Mining micro-seismic signal intelligent identification method and system

Abstract

The invention discloses a mining micro-seismic signal intelligent identification method and system, and belongs to the technical field of data processing.The mining micro-seismic signal intelligent identification method comprises the steps that a micro-seismic monitoring signal to be identified is acquired to serve as a signal to be analyzed; decomposing the to-be-analyzed signal into a plurality of single-type signals by adopting a preset algorithm; performing feature extraction on the decomposed single type of signal to obtain feature data of the signal to be analyzed; and inputting the feature data of the to-be-analyzed signal into a preset recognition model to realize intelligent recognition of the micro-seismic signal. By adopting the mining micro-seismic signal intelligent identification method and system, guidance can be provided for mine micro-seismic intelligent identification, so that safety production of a mine is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and system for intelligently identifying microseismic signals for mining. Background Art

[0002] In recent years, with the development of the mining industry and the increase in mining depth, mine safety issues have become increasingly prominent. Microseismic signal monitoring, as an important means of earthquake prediction and safety assurance, has been widely used in mine safety management. Rapid, accurate, and intelligent identification of mine microseismic signals is crucial for ensuring safe mine production. However, due to the complex mining environment, the signals collected by microseismic monitoring systems often contain a large amount of interference, making intelligent identification of microseismic signals very difficult. Therefore, the development of an efficient and accurate method for intelligent identification of microseismic signals is particularly important.

[0003] Microseismic signals often exhibit nonlinear and nonstationary characteristics, making them difficult to effectively process using traditional linear analysis methods. Traditional methods assume stationary and linear signals, but in reality, microseismic signals often contain complex transient characteristics and nonlinear components. Furthermore, because the geological conditions and mining environments of different mines can vary significantly, a single, unchanging microseismic signal intelligent recognition method may be difficult to adapt to these diverse conditions, making it impossible to maintain high-precision intelligent recognition of microseismic signals across diverse mining environments.

[0004] Therefore, how to make the microseismic intelligent identification method have both high intelligent identification accuracy and gradually adapt to the geological conditions and mining environment of different mines through continuous learning and optimization is crucial for geological research during mining and ensuring safe mine production. Summary of the Invention

[0005] The present invention provides a method and system for intelligent recognition of microseismic signals for mining, so as to solve the technical problem that the prior art cannot maintain high-precision intelligent recognition of microseismic signals in different mining environments.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In one aspect, the present invention provides a method for intelligently identifying microseismic signals for mining, comprising:

[0008] Acquire the microseismic monitoring signal to be identified as the signal to be analyzed;

[0009] Decompose the signal to be analyzed into several single-type signals;

[0010] Perform feature extraction on the decomposed single type of signal to obtain feature data of the signal to be analyzed;

[0011] The characteristic data of the signal to be analyzed is input into the preset recognition model to realize intelligent recognition of microseismic signals.

[0012] Furthermore, decomposing the signal to be analyzed into a plurality of single-type signals includes:

[0013] Step 1: Perform discrete Fourier transform on the signal to obtain the frequency domain representation S[k] as the residual spectrum;

[0014] Step 2: Find the frequency f corresponding to the maximum amplitude in the residual spectrum. max ;

[0015] Step 3, set the harmonic order m = 1, 2, ..., M max , for each possible m, calculate its candidate fundamental frequency f m =f max / m, and solve its harmonic energy function value E(f m ); where the expression of the harmonic energy function is:

[0016]

[0017] Where M is the maximum number of harmonics; round() means rounding to the nearest integer; R[] represents the residual spectrum; f s is the sampling frequency; N is the total number of sampling points of the signal;

[0018] Step 4: Determine the candidate fundamental frequency f corresponding to the maximum harmonic energy function value c-max , construct the optimization function, and obtain the final fundamental frequency f0 by solving the optimization function; the optimization function expression is:

[0019]

[0020] Among them, E(f c-max ) represents f c-max The corresponding harmonic energy function value;

[0021] Step 5: Determine the harmonics m=1, 2, ..., M corresponding to f0 and calculate the frequency k corresponding to each harmonic m :

[0022]

[0023] Step 6: Construct the spectrum S of a single signal i [k], while removing S from the residual spectrum i [k], update the residual spectrum; where the spectrum S of a single signal is constructed i The process of [k] is as follows:

[0024] First, create a spectrum S with the same length as S[k]i [k], and initialize all its values to 0; for each harmonic frequency index, extract the corresponding value from S[k] and assign it to S i [k]; then, let S i [Nk m ]=S[Nk m ], except S i [k m ] and S i [Nk m ], all other positions of S i [k] is 0;

[0025] Step 7, based on the spectrum of the constructed signal, reconstruct a single signal by inverse discrete Fourier transform;

[0026] Step 8, repeating steps 1 to 7 several times to obtain multiple reconstructed signals of a single type;

[0027] Step 9: Return to step 2 and stop iterating until the maximum amplitude of the residual spectrum is lower than a preset threshold or the expected number of signals are extracted.

