Microseismic event identification method, device, equipment and storage medium

Abstract

The present application discloses a microseismic event identification method, apparatus, device, and storage medium. The disclosed microseismic event identification method is applied to a graph neural network, which is composed of multiple interconnected nodes. The present application obtains a response result of each sensor node by having each sensor node respond to a stress signal of a signal node, and the response result includes at least waveform data and audio data; inputs each waveform data into a preset waveform prediction model to obtain an original waveform probability vector corresponding to each sensor node; inputs each audio data into a preset audio prediction model to obtain an original audio probability vector corresponding to each sensor node; and determines the probability that the stress signal is triggered by a preset microseismic event based on each original waveform probability vector and each original audio probability vector, thereby realizing automatic identification of microseismic events without relying on manual experience, effectively improving the recognition speed of microseisms, and thereby improving the timeliness of microseismic warnings.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method, apparatus, device and storage medium for identifying microseismic events. Background Art

[0002] Mining microseismic events are ground pressure hazards caused by changes in rock stress during mining. High-energy mining microseismic events can cause economic losses to mining companies and even endanger the lives of miners, seriously threatening safe production in mines. Therefore, monitoring and identifying microseismic events is crucial for protecting property and life safety.

[0003] At present, the main way to prevent high-energy microseismic events is to bury multiple sensors inside the mine. The sensors capture the stress signals generated by stress changes and output corresponding waveform data and audio data. Technicians can identify microseismic events related to changes in rock mass state by analyzing waveform data and audio data. Based on the identified microseismic events, they can capture early warning signals of high-energy microseismic events, thereby taking scientific and reasonable response measures to ensure the safe and smooth progress of production activities.

[0004] However, since the stress signals detected by the sensors are not only triggered by microseismic events, but may also be triggered by other types of events, such as blasting, noise, and mechanical vibration, the above manual method relies on manual experience, has a large workload, and cannot achieve rapid identification of microseisms, resulting in low timeliness of microseismic warning.

[0005] The above content is only used to assist in understanding the technical solution of this application and does not constitute an admission that the above content is prior art. Summary of the Invention

[0006] The main purpose of this application is to provide a microseismic event identification method, device, equipment and storage medium, aiming to solve the technical problem that the existing technology relies on manual experience, has a large workload, cannot achieve rapid identification of microseisms, and results in low timeliness of microseismic warning.

[0007] To achieve the above objectives, the present application provides a microseismic event identification method, which is applied to a graph neural network. The graph neural network is composed of multiple interconnected nodes, wherein the nodes include sensor nodes and signal nodes. The sensor nodes are composed of sensors, and the signal nodes are composed of locations that generate stress signals. The method includes:

[0008] Obtaining response results of each sensor node by each sensor node responding to the stress signal of the signal node, wherein the response results at least include waveform data and audio data;

[0009] Inputting each of the waveform data into a preset waveform prediction model to obtain an original waveform probability vector corresponding to each of the sensor nodes, wherein the original waveform probability vector represents the probability that the stress signal is triggered by each preset event;

[0010] Inputting each of the audio data into a preset audio prediction model to obtain an original audio probability vector corresponding to each of the sensor nodes, wherein the original audio probability vector represents the probability that the stress signal is triggered by each of the preset events;

[0011] The probability that the signal node is triggered by a preset microseismic event is determined according to each of the original waveform probability vectors and each of the original audio probability vectors.

[0012] In one embodiment, the step of determining the probability that the stress signal is triggered by a preset microseismic event based on each of the original waveform probability vectors and each of the original audio probability vectors includes:

[0013] Acquire environmental parameters of the signal node at a current moment, wherein the environmental parameters include at least a temperature parameter, a humidity parameter, an air pressure parameter, and a regional maximum principal stress value;

[0014] Combining the original waveform probability vector and the original audio probability vector of the same sensor node according to the temperature parameter and the humidity parameter to obtain a combined probability vector corresponding to each sensor node;

[0015] Aggregating the combined probability vectors according to the air pressure parameter and the maximum principal stress value of the region to obtain a target probability vector;

[0016] The probability that the stress signal is triggered by a preset microseismic event is determined based on the target probability vector.

[0017] In one embodiment, the step of combining the original waveform probability vector and the original audio probability vector of the same sensor node according to the temperature parameter and the humidity parameter to obtain a combined probability vector corresponding to each sensor node includes:

[0018] Determining the temperature attenuation coefficient based on the temperature parameter using a preset temperature attenuation coefficient formula;

[0019] Among them, the preset temperature attenuation coefficient formula is:

[0020] ;

[0021] Where, is a signal node, t For the current moment, is the temperature attenuation coefficient of the signal node at the current moment, is the temperature parameter of the signal node at the current moment, is the first learnable attenuation coefficient, is the optimal operating temperature of the sensor, is the temperature influence range parameter;

[0022] Determining a humidity attenuation coefficient based on the humidity parameter using a preset humidity attenuation coefficient formula;

[0023] Among them, the preset humidity attenuation coefficient formula is:

[0024] ;

[0025] Where, is a signal node, t For the current moment, is the humidity attenuation coefficient of the signal node at the current moment, is the humidity parameter of the signal node at the current moment, is the second learnable attenuation coefficient, is a learnable parameter, is the correction factor, The threshold for humidity to take effect;

[0026] Determine a combined probability vector corresponding to each of the sensors based on the temperature attenuation coefficient and the humidity attenuation coefficient using a preset probability formula;

[0027] Among them, the preset probability formula is:

[0028] ;

[0029] Where, is a signal node, j For sensor nodes, t For the current moment, is the combined probability vector of the sensor node at the current moment, is the temperature attenuation coefficient of the signal node at the current moment, is the humidity attenuation coefficient of the signal node at the current moment, is the original waveform probability vector of the sensor node at the current moment, is the original audio probability vector of the sensor node at the current moment.

