction unit 840 is used to construct an initial CNN-LSTM dual-branch feature fusion model;

[0181] The training unit 850 is configured to establish a sample data set of conventional rock failure signals and precursor signals based on the early warning indicator set; train an initial CNN-LSTM dual-branch feature fusion model based on the sample data set to obtain a trained CNN-LSTM dual-branch feature fusion model;

[0182] The early warning unit 860 is used to obtain real-time acoustic emission signals; input the real-time acoustic emission signals into the trained CNN-LSTM dual-branch feature fusion model for recognition, obtain a recognition result, and use the acoustic emission signal with a recognition result of 1 as a warning signal of rock failure.

[0183] Optionally, the characteristic parameters of acoustic emission during the rock deformation and failure process include: acoustic emission event rate, number of events, ring count, cumulative ring count, energy, cumulative energy, rise time, duration, average rise time, average duration and average frequency.

[0184] Optionally, the cluster analysis unit 820 is configured to:

[0185] performing standardization processing on the acoustic emission characteristic parameters to obtain standardized acoustic emission characteristic parameters;

[0186] The normalization process is expressed by the following formula (1):

[0187] (1)

[0188] Among them, x i Indicates the original value; X min Represents the minimum value of the characteristic parameter X; X max Indicates the maximum value of the feature parameter X; min indicates the lower bound of the scaled data range; max indicates the upper bound of the scaled data range; where min is -1 and max is 1;

[0189] Calculate the sum of square errors of acoustic emission parameter clustering corresponding to different K values and determine the optimal number of clusters;

[0190] The process of determining the optimal number of clusters is expressed by the following formula (2):

[0191] (2)

[0192] Where SSE represents the sum of squared errors; K represents the number of acoustic emission parameter classifications; C i represents the data point set of the nth category; x represents a data point in the set, representing the characteristic parameters of acoustic emission every 1 second, μ i Represents the set C i the center of

[0193] The optimal number of clusters is used as the target category number of the Kmeans++ algorithm, and cluster analysis is performed on the standardized acoustic emission characteristic parameters to obtain the cluster labels corresponding to the acoustic emission characteristic parameters.

[0194] Optionally, the first constructing unit 830 is configured to:

[0195] According to the acoustic emission characteristic parameters, the random forest algorithm is used to calculate the importance contribution of the acoustic emission characteristic parameters corresponding to each node in each decision tree;

[0196] According to the importance contribution of the acoustic emission characteristic parameters of each node, the importance score of the acoustic emission characteristic parameters corresponding to the entire decision tree is calculated;

[0197] According to the importance score of the acoustic emission feature parameters corresponding to the entire decision tree, the final importance score of the acoustic emission feature parameters corresponding to all trees in the random forest is calculated.

[0198] Optionally, the process of calculating the importance contribution of the acoustic emission characteristic parameter corresponding to each node in each decision tree is expressed by the following formula (3):

[0199] (3)

[0200] in, Represents the importance contribution of acoustic emission feature parameter j in decision tree node n; I(parent) represents the impurity of the parent node; I(left child) represents the impurity of the left child node; I(right child) represents the impurity of the right child node;

[0201] The process of calculating the importance score of the acoustic emission characteristic parameters corresponding to the entire decision tree is expressed by the following formula (4):

[0202] (4)

[0203] in, represents the importance score of the acoustic emission feature parameter j in the entire decision tree; N represents the node set of the decision tree T;

[0204] The process of calculating the final importance scores of the acoustic emission characteristic parameters corresponding to all trees in the random forest is expressed by the following formula (5):

[0205] (5)

[0206] Among them, all trees in the random forest are represented as {T1, T2, ..., T m}; represents the average importance score of all trees in the random forest for the acoustic emission feature parameter j; m represents the number of decision trees in the random forest, which is gradually increased during the calculation process until the feature parameter importance ranking remains basically unchanged.