[0028] Furthermore, the feature extraction of the decomposed single type of signal to obtain feature data of the signal to be analyzed includes:

[0029] Divide the signal into multiple time windows of fixed length and calculate the statistics of the signal in each time window; the statistics include the mean, variance, skewness and kurtosis of the signal in the corresponding time window;

[0030] Calculate the difference in statistics between adjacent time windows;

[0031] The difference between the statistics of every two adjacent time windows is combined into a feature vector, and all the obtained feature vectors are used as the feature data of the signal to be analyzed.

[0032] Furthermore, the preset recognition model is a BP neural network model optimized by a genetic algorithm.

[0033] Furthermore, the training process of the preset recognition model includes:

[0034] Obtain historical microseismic monitoring signals as sample signals and label them according to their categories;

[0035] Decomposing the sample signal into several single-type signals, and extracting features from the decomposed single-type signals to obtain feature data of the sample signal;

[0036] Using the feature data and label data of the sample signal, a feature database is constructed;

[0037] The feature database is used to train the preset recognition model, and the genetic algorithm is used to optimize the weight and threshold parameters of the BP neural network to enhance the learning ability and generalization ability of the neural network.

[0038] On the other hand, the present invention also provides a mining microseismic signal intelligent recognition system, comprising:

[0039] A data acquisition module is used to acquire the microseismic monitoring signal to be identified as a signal to be analyzed;

[0040] The data processing module is used to decompose the signal to be analyzed into several single-type signals; and extract features from the decomposed single-type signals to obtain feature data of the signal to be analyzed;

[0041] The intelligent recognition module is used to input the characteristic data of the signal to be analyzed obtained by the data processing module into a preset recognition model to realize intelligent recognition of microseismic signals.

[0042] Furthermore, decomposing the signal to be analyzed into a plurality of single-type signals includes:

[0043] Step 1: Perform discrete Fourier transform on the signal to obtain the frequency domain representation S[k] as the residual spectrum;

[0044] Step 2: Find the frequency f corresponding to the maximum amplitude in the residual spectrum. max ;

[0045] Step 3, set the harmonic order m = 1, 2, ..., M max , for each possible m, calculate its candidate fundamental frequency f m =f max / m, and solve its harmonic energy function value E(f m ); where the expression of the harmonic energy function is:

[0046]

[0047] Where M is the maximum number of harmonics; round() means rounding to the nearest integer; R[] represents the residual spectrum; f s is the sampling frequency; N is the total number of sampling points of the signal;

[0048] Step 4: Determine the candidate fundamental frequency f corresponding to the maximum harmonic energy function value c-max , construct the optimization function, and obtain the final fundamental frequency f0 by solving the optimization function; the optimization function expression is:

[0049]

[0050] Among them, E(f c-max) represents f c-max The corresponding harmonic energy function value;

[0051] Step 5: Determine the harmonics m=1, 2, ..., M corresponding to f0 and calculate the frequency k corresponding to each harmonic m :

[0052]

[0053] Step 6: Construct the spectrum S of a single signal i [k], while removing S from the residual spectrum i [k], update the residual spectrum; where the spectrum S of a single signal is constructed i The process of [k] is as follows:

[0054] First, create a spectrum S with the same length as S[k] i [k], and initialize all its values to 0; for each harmonic frequency index, extract the corresponding value from S[k] and assign it to S i [k]; then, let S i [Nk m ]=S[Nk m ], except S i [k m ] and S i [Nk m ], all other positions of S i [k] is 0;

[0055] Step 7, based on the spectrum of the constructed signal, reconstruct a single signal by inverse discrete Fourier transform;

[0056] Step 8, repeating steps 1 to 7 several times to obtain multiple reconstructed signals of a single type;

[0057] Step 9: Return to step 2 and stop iterating until the maximum amplitude of the residual spectrum is lower than a preset threshold or the expected number of signals are extracted.

[0058] Furthermore, the feature extraction of the decomposed single type of signal to obtain feature data of the signal to be analyzed includes:

[0059] Divide the signal into multiple time windows of fixed length and calculate the statistics of the signal in each time window; the statistics include the mean, variance, skewness and kurtosis of the signal in the corresponding time window;

[0060] Calculate the difference in statistics between adjacent time windows;

[0061] The difference between the statistics of every two adjacent time windows is combined into a feature vector, and all the obtained feature vectors are used as the feature data of the signal to be analyzed.

[0062] Furthermore, the preset recognition model is a BP neural network model optimized by a genetic algorithm.