[0030] In one embodiment, the step of aggregating the combined probability vectors according to the air pressure parameter and the regional maximum principal stress value to obtain a target probability vector includes:

[0031] Determining the air pressure attenuation coefficient based on the air pressure parameter using a preset air pressure attenuation coefficient formula;

[0032] The preset pressure attenuation coefficient formula is:

[0033] ;

[0034] Where, is a signal node, t For the current moment, is the air pressure attenuation coefficient of the signal node at the current moment, is the air pressure parameter of the signal node at the current moment, is standard atmospheric pressure, is the sensitivity of the sensor node under standard atmospheric pressure, is the pressure influence coefficient;

[0035] Determining the maximum principal stress value of the sensor corresponding to each of the sensor nodes;

[0036] Determining the stress propagation attenuation coefficient corresponding to each sensor node based on the maximum principal stress value of the region and the maximum principal stress value of each sensor by using a preset stress propagation attenuation coefficient formula;

[0037] The preset stress propagation attenuation coefficient formula is:

[0038] ;

[0039] Where, is a signal node, j For sensor nodes, t For the current moment, is the stress propagation attenuation coefficient between the sensor node and the signal node at the current moment, is the maximum principal stress value in the region, is the maximum principal stress value of the sensor, is the angle between the direction of the regional maximum principal stress of the signal node and the line connecting the signal node and the sensor node, is the material attenuation coefficient of stress propagation, is the distance between the signal node and the sensor node;

[0040] Determining an attention weight between each of the sensor nodes and the signal node according to the air pressure attenuation coefficient and each of the stress propagation attenuation coefficients;

[0041] The combined probability vectors are aggregated according to the attention weights to obtain a target probability vector.

[0042] In one embodiment, the step of determining the attention weight between each of the sensor nodes and the signal node according to the air pressure attenuation coefficient and each of the stress propagation attenuation coefficients includes:

[0043] Determining an attention coefficient between each of the sensor nodes and the signal node based on the air pressure attenuation coefficient and each of the stress propagation attenuation coefficients using a preset attention coefficient formula;

[0044] The preset attention coefficient formula is:

[0045] ;

[0046] Where, is a signal node, j For sensor nodes, t For the current moment, is the attention coefficient between the sensor node and the signal node at the current moment, is the air pressure attenuation coefficient of the signal node at the current moment, is the stress propagation attenuation coefficient between the sensor node and the signal node at the current moment, is a nonlinear activation function, is the learnable parameter vector, T is the matrix transpose operation, is the learnable weight matrix of the shared linear transformation, is the environmental feature vector of the signal node’s environment, is the environmental feature vector of the environment where the sensor node is located;

[0047] The attention coefficients are normalized by a preset attention weight formula to obtain the attention weight corresponding to each sensor node;

[0048] Among them, the preset attention weight formula is:

[0049] ;

[0050] Where, is a signal node, j For sensor nodes, For any sensor node that receives a stress signal, t For the current moment, is the attention weight between the sensor node and the signal node at the current moment, is the attention coefficient between the sensor node and the signal node at the current moment, is the environmental feature vector of the signal node’s environment, is the environmental feature vector of the sensor node’s environment, is the learnable weight matrix of the shared linear transformation, is the set of all sensor nodes that receive stress signals, is the air pressure attenuation coefficient of the signal node at the current moment, is the stress propagation attenuation coefficient between the sensor node and the signal node at the current moment, is the attention coefficient between any sensor node that receives the stress signal and the signal node at the current moment, is the propagation attenuation coefficient between any sensor node that receives the stress signal and the signal node at the current moment.

[0051] In one embodiment, the step of aggregating the combined probability vectors according to the attention weights to obtain a target probability vector includes:

[0052] Aggregating the combined probability vectors based on the attention weights using a first preset output formula to obtain a target probability vector;

[0053] Among them, the first preset output formula is:

[0054] ;

[0055] Where, is a signal node, j For sensor nodes, t For the current moment, is the target probability vector of the signal node at the current moment, is the activation function, is the attention weight between the sensor node and the signal node at the current moment, is the learnable weight matrix of the shared linear transformation, is the combined probability vector of the sensor node at the current moment.

[0056] In one embodiment, the step of aggregating the combined probability vectors according to the attention weights to obtain a target probability vector further includes:

[0057] Aggregating the combined probability vectors based on the attention weights using a second preset output formula to obtain a target probability vector;

[0058] Wherein, the second preset output formula is:

[0059] ;

[0060] Where, is a signal node, j For sensor nodes, t For the current moment, is the target probability vector of the signal node at the current moment, is the activation function, is the attention weight between the sensor node and the signal node at the current moment, is the learnable weight matrix of the shared linear transformation, is the combined probability vector of the sensor node at the current moment, Z is the number of attention heads.

[0061] In addition, to achieve the above objectives, the present application also proposes a microseismic event identification device, the device comprising:

[0062] A data acquisition module, configured to obtain a response result of each sensor node by responding to the stress signal of the signal node, wherein the response result includes at least waveform data and audio data;

[0063] A probability prediction module is used to input each of the waveform data into a preset waveform prediction model to obtain an original waveform probability vector corresponding to each of the sensor nodes, wherein the original waveform probability vector represents the probability that the stress signal is triggered by each preset event;

[0064] The probability prediction module is further configured to input each of the audio data into a preset audio prediction model to obtain an original audio probability vector corresponding to each of the sensor nodes, wherein the original audio probability vector represents the probability that the stress signal is triggered by each of the preset events;

[0065] An event recognition module is used to determine the probability that the stress signal is triggered by a preset microseismic event based on each of the original waveform probability vectors and each of the original audio probability vectors.

[0066] In addition, to achieve the above-mentioned purpose, the present application also proposes a microseismic event identification device, which includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the computer program is configured to implement the steps of the microseismic event identification method as described above.

[0067] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium and stores a computer program. When the computer program is executed by a processor, the steps of the microseismic event identification method described above are implemented.

[0068] One or more technical solutions proposed in this application have at least the following technical effects:

[0069] The present application is applied to a graph neural network, which is composed of multiple interconnected nodes. The nodes include sensor nodes and signal nodes. The sensor nodes are composed of sensors, and the signal nodes are composed of locations that generate stress signals. The method includes: each sensor node responds to the stress signal of the signal node to obtain a response result of each sensor node, and the response result includes at least waveform data and audio data; each waveform data is input into a preset waveform prediction model to obtain an original waveform probability vector corresponding to each sensor node, and the original waveform probability vector represents the probability that the stress signal is triggered by each preset event; each audio data is input into a preset audio prediction model to obtain an original audio probability vector corresponding to each sensor node, and the original audio probability vector represents the probability that the stress signal is triggered by each preset event; and the probability of the stress signal being triggered by a preset microseismic event is determined based on each original waveform probability vector and each original audio probability vector. This application constructs a graph neural network including sensor nodes and signal nodes, adopts the environmental parameters of the signal nodes, and obtains the corresponding original waveform probability vector and original audio probability vector that characterize the probability of the stress signal being triggered by each preset event through a preset waveform prediction model and a preset audio prediction model respectively. Then, the original waveform probability vector and the original audio probability vector are combined to determine the probability of the stress signal being triggered by a preset microseismic event, thereby realizing the prediction of the probability of the stress signal being triggered by each preset event, and then realizing the automatic identification of microseismic events without relying on manual experience, effectively improving the recognition speed of microseisms, and thus improving the timeliness of microseismic warnings. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0071] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0072] Figure 1 This is a flow chart of the first embodiment of the microseismic event identification method of the present application;