[0207] Optionally, the initial CNN-LSTM dual-branch feature fusion model includes: a CNN convolutional neural network and an LSTM bidirectional long short-term memory network, a feature splicing layer, a fully connected layer and an output layer;

[0208] The CNN convolutional neural network includes a convolutional layer, a batch normalization layer, a maximum pooling layer, and a flattening layer.

[0209] The convolutional layer has 64 filters, a convolution kernel size of 3, and an activation function of ReLU; the maximum pooling layer has a pooling window size of 2; and the LSTM bidirectional long short-term memory network consists of one bidirectional LSTM layer containing 64 units.

[0210] Wherein, the CNN convolutional neural network is used to extract local features of the feature matrix;

[0211] The LSTM bidirectional long short-term memory network is used to extract the global temporal features of the feature matrix.

[0212] Optionally, the training unit 850 is configured to:

[0213] The sample data set is input into the initial CNN-LSTM dual-branch feature fusion model, and is processed in parallel by the CNN convolutional neural network and the LSTM bidirectional long short-term memory network. Local features are extracted through the convolution layer to obtain local feature representation; and time series feature processing is performed through the LSTM bidirectional long short-term memory network to obtain time series feature representation;

[0214] The local features are input into the batch normalization layer for processing to obtain the batch normalized feature representation; the batch normalized feature representation is input into the maximum pooling layer and processed through the flattening layer to obtain the feature representation of a one-dimensional vector;

[0215] The one-dimensional vector feature representation and the time series feature representation are input into the feature concatenation layer for feature concatenation to obtain a comprehensive feature representation;

[0216] The comprehensive feature representation is input into the fully connected layer, and the comprehensive feature representation is nonlinearly transformed through the ReLU activation function to obtain the final feature representation; the final feature representation is input into the output layer, and the Sigmoid activation function is used to map the final feature into a binary classification probability; based on the binary classification probability, the binary cross entropy loss function is used for training to obtain the trained CNN-LSTM dual-branch feature fusion model.

[0217] The embodiment of the present invention first obtains acoustic emission characteristic parameters during rock deformation and failure; uses the Kmeans++ algorithm to perform cluster analysis on the acoustic emission characteristic parameters to obtain cluster labels corresponding to the acoustic emission characteristic parameters; uses the random forest algorithm to calculate the importance scores of the acoustic emission characteristic parameters based on the cluster labels; extracts acoustic emission characteristic parameters with importance scores greater than 0.1 to construct an early warning indicator set; secondly, constructs an initial CNN-LSTM dual-branch feature fusion model; establishes a sample data set of conventional signals and precursor signals of rock failure based on the early warning indicator set; trains the initial CNN-LSTM dual-branch feature fusion model based on the sample data set to obtain a trained CNN-LSTM dual-branch feature fusion model; finally, obtains a real-time acoustic emission signal; inputs the real-time acoustic emission signal into the trained CNN-LSTM dual-branch feature fusion model for recognition to obtain a recognition result, and uses the acoustic emission signal with a recognition result of 1 as a warning signal of rock failure.

[0218] The embodiment of the present invention predicts rock failure through multiple acoustic emission parameters, and can automatically identify acoustic emission precursor signals of rock failure and provide early warning. The implementation of the present invention can accurately describe the physical mechanism of complex rock failure processes and provide timely and effective early warning.

[0219] Figure 9 FIG. 1 is a schematic structural diagram of a rock failure short-term warning device based on acoustic emission cluster analysis provided by an embodiment of the present invention. Figure 9 As shown, the rock failure short-term warning equipment based on acoustic emission cluster analysis can include the above Figure 8 The rock failure short-term warning device based on acoustic emission cluster analysis is shown. Optionally, the rock failure short-term warning device based on acoustic emission cluster analysis 910 may include a first processor 2001 .

[0220] Optionally, the rock failure short-term warning device 910 based on acoustic emission cluster analysis may further include a memory 2002 and a transceiver 2003 .

[0221] The first processor 2001, the memory 2002 and the transceiver 2003 may be connected via a communication bus.