[0063] Furthermore, the training process of the preset recognition model includes:

[0064] Obtain historical microseismic monitoring signals as sample signals and label them according to their categories;

[0065] Decomposing the sample signal into several single-type signals, and extracting features from the decomposed single-type signals to obtain feature data of the sample signal;

[0066] Using the feature data and label data of the sample signal, a feature database is constructed;

[0067] The feature database is used to train the preset recognition model, and the genetic algorithm is used to optimize the weight and threshold parameters of the BP neural network to enhance the learning ability and generalization ability of the neural network.

[0068] On the other hand, the present invention further provides an electronic device, comprising a processor and a memory; wherein the memory stores at least one instruction, and the instruction is loaded and executed by the processor to implement the above method.

[0069] In yet another aspect, the present invention further provides a computer-readable storage medium, wherein the storage medium stores at least one instruction, and the instruction is loaded and executed by a processor to implement the above method.

[0070] The beneficial effects brought about by the technical solution provided by the present invention include at least:

[0071] The present invention obtains the microseismic monitoring signal to be identified as the signal to be analyzed; decomposes the signal to be analyzed into several single-type signals; extracts features from the decomposed signals to obtain characteristic data of the signal to be analyzed; and inputs the characteristic data of the signal to be analyzed into a preset recognition model. This allows for high-precision microseismic identification with low computational resource consumption. Furthermore, through continuous learning and optimization, the system can gradually adapt to different mining environments, thus possessing the advantages of wide applicability. It can provide guidance for microseismic identification in mines and ensure safe mine production. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] 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.

[0073] Figure 1 This is a flow chart of a method for intelligent identification of microseismic signals for mining provided by an embodiment of the present invention;

[0074] Figure 2 This is the original signal time domain waveform and frequency domain diagram provided by the embodiment of the present invention;

[0075] Figure 3 This is a time domain waveform and frequency domain diagram of the spectrum comparison diagram of each sound signal after the original signal is decomposed by the "iterative harmonic extraction method" provided by the embodiment of the present invention;

[0076] Figure 4 This is a diagram showing the recognition results of the training set and test set after extracting feature vectors based on local statistics and dynamic differences and using them as neural network inputs, as provided by an embodiment of the present invention;

[0077] Figure 5 is a neural network training performance graph provided by an embodiment of the present invention;

[0078] Figure 6 is a diagram of an output microseismic signal provided by an embodiment of the present invention; wherein (a) to (f) represent different output microseismic signals respectively;

[0079] Figure 7 This is a system block diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0080] To make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0081] First, it should be noted that 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 "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "exemplarily" is intended to present concepts in a concrete manner. In addition, in the embodiments of the present invention, the meaning of "and / or" can be both or either of the two.

[0082] First embodiment

[0083] This embodiment provides a method for intelligent identification of microseismic signals for mining, which can be implemented by an electronic device, which can be a terminal or a server. Figure 1 As shown, the following steps are included:

[0084] S1, obtaining the microseismic monitoring signal to be identified as the signal to be analyzed;

[0085] S2, decomposing the signal to be analyzed into several single-type signals;

[0086] It should be noted that the composite sound signal s(t) can be regarded as multiple single sound signals s i (t), that is, s(t) = s1(t) + s2(t) + ... + s n (t), assuming that each single sound signal is a harmonic signal with a fundamental frequency f 0i and its harmonic component kf 0i (where k = 1, 2, 3, ...).

[0087] Based on the above, this embodiment uses the "iterative harmonic extraction method" to decompose the composite sound signal into several single-type sound signals, thereby constructing a sound signal database. Specifically, the following steps are included:

[0088] S21, perform discrete Fourier transform DFT on the composite sound signal to obtain the frequency domain representation:

[0089]

[0090] Among them, S[k] is the complex amplitude of the composite sound signal at frequency index k after discrete Fourier transform, which is the representation of the signal in the frequency domain; s[n] is the sampled composite sound signal, n is the time index, the value range is n=0,1,...,N-1, representing the discrete signal sample in the time domain; N is the total number of sampling points of the signal, that is, the length of the signal; k is the frequency index, the value range is k=0,1,...N-1, corresponding to the specific frequency component, the actual frequency is f k =kf s / N, where f s is the sampling frequency; j is the imaginary unit, defined as Used to represent the imaginary part of a complex number; e -j2πkn / N It is the basis function of DFT, a complex exponential function that represents the rotation factor at frequency k and is used to project the time domain signal into the frequency domain; is a summation operation over all time indices n, computing the total contribution of the signal at frequency k.

[0091] S22, setting the frequency domain signal as the initial residual spectrum, finding the frequency corresponding to the maximum value (i.e., the significant frequency) in the amplitude of the residual spectrum; and estimating the possible fundamental frequency based on this frequency; the specific process is as follows:

[0092] First initialize the residual spectrum R[k]:

[0093] R[k]=S[k]

[0094] In the amplitude |R[k]| of the residual spectrum R[k], find the frequency f corresponding to the maximum value max :

[0095] f max =k max f s / N

[0096] Among them, k max is the frequency position with the strongest energy in the spectrum when the residual spectrum amplitude reaches the maximum value, and its value range is k=0,1,...,N-1; f s is the sampling frequency, in Hertz (Hz), which represents the number of samples collected per second.