[0073] Figure 2 This is a flow chart of the second embodiment of the microseismic event identification method of the present application;

[0074] Figure 3 This is a flow chart of the third embodiment of the microseismic event identification method of the present application;

[0075] Figure 4 This is a module structure diagram of the microseismic event identification device of this application;

[0076] Figure 5 This is a schematic diagram of the hardware structure of the microseismic event identification device of this application.

[0077] The realization of the objectives, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0078] It should be understood that the specific embodiments described herein are merely used to explain the technical solutions of the present application and are not intended to limit the present application.

[0079] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0080] The main solution of the embodiment of the present application is applied to a graph neural network, which is composed of multiple interconnected nodes. The nodes include sensor nodes and signal nodes. The sensor nodes are composed of sensors, and the signal nodes are composed of locations that generate stress signals. The solution includes: each sensor node responds to the stress signal of the signal node to obtain a response result of each sensor node, and the response result includes at least waveform data and audio data; each waveform data is input into a preset waveform prediction model to obtain an original waveform probability vector corresponding to each sensor node, and the original waveform probability vector represents the probability that the stress signal is triggered by each preset event; each audio data is input into a preset audio prediction model to obtain an original audio probability vector corresponding to each sensor node, and the original audio probability vector represents the probability that the stress signal is triggered by each preset event; the environmental parameters of the signal node at the current moment are obtained; and the probability that the stress signal is triggered by a preset microseismic event is determined based on the environmental parameters, each original waveform probability vector, and each original audio probability vector.

[0081] In the process of manual microseismic event identification, the stress signals monitored by the sensors are not only triggered by microseismic events, but may also be triggered by other types of events, such as blasting, noise, and mechanical vibration. Therefore, the above manual method relies on manual experience, has a large workload, and cannot achieve rapid identification of microseisms, resulting in low timeliness of microseismic warning.

[0082] The present application provides a solution by constructing a graph neural network including sensor nodes and signal nodes, using the environmental parameters of the signal nodes, and combining the environmental parameters with the original waveform probability vector and the original audio probability vector that can characterize the probability of triggering a preset event to determine the probability that the stress signal is triggered by a preset microseismic event. In this way, based on the probability of identifying the stress signal being triggered by each preset event, microseismic events can be automatically identified without relying on manual experience, effectively improving the speed of microseismic identification and thereby improving the timeliness of microseismic warning.

[0083] Based on this, this application provides a microseismic event identification method, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the microseismic event identification method of the present application.

[0084] This embodiment is applied to a graph neural network, which is composed of a plurality of interconnected nodes. The nodes include sensor nodes and signal nodes. The sensor nodes are composed of sensors, and the signal nodes are composed of locations that generate stress signals. The microseismic event recognition method includes steps S10 to S40:

[0085] Step S10 : obtaining a response result of each sensor node by having each sensor node respond to the stress signal of the signal node, wherein the response result at least includes waveform data and audio data.

[0086] It should be noted that the method of this embodiment can be executed by a computing service device capable of microseismic event identification, network communication, and program execution, such as a mobile phone, tablet computer, or personal computer. It can also be other electronic devices that perform the same or similar functions, such as devices that comprise an intelligent mine safety monitoring system. This embodiment and the following embodiments will be described below using a microseismic event identification device (referred to as the identification device).

[0087] It is understood that the aforementioned sensor can be a detection device that captures stress signals generated within a mine. This stress signal can be a mechanical wave signal generated by changes in rock stress during mining. Stress changes can be triggered by corresponding events within the mine, such as microseismic events, blasting events, noise events, mechanical vibration events, etc.

[0088] It should be noted that the three-dimensional spatial image of the mine address can be divided into multiple grid regions of different shapes based on the address characteristics of the ore body and the geological structure of the mining area. The overall grid region formed by these grid regions covers the entire mining area. Each sensor can be located in the grid region as a sensor node, and the location generating the stress signal can be located in the grid region as a signal node. Each sensor node and signal node is connected to form edges in the grid region, forming a graph neural network.

[0089] For example, suppose the coordinates of the signal node are , the coordinates of the sensor nodes are , then the distance between the signal node and any sensor node can be expressed by the preset distance formula. The preset distance formula is:

[0090] ;

[0091] Where, is a signal node, jFor sensor nodes, is the distance between the signal node and the sensor node.

[0092] It is understood that waveform data can be a record of the continuous changes in the stress signal in the time domain. After receiving the stress signal, the sensor can convert the stress signal into an electrical signal. After analog-to-digital conversion, the resulting digitized waveform data can be used to represent the changes in stress vibration amplitude, frequency, and phase over time.

[0093] It should be noted that audio data can be the propagation of sound in the air when the event corresponding to the stress signal occurs, including the sound intensity, frequency components, duration, etc. After receiving the stress signal, the sensor can obtain audio data by collecting the sound or sound wave signal when the event corresponding to the stress signal occurs.

[0094] In a specific implementation, when a stress signal appears in a mine's internal grid area, i.e., within the graph neural network, each sensor receives and identifies the stress signal, outputting its corresponding waveform data and audio data. The waveform data and audio data output by each sensor constitute the response result. The aforementioned recognition device communicates with each sensor and obtains the response result of each sensor based on the stress signal received from each sensor's response signal node.

[0095] In step S20 , each of the waveform data is input into a preset waveform prediction model to obtain an original waveform probability vector corresponding to each of the sensor nodes, wherein the original waveform probability vector represents the probability that the stress signal is triggered by each preset event.

[0096] Step S30: Input each of the audio data into a preset audio prediction model to obtain an original audio probability vector corresponding to each of the sensor nodes, wherein the original audio probability vector represents the probability that the stress signal is triggered by each of the preset events.

[0097] It should be noted that the preset waveform prediction model can be a machine learning model or a deep learning model that predicts the probability of the stress signal being triggered by different preset events based on waveform data.