[0222] The following combination Figure 9The following is a detailed introduction to the various components of the rock failure short-term warning device 910 based on acoustic emission cluster analysis:

[0223] The first processor 2001 is the control center of the rock failure short-term warning device 910 based on acoustic emission cluster analysis, and can be a single processor or a collective term for multiple processing elements. For example, the first processor 2001 can be one or more central processing units (CPUs), or application-specific integrated circuits (ASICs), or one or more integrated circuits configured to implement embodiments of the present invention, such as one or more digital signal processors (DSPs) or one or more field programmable gate arrays (FPGAs).

[0224] Optionally, the first processor 2001 can execute various functions of the rock damage short-term warning device 910 based on acoustic emission cluster analysis by running or executing a software program stored in the memory 2002 and calling data stored in the memory 2002.

[0225] In a specific implementation, as an embodiment, the first processor 2001 may include one or more CPUs, such as Figure 9 CPU0 and CPU1 are shown in FIG.

[0226] In a specific implementation, as an embodiment, the rock failure short-term warning device 910 based on acoustic emission cluster analysis may also include multiple processors, such as Figure 9 1 and 2. The first processor 2001 and the second processor 2004 are shown in FIG. Each of these processors can be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). A processor herein can refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0227] The memory 2002 is used to store the software program for executing the solution of the present invention, and is controlled by the first processor 2001 for execution. The specific implementation method can refer to the above method embodiment and will not be repeated here.

[0228] Alternatively, the memory 2002 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, a random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, an optical disc storage (including a compact disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and capable of being accessed by a computer, but not limited thereto. The memory 2002 may be integrated with the first processor 2001 or exist independently and be accessed through the interface circuit ( Figure 9 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0229] The transceiver 2003 is used to communicate with a network device or a terminal device.

[0230] Optionally, the transceiver 2003 may include a receiver and a transmitter ( Figure 9 The receiver is used to implement a receiving function, and the transmitter is used to implement a sending function.

[0231] Optionally, the transceiver 2003 may be integrated with the first processor 2001 or may exist independently and be connected to the interface circuit of the rock damage short-term warning device 910 based on acoustic emission cluster analysis ( Figure 9 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0232] It should be noted that Figure 9 The structure of the rock failure short-term warning device 910 based on acoustic emission cluster analysis shown in the figure does not constitute a limitation on the router. The actual knowledge structure recognition device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0233] In addition, the technical effects of the rock damage short-term warning device 910 based on acoustic emission cluster analysis can refer to the technical effects of the rock damage short-term warning method based on acoustic emission cluster analysis described in the above method embodiment, which will not be repeated here.

[0234] It should be understood that the first processor 2001 in the embodiment of the present invention may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor, or the processor may be any conventional processor, etc.

[0235] It should also be understood that the memory in the embodiments of the present invention may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory may be random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0236] The above embodiments can be implemented in whole or in part via software, hardware (e.g., circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product comprises one or more computer instructions or computer programs. When loaded or executed on a computer, the processes or functions described in accordance with the embodiments of the present invention are fully or partially performed. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired means (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server or data center that contains a collection of one or more available media. The available medium can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0237] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the associated objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0238] In this disclosure, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, "at least one of a, b, or c" can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or plural.

[0239] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0240] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0241] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0242] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interface, indirect coupling or communication connection of the device or unit, which can be electrical, mechanical or other forms.

[0243] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0244] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0245] If the functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or the portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage media include various media that can store program code, such as USB flash drives, mobile hard drives, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical disks.

[0246] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A short-term early warning method for rock failure based on acoustic emission cluster analysis, characterized in that: The method comprises: S1. Obtaining characteristic parameters of acoustic emission during rock deformation and failure; S2. Performing cluster analysis on the acoustic emission characteristic parameters using the Kmeans++ algorithm to obtain cluster labels corresponding to the acoustic emission characteristic parameters; S3. Calculate the importance scores of the acoustic emission characteristic parameters using a random forest algorithm based on the cluster labels; extract the acoustic emission characteristic parameters with an importance score greater than 0.1 and construct a set of early warning indicators; S4. Build the initial CNN-LSTM dual-branch feature fusion model; S5. Establish a sample data set of conventional signals and precursor signals of rock failure based on the set of early warning indicators; train an initial CNN-LSTM dual-branch feature fusion model based on the sample data set to obtain a trained CNN-LSTM dual-branch feature fusion model; S6. Acquire real-time acoustic emission signals; input the real-time acoustic emission signals into the trained CNN-LSTM dual-branch feature fusion model for recognition to obtain recognition results; and use the acoustic emission signals with a recognition result of 1 as early warning signals of rock damage.