[0097] f max It is likely to be a harmonic component of a fundamental frequency, so the possible fundamental frequency is estimated based on this frequency.

[0098] S23, estimate the fundamental frequency and calculate the harmonic energy: Assume f max =mf0, where m is the harmonic order, m=1,2,...,M max , M max It is the maximum possible value of the harmonic order, and is usually taken as M in experience. max =5; for each possible m, calculate its candidate fundamental frequency f m =f max / m, and solve its harmonic energy function value E(f m ), calculate the sum of the energies of the first several harmonics in the residual spectrum; the expression of the harmonic energy function is:

[0099]

[0100] Among them, E(f m ) is f m The corresponding harmonic energy function value represents the total energy of the harmonic components of the corresponding fundamental frequency in the residual spectrum, which is used to evaluate the significance of a certain fundamental frequency; f mis the candidate fundamental frequency, that is, the assumed fundamental frequency of the signal, in Hertz (Hz); M is the maximum number of harmonics considered, for example: M = 5 means calculating the energy of the first 5 harmonics; m is the harmonic number, ranging from m = 1, 2, ..., M, indicating the number of harmonics (for example, m = 1 is the fundamental frequency, m = 2 is the double frequency); R[] represents the residual spectrum, which is initially equal to S[k] (DFT result) and is updated in each iteration to indicate the remaining unextracted spectral components; round(x) is a rounding function that rounds the input value x to the nearest integer to find the frequency index closest to the harmonic frequency; Indicates the approximate position of the frequency index corresponding to the mth harmonic frequency, where N / f s is the time length; |R[k]| 2 is the energy of the residual spectrum at frequency index k, calculated as the square of the complex magnitude.

[0101] The harmonic energy function value corresponding to each k can be calculated through the above formula;

[0102] S24, determining k and candidate fundamental frequency f corresponding to the maximum harmonic energy function value c-max , construct the optimization function, and obtain the final fundamental frequency f0 by solving the optimization function; the optimization function expression is:

[0103]

[0104] in, Indicates that among all possible f c-max Among the values, find the value that makes E(f fc-max )The largest f c-max .

[0105] In this embodiment, the method for solving the above optimization function is as follows:

[0106] f c-max As a starting point, use numerical optimization methods such as gradient ascent or solve at f c-max Numerical optimization methods such as grid search in a small nearby area are used to solve the harmonic energy equation to maximize the harmonic energy. Since the harmonic energy function is a function of the discrete spectrum, its gradient can be approximated by numerical differentiation.

[0107] After obtaining the final fundamental frequency, the frequency indices corresponding to the first several harmonics are calculated to construct the spectrum of a single signal. The spectrum is then converted back to the time domain using an inverse discrete Fourier transform (IDFT) to reconstruct the single signal. The extracted harmonic components are then removed from the residual spectrum to complete the separation of an independent signal and update the residual for the next iteration. The specific implementation process is as follows:

[0108] S25, after determining f0, extract its harmonic set. For each harmonic m=1,2,...,M, calculate the corresponding frequency k m :

[0109]

[0110] Among them, k m is the frequency index corresponding to the mth harmonic, which is used to locate the harmonic position in the spectrum; It is to calculate the exact frequency index position corresponding to the mth harmonic, which is usually a decimal and is rounded to an integer.

[0111] S26, construct the spectrum S of a single signal i [k];

[0112] First create a spectrum S with the same length as S[k] i [k], and initialize all values to 0. For each harmonic frequency index k m (m=1,2,...,M) extract the corresponding value from the original spectrum S[k] and assign it to S i [k]: S i [k m ]=S[k m ]; Considering the conjugate symmetry of the Fourier transform spectrum, let S i [Nk m ]=S[Nk m ], except for the above S i [k m ] and S i [Nk m ], all other positions of S i [k] is set to 0.

[0113] S27, based on the signal spectrum, reconstructs a single signal s by inverse discrete Fourier transform i [n]:

[0114]

[0115] Among them, s i [n] is the i-th extracted single sound signal, n is the time index, the value range is n=0,1,...,N-1, representing the reconstructed time domain signal; s i [k] is the spectrum of the i-th signal, which usually has values only at the harmonic frequency index and is 0 at other positions; e j2πkn / N It is the basis function of the inverse discrete Fourier transform (IDFT), which is a complex exponential function used to convert the frequency domain signal back to the time domain; Sum all frequency indices k to complete the reconstruction from frequency domain to time domain; is a normalization factor used to ensure that the magnitude of the inverse transform is correct.

[0116] S28, repeating S21 to S27 several times, gradually extracting all main sound components in the composite signal through multiple iterations, and completing the separation of each sound signal.