[0098] It is understandable that the preset audio prediction model may be a machine learning model or a deep learning model that predicts the probability of the stress signal being triggered by different preset events based on audio data.

[0099] In a specific implementation, a large amount of waveform data and audio data of known event types can be collected and preprocessed separately, such as normalization, denoising, and resampling. For waveform data, methods such as short-time Fourier transform and wavelet transform can be used to extract time-frequency domain features, such as amplitude, frequency, energy, and duration. Waveform training data is then constructed based on the extracted time-frequency domain features and corresponding labels (constructed by event type). For audio data, methods such as mel-spectrograms and spectrograms can be used to convert audio signals into image features, such as mel-frequency cepstral coefficients, spectrograms, and zero-crossing rates. Audio training data is then constructed based on the extracted image features and corresponding labels (constructed by event type). An initial model is trained using the waveform training data to obtain a preset waveform prediction model, and the initial model is trained using the audio training data to obtain a preset audio prediction model. The initial model can be a convolutional neural network, a recurrent neural network, or the like, which is not limited in this embodiment.

[0100] Furthermore, after the above-mentioned recognition device receives the waveform data and audio data fed back by each sensor at the current moment, it can call the preset waveform prediction model and the preset audio prediction model, input each waveform data into the preset waveform prediction model respectively, and obtain the original waveform probability vector corresponding to each sensor, wherein the original waveform probability vector represents the probability that the stress signal is triggered by each preset event; input each audio data into the preset audio prediction model respectively, and obtain the original audio probability vector corresponding to each sensor, wherein the original audio probability vector represents the probability that the stress signal is triggered by each preset event.

[0101] It should be noted that the above-mentioned preset events correspond to the labels used in the training process of the preset waveform prediction model and the preset audio prediction model, and the specific number of event types can be determined according to actual needs. For example, when the labels used in the training process of the above-mentioned preset waveform prediction model and the preset audio prediction model are microseismic events, blasting events, noise events, and mechanical vibration events, the corresponding preset events can be microseismic events, blasting events, noise events, and mechanical vibration events. Assuming that the type of preset events is N, for any sensor node, the obtained original waveform probability vector and original audio probability vector are expressed as follows:

[0102] ;

[0103] Where, is a signal node, j For sensor nodes, t For the current moment, is the original waveform probability vector, is the original audio probability vector, is the original waveform probability of the stress signal being triggered by the first type of event, is the original waveform probability of the stress signal being triggered by the second type of event ， is the original waveform probability of the stress signal being triggered by the Nth type of event, is the original audio probability that the stress signal is triggered by the first type of event, is the original audio probability that the stress signal is triggered by the second type of event, is the original audio probability that the stress signal is triggered by the Nth type of event.

[0104] Step S40 : determining the probability that the stress signal is triggered by a preset microseismic event based on each of the original waveform probability vectors and each of the original audio probability vectors.

[0105] It should be noted that the above-mentioned preset microseismic event may be a pre-set event used to characterize a microseismic event, and the preset microseismic event may be included in the preset events.

[0106] In a specific implementation, when obtaining the raw waveform probability vector and raw audio probability vector for each sensor, the recognition device performs an arithmetic average of the elements of the same event type in the raw waveform probability vector and raw audio probability vector for the same sensor, thereby fusing the raw waveform probability vector and raw audio probability vector for that sensor to obtain a fused probability vector. The above process can be repeated to obtain a fused probability vector for each sensor, and then the fused probability vectors of all sensors are averaged to obtain a target probability vector. Each element in the target probability vector represents the probability that the stress signal is triggered by each preset event. In the target probability vector, the event that triggers the stress signal can be the element with the highest probability. If the probability of triggering the preset microseismic event is the highest, then the stress signal is determined to be triggered by the preset microseismic event.

[0107] This embodiment is applied to a graph neural network, which is composed of multiple interconnected nodes. The nodes include sensor nodes and signal nodes. The sensor nodes are composed of sensors, and the signal nodes are composed of locations that generate stress signals. The method includes: each sensor node responds to the stress signal of the signal node to obtain a response result of each sensor node, and the response result includes at least waveform data and audio data; each waveform data is input into a preset waveform prediction model to obtain an original waveform probability vector corresponding to each sensor node, and the original waveform probability vector represents the probability that the stress signal is triggered by each preset event; each audio data is input into a preset audio prediction model to obtain an original audio probability vector corresponding to each sensor, and the original audio probability vector represents the probability that the stress signal is triggered by each preset event; and the probability of the stress signal being triggered by a preset microseismic event is determined based on each original waveform probability vector and each original audio probability vector. This embodiment constructs a graph neural network including sensor nodes and signal nodes, adopts the environmental parameters of the signal nodes, and obtains the corresponding original waveform probability vector and original audio probability vector that characterize the probability of the stress signal being triggered by each preset event through a preset waveform prediction model and a preset audio prediction model respectively. Then, the original waveform probability vector and the original audio probability vector are combined to determine the probability of the stress signal being triggered by the preset microseismic event, thereby realizing the prediction of the probability of the stress signal being triggered by each preset event, and then realizing the automatic identification of microseismic events without relying on manual experience, effectively improving the recognition speed of microseisms, and thus improving the timeliness of microseismic warnings.

[0108] Based on the first embodiment of the present application, the second embodiment of the present application is proposed. In the second embodiment of the present application, the same or similar contents as those in the first embodiment can be referred to the above introduction and will not be repeated hereafter. Figure 2 , Figure 2 This is a flow chart of the second embodiment of the microseismic event identification method of the present application.

[0109] In this embodiment, step S40 includes steps S41 to S44:

[0110] Step S41 : obtaining the environmental parameters of the signal node at the current moment, wherein the environmental parameters at least include temperature parameters, humidity parameters, air pressure parameters, and regional maximum principal stress values.

[0111] It should be noted that the maximum principal stress value in the above region may be the maximum stress among the three principal axis stresses of the internal stress state of the signal node when subjected to force.

[0112] In a specific implementation, corresponding environmental sensors for collecting environmental parameters can be additionally set at the location of each node, such as capacitive humidity sensors, thermistor temperature sensors, semiconductor air pressure sensors, etc. The above-mentioned identification devices can communicate with each environmental sensor, and while obtaining the response results of each sensor, obtain the initial environmental parameters fed back by each environmental sensor, and then normalize and time-align each initial environmental parameter to obtain the environmental parameters for subsequent use.