2. The rock failure short-term warning method based on acoustic emission cluster analysis according to claim 1 is characterized in that: The acoustic emission characteristic parameters during the rock deformation and failure process include: acoustic emission event rate, event number, ring count, cumulative ring count, energy, cumulative energy, rise time, duration, average rise time, average duration and average frequency.

3. The rock failure short-term early warning method based on acoustic emission cluster analysis according to claim 1 is characterized in that: The S2 uses the Kmeans++ algorithm to perform cluster analysis on the acoustic emission characteristic parameters to obtain cluster labels corresponding to the acoustic emission characteristic parameters, including: S21, performing standardization processing on the acoustic emission characteristic parameters to obtain standardized acoustic emission characteristic parameters; The normalization process is expressed by the following formula (1): （1） Among them, x i Indicates the original value; X min Represents the minimum value of the characteristic parameter X; X max Indicates the maximum value of the feature parameter X; min indicates the lower bound of the scaled data range; max indicates the upper bound of the scaled data range; where min is -1 and max is 1; S22, calculating the sum of squares of the acoustic emission parameter clustering corresponding to different K values, and determining the optimal number of clusters; The process of determining the optimal number of clusters is expressed by the following formula (2): （2） Where SSE represents the sum of squared errors; K represents the number of acoustic emission parameter classifications; C i represents the data point set of the nth category; x represents a data point in the set, representing the characteristic parameters of acoustic emission every 1 second, μ i Represents the set C i the center of S23. Using the optimal number of clusters as the target category number of the Kmeans++ algorithm, cluster analysis is performed on the standardized acoustic emission characteristic parameters to obtain cluster labels corresponding to the acoustic emission characteristic parameters.

4. The rock failure short-term warning method based on acoustic emission cluster analysis according to claim 1 is characterized in that: The step S3 calculates the importance score of the acoustic emission characteristic parameter using a random forest algorithm based on the cluster labels, including: S31. Calculate the importance contribution of the acoustic emission characteristic parameters corresponding to each node in each decision tree using a random forest algorithm based on the acoustic emission characteristic parameters; S32. Calculate the importance score of the acoustic emission characteristic parameter corresponding to the entire decision tree based on the importance contribution of the acoustic emission characteristic parameter of each node; S33. Calculate the final importance scores of the acoustic emission characteristic parameters corresponding to all trees in the random forest based on the importance scores of the acoustic emission characteristic parameters corresponding to the entire decision tree.

5. The rock failure short-term warning method based on acoustic emission cluster analysis according to claim 4 is characterized in that: The process of calculating the importance contribution of the acoustic emission characteristic parameters corresponding to each node in each decision tree is expressed by the following formula (3): （3） in, Represents the importance contribution of acoustic emission feature parameter j in decision tree node n; I(parent) represents the impurity of the parent node; I(left child) represents the impurity of the left child node; I(right child) represents the impurity of the right child node; The process of calculating the importance score of the acoustic emission characteristic parameters corresponding to the entire decision tree is expressed by the following formula (4): （4） in, represents the importance score of the acoustic emission feature parameter j in the entire decision tree; N represents the node set of the decision tree T; The process of calculating the final importance scores of the acoustic emission characteristic parameters corresponding to all trees in the random forest is expressed by the following formula (5): （5） Among them, all trees in the random forest are represented as {T1, T2, ..., T m }; represents the average importance score of all trees in the random forest to the acoustic emission feature parameter j; m represents the number of decision trees in the random forest, which is gradually increased during the calculation process until the feature parameter importance ranking remains basically unchanged.