[0117] In step S29, return to step S22 and use the updated residual spectrum R[k] to continue searching for the next significant frequency, repeating steps S23 and S28. The iteration stops when the maximum amplitude of the residual spectrum falls below a preset threshold or the expected number of signals has been extracted. This iterative process ensures that the method can adaptively decompose complex signals of varying complexity, ultimately completing the separation of single sound signals and constructing a single sound signal database.

[0118] S3, extracting features from the decomposed single type of signal to obtain feature data of the signal to be analyzed;

[0119] Specifically, in this embodiment, a feature database is extracted from the sound signal database obtained in S2 by extracting feature vectors based on local statistics and dynamic differences, and used as the input of the neural network to improve the neural network's ability to recognize data. The specific steps are as follows:

[0120] S31, divide the sound signal into multiple time windows of fixed time length (for example, each window contains 100 sampling points). In each time window, calculate the following statistics: mean (reflecting the average strength of the signal in the window), variance (reflecting the degree of signal fluctuation), skewness (reflecting the symmetry of the signal distribution), and kurtosis (reflecting the sharpness of the signal peak). Then calculate the difference in statistics between adjacent time windows, for example: mean difference (reflecting the change in the mean of adjacent windows); variance difference (reflecting the change in the variance of adjacent windows);

[0121] S32, combining the statistics (mean, variance, skewness, kurtosis) of each window and the differences between adjacent windows into a feature vector, and obtaining a feature database corresponding to the signal to be analyzed as the input of the network.

[0122] S4, inputting the characteristic data of the signal to be analyzed into a preset recognition model to realize intelligent recognition of microseismic signals.

[0123] Among them, the preset recognition model adopted in this embodiment is the BP neural network model GA-BP optimized by genetic algorithm. The BP neural network consists of an input layer, a hidden layer (also called an intermediate layer) and an output layer, wherein the hidden layer has one or more layers. Each layer can have several nodes. The connection status of the nodes between layers is reflected by weights. First, forward propagation is performed: after the data is input from the input end, it is multiplied by the corresponding weight along the direction of the network and then added. The result is then used as input to calculate in the activation function, and the calculated result is passed as input to the next node; calculation is performed in sequence until the final result is obtained. Then back propagation is performed: the output result is compared with the expected output result, and the error generated by the comparison is back propagated through the network. Through multiple iterations, the weights between each node on the network are continuously adjusted (updated).

[0124] The activation function can be selected from sigmoid function, Tanh function, ReLU function, etc. If the sigmoid function is selected, its expression is:

[0125]

[0126] Here, x represents the linear combination of input features, specifically x=w·z+b, where w is the weight vector, z is the input feature vector, and b is the bias term.

[0127] There is often an error between the output result after forward propagation and the expected output result. The loss function is used to measure the size of the error:

[0128]

[0129] In the above formula, y i The expected output result for item i; is the actual output result of the i-th item; n is the total number of samples in the training data set.

[0130] The weight adjustment (update) can be iterated using the gradient descent method to minimize the loss function. Other algorithms can also be used to update the weights, such as mini-batch gradient descent, Adam, etc. If the gradient descent method is used, its algorithm expression is:

[0131]

[0132] In the above formula, W (t) is the parameter after the tth update; η is the learning rate, which controls the step size of each parameter update; is the gradient of the loss function with respect to the parameter w; α[W (t) -W (t-1) ] is a smoothing term.

[0133] The training process of the recognition model includes:

[0134] Obtain historical microseismic monitoring signals as sample signals and label them according to their categories;

[0135] The "iterative harmonic extraction method" is used to decompose the sample signal into several single-type sound signals, and the feature vectors are extracted based on "local statistics and dynamic differences" to form the feature data of the sample signal;

[0136] Using the feature data and label data of the sample signal, a feature database is constructed;

[0137] The feature database is used to train the preset recognition model, and the genetic algorithm is used to optimize the weight and threshold parameters of the BP neural network, thereby enhancing the learning ability and generalization ability of the neural network and improving the accuracy.

[0138] The basic idea behind genetic algorithms for optimizing BP neural networks is to use individuals to represent the network's initial weights and thresholds, taking the norm of the BP neural network's test error for the predicted sample as the output of the objective function, and then calculating the fitness of that individual. Through selection, crossover, and mutation operations, the optimal individual, i.e., the optimal initial weights and thresholds for the BP neural network, is found. In addition to genetic algorithms, particle swarm optimization and ant colony optimization can also be used to optimize the initial weights and thresholds of BP neural networks.