[0113] Step S42 : combining the original waveform probability vector and the original audio probability vector of the same sensor node according to the temperature parameter and the humidity parameter to obtain a combined probability vector corresponding to each sensor node.

[0114] In a specific implementation, for any sensor, the recognition device can modify its original waveform probability vector and the original audio probability vector based on the sensor's corresponding temperature and humidity parameters. Specifically, weights corresponding to the temperature and humidity parameters can be pre-set, and the original waveform probability vector and the original audio probability vector can be weighted averaged. This allows the resulting combined probability vector to reflect the sensor node's environment, improving its accuracy.

[0115] In a feasible implementation, step S42 includes:

[0116] Step S421 : determining a temperature attenuation coefficient based on the temperature parameter using a preset temperature attenuation coefficient formula.

[0117] Among them, the preset temperature attenuation coefficient formula is:

[0118] ;

[0119] Where, is a signal node, t For the current moment, is the temperature attenuation coefficient of the signal node at the current moment, is the temperature parameter of the signal node at the current moment, is the first learnable attenuation coefficient, is the optimal operating temperature of the sensor, is the temperature influence range parameter.

[0120] In a specific implementation, in a high-temperature environment, the sensor is susceptible to noise interference, resulting in errors in the waveform data output by the sensor, which in turn affects the accuracy of the original waveform probability vector. Therefore, a preset temperature attenuation coefficient formula can be pre-constructed. After obtaining the waveform data of each sensor and the temperature parameters of the signal node at the current moment, the above-mentioned recognition device can read the pre-saved first learnable attenuation coefficient, the sensor's optimal operating temperature, and the temperature influence range parameters, and substitute the temperature parameters, the first learnable attenuation coefficient, the sensor's optimal operating temperature, and the temperature influence range parameters into the above-mentioned preset temperature attenuation coefficient formula to obtain the temperature attenuation coefficient of the signal node at the current moment. This temperature attenuation coefficient can reduce the impact of temperature on the waveform data and improve the accuracy of the original waveform probability vector.

[0121] It should be understood that the above-mentioned first learnable attenuation coefficient can be a parameter applicable to the preset temperature attenuation coefficient formula, and technicians can adjust it according to relevant test data to improve the accuracy of the preset temperature attenuation coefficient formula.

[0122] Step S422: determining the humidity attenuation coefficient based on the humidity parameter using a preset humidity attenuation coefficient formula.

[0123] Among them, the preset humidity attenuation coefficient formula is:

[0124] ;

[0125] Where, is a signal node, t For the current moment, is the humidity attenuation coefficient of the signal node at the current moment, is the humidity parameter of the signal node at the current moment, is the second learnable attenuation coefficient, is a learnable parameter, is the correction factor, The threshold at which humidity affects the effect.

[0126] In a specific implementation, in a mining environment, high humidity can lead to increased energy loss during audio signal propagation, resulting in errors in the audio data output by the sensor, and thus affecting the accuracy of the original audio probability vector. Therefore, a preset humidity attenuation coefficient formula can be pre-constructed. After obtaining the audio data of each sensor and the humidity parameters of the signal node at the current moment, the above-mentioned recognition device can read the pre-saved second learnable attenuation coefficient, learnable parameters, correction coefficient, and humidity influence threshold, and substitute the humidity parameters, first learnable attenuation coefficient, second learnable attenuation coefficient, learnable parameters, correction coefficient, and humidity influence threshold into the above-mentioned preset humidity attenuation coefficient formula to obtain the humidity attenuation coefficient of the signal node at the current moment. The humidity attenuation coefficient can reduce the impact of humidity on the audio data and improve the accuracy of the original audio probability vector.

[0127] It should be understood that the above-mentioned second learnable attenuation coefficient can be a parameter applicable to the preset humidity attenuation coefficient formula, and technicians can adjust it according to relevant test data to improve the accuracy of the preset humidity attenuation coefficient formula.

[0128] Step S423 : determining a combined probability vector corresponding to each of the sensors based on the temperature attenuation coefficient and the humidity attenuation coefficient using a preset probability formula.

[0129] Among them, the preset probability formula is:

[0130] ;

[0131] Where, is a signal node, j For sensor nodes, t For the current moment, is the combined probability vector of the sensor node at the current moment, is the temperature attenuation coefficient of the signal node at the current moment, is the humidity attenuation coefficient of the signal node at the current moment, is the original waveform probability vector of the sensor node at the current moment, is the original audio probability vector of the sensor node at the current moment.

[0132] In a specific implementation, for the same sensor node, after obtaining the temperature attenuation coefficient and the humidity attenuation coefficient, the above-mentioned identification device can substitute the temperature attenuation coefficient, the humidity attenuation coefficient, the original waveform probability vector, and the original audio probability vector into a pre-constructed preset probability formula to obtain the combined probability vector corresponding to the sensor node, thereby realizing the correction of the original waveform probability vector and the original audio probability vector through the temperature attenuation coefficient and the humidity attenuation coefficient, reducing the influence of temperature and humidity on the original waveform probability vector and the original audio probability vector, and improving the accuracy of the combined probability vector.

[0133] It should be understood that for each sensor node, the above preset probability formula can be called to obtain the corresponding combined probability vector.

[0134] Step S43 , aggregating the combined probability vectors according to the air pressure parameter and the regional maximum principal stress value to obtain a target probability vector.

[0135] In a specific implementation, after obtaining the combined probability vector for each sensor node, the identification device updates the edge weights connecting each sensor node to the signal node using the air pressure parameter and the maximum principal stress value. Each edge weight serves as a measure of the reliability of the combined probability vector corresponding to each sensor node. A weighted average is then performed based on each edge weight and the combined probability vector for each sensor node to obtain a target probability vector. Compared to averaging the fused probability vectors of all sensors to obtain the target probability vector, updating the edge weights connecting each sensor node to the signal node using the air pressure parameter and the maximum principal stress value effectively improves the accuracy of the target probability vector, thereby enhancing the accuracy of microseismic event identification.

[0136] Step S44: determining the probability that the stress signal is triggered by a preset microseismic event based on the target probability vector.

[0137] In a specific implementation, after obtaining the target probability vector, the identification device uses the element corresponding to the preset microseismic event in the target probability vector as the probability that the stress signal was triggered by the preset microseismic event. Specifically, if the element in the target probability vector with the highest probability of triggering the preset microseismic event is the one with the highest probability of triggering the preset microseismic event, the stress signal is determined to have been triggered by the preset microseismic event.