6. The rock failure short-term warning method based on acoustic emission cluster analysis according to claim 1 is characterized in that: The initial CNN-LSTM dual-branch feature fusion model of S4 includes: a CNN convolutional neural network and an LSTM bidirectional long short-term memory network, a feature splicing layer, a fully connected layer, and an output layer; The CNN convolutional neural network includes a convolutional layer, a batch normalization layer, a maximum pooling layer, and a flattening layer. The convolutional layer has 64 filters, a convolution kernel size of 3, and an activation function of ReLU; the maximum pooling layer has a pooling window size of 2; and the LSTM bidirectional long short-term memory network consists of one bidirectional LSTM layer containing 64 units. Wherein, the CNN convolutional neural network is used to extract local features of the feature matrix; The LSTM bidirectional long short-term memory network is used to extract the global temporal features of the feature matrix.

7. The rock failure short-term warning method based on acoustic emission cluster analysis according to claim 1 is characterized in that: The step S5 trains the initial CNN-LSTM dual-branch feature fusion model according to the sample data set to obtain a trained CNN-LSTM dual-branch feature fusion model, including: S51, inputting the sample data set into the initial CNN-LSTM dual-branch feature fusion model, performing parallel processing through the CNN convolutional neural network and the LSTM bidirectional long short-term memory network, performing local feature extraction through the convolution layer to obtain local feature representation; performing temporal feature processing through the LSTM bidirectional long short-term memory network to obtain temporal feature representation; S52, inputting the local features into a batch normalization layer for processing to obtain a batch-normalized feature representation; inputting the batch-normalized feature representation into a maximum pooling layer, processing it through a flattening layer, and obtaining a one-dimensional vector feature representation; S53, inputting the one-dimensional vector feature representation and the time series feature representation into the feature concatenation layer for feature concatenation to obtain a comprehensive feature representation; S54. Input the comprehensive feature representation into the fully connected layer, perform nonlinear transformation on the comprehensive feature representation through the ReLU activation function to obtain the final feature representation; input the final feature representation into the output layer, use the Sigmoid activation function to map the final feature into a binary classification probability; according to the binary classification probability, use the binary cross entropy loss function for training to obtain the trained CNN-LSTM dual-branch feature fusion model.

8. A short-term warning device for rock failure based on acoustic emission cluster analysis, wherein the short-term warning device for rock failure based on acoustic emission cluster analysis is used to implement the short-term warning method for rock failure based on acoustic emission cluster analysis according to any one of claims 1 to 7, characterized in that: The device comprises: An acquisition unit, used to obtain characteristic parameters of acoustic emission during rock deformation and failure; A cluster analysis unit, configured to perform cluster analysis on the acoustic emission characteristic parameters using a Kmeans++ algorithm to obtain cluster labels corresponding to the acoustic emission characteristic parameters; A first construction unit is configured to calculate the importance scores of the acoustic emission characteristic parameters using a random forest algorithm based on the cluster labels; extract the acoustic emission characteristic parameters with an importance score greater than 0.1 to construct a warning indicator set; The second construction unit is used to build the initial CNN-LSTM dual-branch feature fusion model; A training unit is used to establish a sample data set of conventional signals and precursor signals of rock failure based on the early warning indicator set; train an initial CNN-LSTM dual-branch feature fusion model based on the sample data set to obtain a trained CNN-LSTM dual-branch feature fusion model; The early warning unit is used to obtain real-time acoustic emission signals; the real-time acoustic emission signals are input into the trained CNN-LSTM dual-branch feature fusion model for recognition to obtain the recognition results, and the acoustic emission signals with the recognition result of 1 are used as early warning signals of rock failure.

9. A short-term early warning device for rock failure based on acoustic emission cluster analysis, characterized in that: The rock failure short-term warning device based on acoustic emission cluster analysis includes: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program code, which can be called by a processor to execute the method according to any one of claims 1 to 7.

Images