[0139] The following is an example of using the method of the present invention to perform intelligent identification of microseismic signals:

[0140] Two days of data from a mine were randomly selected as a composite sound signal set. After complex manual identification, it contained 18 rock microseismic signals, 34 electrical noise signals, 28 blasting signals and 14 other types of signals. Figure 2 The time domain waveform and frequency domain image of the original signal are shown in Figure 2. After the BP neural network model is trained using the manually labeled previous mine microseismic data, the mine data is intelligently identified using microseismic signals according to the above steps S1 to S3. Figure 3 The time domain waveform and frequency domain diagram of the spectrum comparison diagram of each sound signal after the original signal is decomposed by the "iterative harmonic extraction method". Figure 4 The training set and test set recognition results are shown after the feature vector is extracted based on local statistics and dynamic differences and used as the input of the neural network. Figure 5 Graph of neural network training performance. Figure 6This is the microseismic signal output after the method of this invention. Analysis of the output results shows that this intelligent identification process output a total of 21 signals, including 18 rock mass microseismic signals, with an accuracy rate of 85.7%. This demonstrates the high reliability and practicality of this microseismic signal intelligent identification model. In practical applications, this intelligent recognition rate can effectively help personnel quickly and intelligently identify and analyze microseismic signals, thereby making timely decisions and responses.

[0141] In summary, this embodiment provides a method for intelligently identifying microseismic signals for use in mines. This method can intelligently identify microseismic signals with high precision while gradually adapting to different mining environments through continuous learning and optimization. This method can provide guidance for intelligent microseismic identification in mines and ensure safe production in mines. It has the advantage of a wide range of applications.

[0142] Second embodiment

[0143] This embodiment provides a system for intelligently identifying microseismic signals for mining, including the following modules:

[0144] A data acquisition module is used to acquire the microseismic monitoring signal to be identified as a signal to be analyzed;

[0145] The data processing module is used to decompose the signal to be analyzed into several single-type signals; and extract features from the decomposed single-type signals to obtain feature data of the signal to be analyzed;

[0146] The intelligent recognition module is used to input the characteristic data of the signal to be analyzed obtained by the data processing module into a preset recognition model to realize intelligent recognition of microseismic signals.

[0147] It should be noted that, for the sake of convenience, the intelligent identification system for microseismic signals used in mines of this embodiment corresponds to the intelligent identification method for microseismic signals used in mines of the first embodiment; the functions implemented by each functional module in the intelligent identification system for microseismic signals used in mines of this embodiment correspond one-to-one to each process step in the intelligent identification method for microseismic signals used in mines of the first embodiment; therefore, they will not be repeated here.

[0148] Third embodiment

[0149] This embodiment provides an electronic device, such as Figure 3 As shown, the electronic device includes: at least one processor 71, at least one memory 72, at least one network interface 74 and other user interfaces 73. The various components in the evaluation device are coupled together via a bus system 75. It is understood that the bus system 75 is used to achieve connection and communication between these components. In addition to the data bus, the bus system 75 also includes a power bus, a control bus and a status signal bus. However, for the sake of clarity, Figure 7Various buses are labeled as bus system 75.

[0150] The user interface 73 may include a display, a keyboard, or a pointing device (eg, a mouse, a trackball, or a touchpad).

[0151] It is understood that the memory 72 in this embodiment can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct RAM bus random access memory (DRRAM). The memory 72 described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0152] In some embodiments, the memory 72 stores the following elements, executable units or data structures, or a subset thereof, or an extended set thereof: an operating system 721 and application programs 722 .

[0153] The operating system 721 includes various system programs, such as a framework layer, a core library layer, and a driver layer, for implementing various basic services and handling hardware-based tasks. Application programs 722 include various applications, such as an industrial control equipment operation and management system, for implementing various application services. Programs implementing the methods of the embodiments of the present invention may be included in application programs 722.

[0154] In the embodiment of the present invention, the processor 71 calls the program or instruction stored in the memory 72, specifically, the program or instruction stored in the application 722, and the processor 71 is used to execute the method steps provided in the first embodiment.

[0155] The methods disclosed in the above embodiments of the present invention can be applied to or implemented by processor 71. Processor 71 may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by hardware integrated logic circuits in processor 71 or by software instructions. The above processor 71 may be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The methods, steps, and logic block diagrams disclosed in the embodiments of the present invention can be implemented or executed. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in conjunction with the embodiments of the present invention can be directly implemented and executed by a hardware decoding processor, or by a combination of hardware and software units in the decoding processor. The software units can be located in storage media well-known in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory 72 , and the processor 71 reads the information in the memory 72 and completes the steps of the above method in combination with its hardware.

[0156] In addition, it should be noted that Figure 7 The structure of the electronic device shown in the figure does not constitute a limitation on the device. The actual device may include more or fewer components than shown, or may combine certain components, or arrange the components differently. In addition, the technical effects achieved by the electronic device when executing the method of the first embodiment can refer to the technical effects described in the first embodiment above, and therefore will not be repeated here.

[0157] Fourth embodiment

[0158] This embodiment provides a computer-readable storage medium storing at least one instruction, which is loaded and executed by a processor to implement the method of the first embodiment described above. The computer-readable storage medium may be a ROM, random access memory, CD-ROM, magnetic tape, floppy disk, or optical data storage device. The instructions stored therein can be loaded by a processor in a terminal to execute the method described above.