[0138] This embodiment obtains the environmental parameters of the signal node at the current moment, which include at least temperature, humidity, air pressure, and the regional maximum principal stress value. The original waveform probability vector and the original audio probability vector of the same sensor node are combined according to the temperature and humidity parameters to obtain a combined probability vector corresponding to each sensor node. The combined probability vectors are aggregated according to the air pressure parameter and the regional maximum principal stress value to obtain a target probability vector. Based on the target probability vector, the probability that the stress signal is triggered by a preset microseismic event is determined, thereby reducing the influence of temperature and humidity on the sensor response results and improving the accuracy of microseismic identification.

[0139] Based on the first and second embodiments of the present application, the third embodiment of the present application is proposed. In the third embodiment of the present application, the same or similar contents as those of the first and second embodiments can be referred to above and will not be described in detail later. Figure 3 , Figure 3 This is a flow chart of the third embodiment of the microseismic event identification method of the present application.

[0140] In this embodiment, step S43 includes steps S431 to S434:

[0141] Step S431 : determining the pressure attenuation coefficient based on the pressure parameter using a preset pressure attenuation coefficient formula.

[0142] The preset pressure attenuation coefficient formula is:

[0143] ;

[0144] Where, is a signal node, t For the current moment, is the air pressure attenuation coefficient of the signal node at the current moment, is the air pressure parameter of the signal node at the current moment, is standard atmospheric pressure, is the sensitivity of the sensor node under standard atmospheric pressure, is the pressure influence coefficient.

[0145] In a specific implementation, the sensitivity of sensors in a mining environment is affected by air pressure. Since the air pressure inside a mine is high, it may reduce the sensitivity of buried sensors. Therefore, a preset air pressure attenuation coefficient formula can be pre-established. After obtaining the air pressure parameters, the above-mentioned identification device can read the pre-saved standard atmospheric pressure and the sensitivity of the sensor node under standard atmospheric pressure. The air pressure parameters, standard atmospheric pressure, and the sensitivity of the sensor node under standard atmospheric pressure are then substituted into the preset air pressure attenuation coefficient formula to obtain the air pressure attenuation coefficient. This air pressure attenuation coefficient can be used to reduce the impact of air pressure on sensor sensitivity, thereby improving the accuracy of microseismic identification.

[0146] Step S432: Determine the maximum principal stress value of the sensor corresponding to each sensor node.

[0147] It should be noted that the maximum principal stress value of the sensor may be the maximum stress among the three principal axis stresses of the internal stress state of the sensor node when subjected to force.

[0148] Step S433 : determining the stress propagation attenuation coefficient corresponding to each sensor node based on the maximum principal stress value of the region and the maximum principal stress value of each sensor using a preset stress propagation attenuation coefficient formula.

[0149] The preset stress propagation attenuation coefficient formula is:

[0150] ;

[0151] Where, is a signal node, j For sensor nodes, t For the current moment, is the stress propagation attenuation coefficient between the sensor node and the signal node at the current moment, is the maximum principal stress value in the region, is the maximum principal stress value of the sensor, is the angle between the direction of the regional maximum principal stress of the signal node and the line connecting the signal node and the sensor node, is the material attenuation coefficient of stress propagation, is the distance between the signal node and the sensor node.

[0152] In a specific implementation, a preset propagation attenuation coefficient formula can be pre-established. For any sensor node, after obtaining the regional maximum principal stress value and the sensor maximum principal stress value, the identification device can read the pre-stored angle between the regional maximum principal stress direction of the signal node and the line connecting the signal node and the sensor node, as well as the material attenuation coefficient for stress propagation. The regional maximum principal stress value, the sensor maximum principal stress value, the angle, and the material attenuation coefficient for stress propagation are then substituted into the preset stress propagation attenuation coefficient formula to obtain the stress propagation attenuation coefficient between the sensor node and the signal node at the current moment. This process is repeated for other sensor nodes to obtain the stress propagation attenuation coefficient between each sensor and signal node.

[0153] Step S434 : determining the attention weight between each of the sensor nodes and the signal node according to the air pressure attenuation coefficient and each of the stress propagation attenuation coefficients.

[0154] In a specific implementation, the above-mentioned recognition device can determine the edge weight of the connection between each sensor node and the signal node according to the air pressure attenuation coefficient and each stress propagation attenuation coefficient, and this weight is the attention weight.

[0155] In a feasible implementation, step S434 includes: steps S4341-S4342:

[0156] S4341: Determine the attention coefficient between each of the sensor nodes and the signal node based on the air pressure attenuation coefficient and each of the stress propagation attenuation coefficients using a preset attention coefficient formula.

[0157] The preset attention coefficient formula is:

[0158] ;

[0159] Where, is a signal node, j For sensor nodes, t For the current moment, is the attention coefficient between the sensor node and the signal node at the current moment, is the air pressure attenuation coefficient of the signal node at the current moment, is the stress propagation attenuation coefficient between the sensor node and the signal node at the current moment, is a nonlinear activation function, is the learnable parameter vector, T is the matrix transpose operation, is the learnable weight matrix of the shared linear transformation, is the environmental feature vector of the signal node’s environment, is the environmental feature vector of the environment where the sensor node is located.

[0160] It should be noted that the environmental feature vector of the environment in which the signal node is located may include various environmental parameters at the signal node, and the environmental feature vector of the environment in which the sensor node is located may include various environmental parameters at the sensor node.

[0161] In a specific implementation, a preset attention coefficient formula can be pre-constructed. For any sensor node, after determining the air pressure attenuation coefficient and the stress propagation attenuation coefficient, the recognition device can obtain a nonlinear activation function, a learnable parameter vector, a learnable weight matrix of a shared linear transformation, the environmental feature vector of the signal node's environment, and the environmental feature vector of the sensor node's environment. The pressure attenuation coefficient, the stress propagation attenuation coefficient, the nonlinear activation function, the learnable parameter vector, the learnable weight matrix of a shared linear transformation, the environmental feature vector of the signal node's environment, and the environmental feature vector of the sensor node's environment are substituted into the preset attention coefficient formula to obtain the attention coefficient between the sensor node and the signal node at the current moment. Repeating the above process for other sensor nodes will yield the attention coefficient between each sensor and the signal node.