[0159] Furthermore, it should be noted that the present invention may be provided as a method, apparatus, or computer program product. Therefore, embodiments of the present invention may take the form of a fully or partially hardware embodiment, a fully or partially software embodiment, or an embodiment combining software and hardware aspects. Furthermore, when implemented using software, embodiments of the present invention may take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The computer program product comprises one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the processes or functions described in accordance with the embodiments of the present invention are fully or partially generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired connection (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium may 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 may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a DVD), or a semiconductor medium. The semiconductor medium may be a solid state drive.

[0160] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the process in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0161] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device, so that a series of operation steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for implementing the process in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0162] It should also be noted that, in this document, relational terms such as first and second are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any actual relationship or order between these entities or operations. The terms "include," "comprises," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or terminal device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or terminal device. Without further limitation, an element defined by the phrase "comprising a..." does not preclude the presence of other identical elements in the process, method, article, or terminal device comprising the element. In addition, the term "and / or" is merely a description of an associative relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent: the presence of A alone, the presence of A and B simultaneously, or the presence of B alone, where A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding. "At least one" means one or more, and "more" means two or more. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b or c can be represented by: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0163] In addition, it can 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.

[0164] 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.

[0165] In the several embodiments provided herein, 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 functional modules / units is merely a logical functional division. In actual implementation, other division methods may be used, such as multiple units or components being combined or integrated into another device, or some features being ignored or not implemented. Furthermore, the coupling or direct coupling or communication connection shown or discussed between each other may be through some interface, indirect coupling or communication connection between devices or units, and may be electrical, mechanical, or other forms. Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, i.e., they may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of the present embodiment according to actual needs. In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit.

[0166] If the method is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The 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 perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0167] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention. It should be noted that, although preferred embodiments of the present invention have been described, those skilled in the art, once understanding the basic inventive concepts of the present invention, may make various improvements and modifications without departing from the principles of the present invention. Such improvements and modifications should also be considered within the scope of protection of the present invention. Therefore, the appended claims are intended to be interpreted as covering the preferred embodiments and all variations and modifications that fall within the scope of the embodiments of the present invention.

Claims

1. A method for intelligent identification of microseismic signals for mining, characterized in that: include: Acquire the microseismic monitoring signal to be identified as the signal to be analyzed; Decompose the signal to be analyzed into several single-type signals; Perform feature extraction on the decomposed single type of signal to obtain feature data of the signal to be analyzed; The characteristic data of the signal to be analyzed is input into the preset recognition model to realize intelligent recognition of microseismic signals.

2. The method for intelligent identification of microseismic signals for mining according to claim 1, characterized in that: The decomposition of the signal to be analyzed into several single-type signals includes: Step 1: Perform discrete Fourier transform on the signal to obtain the frequency domain representation S[k] as the residual spectrum; Step 2: Find the frequency f corresponding to the maximum amplitude in the residual spectrum. max ; Step 3, set the harmonic order m = 1, 2, ..., M max , for each possible m, calculate its candidate fundamental frequency f m =f max / m, and solve its harmonic energy function value E(f m ); where the expression of the harmonic energy function is: Where M is the maximum number of harmonics; round() means rounding to the nearest integer; R[] represents the residual spectrum; f s is the sampling frequency; N is the total number of sampling points of the signal; Step 4: Determine the candidate fundamental frequency f corresponding to the maximum harmonic energy function value c-max , construct the optimization function, and obtain the final fundamental frequency f0 by solving the optimization function; the optimization function expression is: Among them, E(f c-max ) represents f c-max The corresponding harmonic energy function value; Step 5: Determine the harmonics m=1, 2, ..., M corresponding to f0 and calculate the frequency k corresponding to each harmonic m : Step 6: Construct the spectrum S of a single signal i [k], while removing S from the residual spectrum i [k], update the residual spectrum; where the spectrum S of a single signal is constructed i The process of [k] is as follows: First, create a spectrum S with the same length as S[k] i [k], and initialize all its values to 0; for each harmonic frequency index, extract the corresponding value from S[k] and assign it to S i [k]; then, let S i [Nk m ]=S[Nk m ], except S i [k m ] and S i [Nk m ], all other positions of S i [k] is 0; Step 7, based on the spectrum of the constructed signal, reconstruct a single signal by inverse discrete Fourier transform; Step 8, repeating steps 1 to 7 several times to obtain multiple reconstructed signals of a single type; Step 9: Return to step 2 and stop iterating until the maximum amplitude of the residual spectrum is lower than a preset threshold or the expected number of signals are extracted.