[0162] S4342, normalizing each of the attention coefficients using a preset attention weight formula to obtain an attention weight corresponding to each of the sensor nodes;

[0163] Among them, the preset attention weight formula is:

[0164] ;

[0165] Where, is a signal node, j For sensor nodes, For any sensor node that receives a stress signal, t For the current moment, is the attention weight between the sensor node and the signal node at the current moment, is the attention coefficient between the sensor node and the signal node at the current moment, is the environmental feature vector of the signal node’s environment, is the environmental feature vector of the sensor node’s environment, is the learnable weight matrix of the shared linear transformation, is the set of all sensor nodes that receive stress signals triggered by signal nodes, is the air pressure attenuation coefficient of the signal node at the current moment, is the stress propagation attenuation coefficient between the sensor node and the signal node at the current moment, is the attention coefficient between any sensor node that receives the stress signal and the signal node at the current moment, is the propagation attenuation coefficient between any sensor node that receives the stress signal and the signal node at the current moment.

[0166] In a specific implementation, a preset attention weight formula can be pre-established. The recognition device can obtain the attention coefficient of each sensor node (i.e., all sensor nodes that receive stress signals). The attention coefficient of any sensor node can be normalized by combining the attention coefficients of all sensor nodes to obtain the attention weight of that sensor node. Specifically, for any sensor node, the recognition device can substitute the environmental feature vector of the signal node's environment, the environmental feature vector of the sensor node's environment, a learnable weight matrix of a shared linear transformation, and the attention coefficients of each sensor node into the preset attention weight formula to normalize the attention coefficient of that sensor node and obtain the attention weight of that sensor node. This process can be repeated for other sensor nodes to normalize the attention coefficients of each sensor node and obtain the attention weight of each sensor node.

[0167] Step S435 , aggregating the combined probability vectors according to the attention weights to obtain a target probability vector.

[0168] In a specific implementation, the above-mentioned recognition device can aggregate the combined probability vectors through the attention weights corresponding to each sensor node, so that the obtained target probability vector combines multimodal data such as temperature parameters, humidity, air pressure, stress, waveform data and audio data, effectively improving the reliability of the single sensor discrimination results, and further improving the accuracy of microseismic signal recognition by aggregating the combined probability vectors of each sensor through attention weights.

[0169] In a feasible implementation, step S435 includes step S4351:

[0170] Step S4351: Aggregate the combined probability vectors based on the attention weights using a first preset output formula to obtain a target probability vector.

[0171] Among them, the first preset output formula is:

[0172] ;

[0173] Where, is a signal node, j For sensor nodes, t For the current moment, is the target probability vector of the signal node at the current moment, is the activation function, is the attention weight between the sensor node and the signal node at the current moment, is the learnable weight matrix of the shared linear transformation, is the combined probability vector of the sensor node at the current moment.

[0174] In a specific implementation, a first preset output formula can be constructed in advance, and the above-mentioned recognition device can substitute the activation function, attention weight, learnable weight matrix of shared linear transformation, and the combined probability vector of each sensor node into the above-mentioned first preset output formula to aggregate the combined probability vector of each sensor node to obtain the target probability vector.

[0175] In another feasible implementation, step S435 includes step S4352:

[0176] Step S4352: Aggregate the combined probability vectors based on the attention weights using a second preset output formula to obtain a target probability vector.

[0177] Wherein, the second preset output formula is:

[0178] ;

[0179] Where, is a signal node, j is the sensor node, t is the current time, is the target probability vector of the signal node at the current moment, is the activation function, is the attention weight between the sensor node and the signal node at the current moment, is the learnable weight matrix of the shared linear transformation, is the combined probability vector of the sensor node at the current moment, Z is the number of attention heads.

[0180] In a specific implementation, in order to improve the accuracy of the target probability vector, a multi-head attention mechanism can be used in advance to introduce multiple groups of learnable weight matrices of shared linear transformations to construct a second preset output formula. The above-mentioned recognition device can substitute the activation function, attention weight, learnable weight matrix of shared linear transformation, number of attention heads, and the combined probability vector of each sensor node into the above-mentioned second preset output formula to obtain the accuracy of the target probability vector, thereby improving the accuracy of the target probability vector.

[0181] This embodiment determines the air pressure attenuation coefficient based on the air pressure parameters through a preset air pressure attenuation coefficient formula; determines the maximum principal stress value of the sensor corresponding to each sensor node; determines the stress propagation attenuation coefficient corresponding to each sensor node based on the regional maximum principal stress value and the maximum principal stress value of each sensor through a preset propagation attenuation coefficient formula; determines the attention weight between each sensor node and the signal node according to the air pressure attenuation coefficient and each stress propagation attenuation coefficient; aggregates each combined probability vector according to each attention weight to obtain a target probability vector, thereby realizing the integration of multimodal data into the graph neural network and effectively improving the recognition accuracy of microseismic events.

[0182] This application also provides a microseismic event identification device, please refer to Figure 4 , Figure 4 This is a block diagram of the module structure of the microseismic event identification device of the present application, which includes:

[0183] The data acquisition module 10 is configured to obtain a response result of each sensor node by responding to the stress signal of the signal node of each sensor node, wherein the response result at least includes waveform data and audio data.

[0184] The probability prediction module 20 is used to input each of the waveform data into a preset waveform prediction model to obtain an original waveform probability vector corresponding to each of the sensor nodes, wherein the original waveform probability vector represents the probability that the stress signal is triggered by each preset event.

[0185] The probability prediction module 20 is further used to input each of the audio data into a preset audio prediction model to obtain an original audio probability vector corresponding to each of the sensor nodes, wherein the original audio probability vector represents the probability that the stress signal is triggered by each of the preset events.

[0186] The event identification module 30 is configured to determine the probability that the stress signal is triggered by a preset microseismic event based on each of the original waveform probability vectors and each of the original audio probability vectors.

[0187] The microseismic event identification device provided in this application utilizes the microseismic event identification method of the aforementioned embodiment, and can address the technical issues of the prior art, which relies on manual experience, is labor-intensive, and cannot rapidly identify microseisms, resulting in low timeliness of microseismic early warnings. Compared to the prior art, the beneficial effects of the microseismic event identification device provided in this application are the same as those of the microseismic event identification method provided in the aforementioned embodiment, and the other technical features of the microseismic event identification device are the same as those disclosed in the aforementioned microseismic event identification method, and are not further described here.