3. The method for intelligent identification of microseismic signals for mining according to claim 1, characterized in that: The feature extraction of the decomposed single type of signal to obtain feature data of the signal to be analyzed includes: Divide the signal into multiple time windows of fixed length and calculate the statistics of the signal in each time window; the statistics include the mean, variance, skewness and kurtosis of the signal in the corresponding time window; Calculate the difference in statistics between adjacent time windows; The difference between the statistics of every two adjacent time windows is combined into a feature vector, and all the obtained feature vectors are used as the feature data of the signal to be analyzed.

4. The method for intelligent identification of microseismic signals for mining according to claim 1, characterized in that: The preset recognition model is a BP neural network model optimized by genetic algorithm.

5. The method for intelligent identification of microseismic signals for mining according to claim 4, characterized in that: The training process of the preset recognition model includes: Obtain historical microseismic monitoring signals as sample signals and label them according to their categories; Decomposing the sample signal into several single-type signals, and extracting features from the decomposed single-type signals to obtain feature data of the sample signal; Using the feature data and label data of the sample signal, a feature database is constructed; The feature database is used to train the preset recognition model, and the genetic algorithm is used to optimize the weight and threshold parameters of the BP neural network to enhance the learning ability and generalization ability of the neural network.

6. A mine-used microseismic signal intelligent identification system, characterized in that: include: A data acquisition module is used to acquire the microseismic monitoring signal to be identified as a signal to be analyzed; The data processing module is used to decompose the signal to be analyzed into several single-type signals; and extract features from the decomposed single-type signals to obtain feature data of the signal to be analyzed; The intelligent recognition module is used to input the characteristic data of the signal to be analyzed obtained by the data processing module into a preset recognition model to realize intelligent recognition of microseismic signals.

7. The intelligent identification system for microseismic signals used in mines according to claim 6, characterized in that: The decomposition of the signal to be analyzed into several single-type signals includes: Step 1: Perform discrete Fourier transform on the signal to obtain the frequency domain representation S[k] as the residual spectrum; Step 2: Find the frequency f corresponding to the maximum amplitude in the residual spectrum. max ; Step 3, set the harmonic order m = 1, 2, ..., M max , for each possible m, calculate its candidate fundamental frequency f m =f max / m, and solve its harmonic energy function value E(f m ); where the expression of the harmonic energy function is: Where M is the maximum number of harmonics; round() means rounding to the nearest integer; R[] represents the residual spectrum; f s is the sampling frequency; N is the total number of sampling points of the signal; Step 4: Determine the candidate fundamental frequency f corresponding to the maximum harmonic energy function value c-max , construct the optimization function, and obtain the final fundamental frequency f0 by solving the optimization function; the optimization function expression is: Among them, E(f c-max ) represents f c-max The corresponding harmonic energy function value; Step 5: Determine the harmonics m=1, 2, ..., M corresponding to f0 and calculate the frequency k corresponding to each harmonic m : Step 6: Construct the spectrum S of a single signal i [k], while removing S from the residual spectrum i [k], update the residual spectrum; where the spectrum S of a single signal is constructed i The process of [k] is as follows: First, create a spectrum S with the same length as S[k] i [k], and initialize all its values to 0; for each harmonic frequency index, extract the corresponding value from S[k] and assign it to S i [k]; then, let S i [Nk m ]=S[Nk m ], except S i [k m ] and S i [Nk m ], all other positions of S i [k] is 0; Step 7, based on the spectrum of the constructed signal, reconstruct a single signal by inverse discrete Fourier transform; Step 8, repeating steps 1 to 7 several times to obtain multiple reconstructed signals of a single type; Step 9: Return to step 2 and stop iterating until the maximum amplitude of the residual spectrum is lower than a preset threshold or the expected number of signals are extracted.

8. The intelligent identification system for microseismic signals used in mines according to claim 6, characterized in that: The feature extraction of the decomposed single type of signal to obtain feature data of the signal to be analyzed includes: Divide the signal into multiple time windows of fixed length and calculate the statistics of the signal in each time window; the statistics include the mean, variance, skewness and kurtosis of the signal in the corresponding time window; Calculate the difference in statistics between adjacent time windows; The difference between the statistics of every two adjacent time windows is combined into a feature vector, and all the obtained feature vectors are used as the feature data of the signal to be analyzed.

9. The intelligent identification system for microseismic signals used in mines according to claim 6, characterized in that: The preset recognition model is a BP neural network model optimized by genetic algorithm.

10. The intelligent identification system for microseismic signals used in mines according to claim 9, characterized in that: The training process of the preset recognition model includes: Obtain historical microseismic monitoring signals as sample signals and label them according to their categories; Decomposing the sample signal into several single-type signals, and extracting features from the decomposed single-type signals to obtain feature data of the sample signal; Using the feature data and label data of the sample signal, a feature database is constructed; The feature database is used to train the preset recognition model, and the genetic algorithm is used to optimize the weight and threshold parameters of the BP neural network to enhance the learning ability and generalization ability of the neural network.

Images