[0188] The present application provides an electronic device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the above-mentioned security verification method applied to a head-mounted display device or a wearable device.

[0189] Reference below Figure 5 , Figure 5 The following is a hardware diagram of the microseismic event identification device of the present application. The microseismic event identification device involved in the embodiments of the present application may include, but is not limited to, mobile terminals such as laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), and in-vehicle terminals (e.g., in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Figure 5 The microseismic event identification device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0190] like Figure 5As shown, the microseismic event identification device may include a processing device 1001 (e.g., a central processing unit, graphics processing unit, etc.), which can perform various appropriate actions and processes based on programs stored in a read-only memory (ROM) 1002 or programs loaded from a storage device 1003 into a random access memory (RAM) 1004. RAM 1004 also stores various programs and data required for the operation of the microseismic event identification device. Processing device 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007, such as a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008, such as a liquid crystal display (LCD), speaker, vibrator, etc.; storage device 1003, such as a magnetic tape or hard disk; and communication devices 1009. The communication device 1009 can allow the microseismic event identification device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows a microseismic event identification device with various systems, it should be understood that it is not required to implement or have all of the systems shown. More or fewer systems can be implemented or have alternatively.

[0191] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program comprising program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0192] The microseismic event identification device provided in this application utilizes the microseismic event identification method of the aforementioned embodiment, resolving the technical issues of the prior art, which relies on manual experience, is labor-intensive, and cannot rapidly identify microseisms, resulting in low timeliness of microseismic warnings. Compared to the prior art, the beneficial effects of the microseismic event identification device provided in this application are the same as those of the microseismic event identification method provided in the aforementioned embodiment, and the other technical features of the microseismic event identification device are the same as those disclosed in the aforementioned embodiment, and are not further elaborated here.

[0193] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0194] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0195] The present application provides a storage medium having computer-readable program instructions (ie, a computer program) stored thereon, wherein the computer-readable program instructions are used to execute the microseismic event identification method in the above-mentioned embodiment.

[0196] The computer-readable storage medium provided herein may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems, or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including, but not limited to, wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0197] The computer-readable storage medium may be included in the microseismic event identification device, or may exist independently without being incorporated into the microseismic event identification device.

[0198] The above-mentioned computer-readable storage medium carries one or more programs. When the above-mentioned one or more programs are executed by the microseismic event identification device, the microseismic event identification device: obtains the response result of each sensor node through each sensor node responding to the stress signal of the signal node, and the response result at least includes waveform data and audio data; inputs each waveform data into a preset waveform prediction model to obtain the original waveform probability vector corresponding to each sensor node, and the original waveform probability vector represents the probability that the stress signal is triggered by each preset event; inputs each audio data into a preset audio prediction model to obtain the original audio probability vector corresponding to each sensor node, and the original audio probability vector represents the probability that the stress signal is triggered by each preset event; determines the probability that the signal node is triggered by the preset microseismic event based on each original waveform probability vector and each original audio probability vector.

[0199] Computer program code for performing the operations of the present application may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0200] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, program segment or a part of code, and the module, program segment or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs the specified function or operation, or can be implemented by a combination of dedicated hardware and computer instructions.

[0201] The modules described in the embodiments of the present application may be implemented in software or hardware, wherein the name of a module does not necessarily limit the unit itself.

[0202] The computer-readable storage medium provided in this application stores computer-readable program instructions (i.e., a computer program) for executing the aforementioned microseismic event identification method. This computer-readable storage medium can address the technical issues of existing technologies that rely on manual experience, are labor-intensive, and fail to rapidly identify microseisms, resulting in low microseismic early warning effectiveness. Compared to existing technologies, the beneficial effects of the computer-readable storage medium provided in this application are similar to those of the microseismic event identification method provided in the aforementioned embodiments, and are not further elaborated here.

[0203] The above description is only part of the embodiments of the present application and does not limit the patent scope of the present application. All equivalent structural transformations made by using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect application in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A microseismic event identification method, characterized in that: The method is applied to a graph neural network, which is composed of a plurality of interconnected nodes, wherein the nodes include sensor nodes and signal nodes, wherein the sensor nodes are composed of sensors, and the signal nodes are composed of locations generating stress signals. The method includes: Obtaining response results of each sensor node by each sensor node responding to the stress signal of the signal node, wherein the response results at least include waveform data and audio data; Inputting each of the waveform data into a preset waveform prediction model to obtain an original waveform probability vector corresponding to each of the sensor nodes, wherein the original waveform probability vector represents the probability that the stress signal is triggered by each preset event; Inputting each of the audio data into a preset audio prediction model to obtain an original audio probability vector corresponding to each of the sensor nodes, wherein the original audio probability vector represents the probability that the stress signal is triggered by each of the preset events; Determining the probability that the signal node is triggered by a preset microseismic event according to each of the original waveform probability vectors and each of the original audio probability vectors; The step of determining the probability that the stress signal is triggered by a preset microseismic event based on each of the original waveform probability vectors and each of the original audio probability vectors includes: Acquire environmental parameters of the signal node at a current moment, wherein the environmental parameters include at least a temperature parameter, a humidity parameter, an air pressure parameter, and a regional maximum principal stress value; Combining the original waveform probability vector and the original audio probability vector of the same sensor node according to the temperature parameter and the humidity parameter to obtain a combined probability vector corresponding to each sensor node; Aggregating the combined probability vectors according to the air pressure parameter and the maximum principal stress value of the region to obtain a target probability vector; The probability that the stress signal is triggered by a preset microseismic event is determined based on the target probability vector.

2. The microseismic event identification method according to claim 1, wherein: The step of combining the original waveform probability vector and the original audio probability vector of the same sensor node according to the temperature parameter and the humidity parameter to obtain a combined probability vector corresponding to each sensor node includes: Determining the temperature attenuation coefficient based on the temperature parameter using a preset temperature attenuation coefficient formula; Among them, the preset temperature attenuation coefficient formula is: ； Where, is a signal node, t For the current moment, is the temperature attenuation coefficient of the signal node at the current moment, is the temperature parameter of the signal node at the current moment, is the first learnable attenuation coefficient, is the optimal operating temperature of the sensor, is the temperature influence range parameter; Determining a humidity attenuation coefficient based on the humidity parameter using a preset humidity attenuation coefficient formula; Among them, the preset humidity attenuation coefficient formula is: ； Where, is a signal node, t For the current moment, is the humidity attenuation coefficient of the signal node at the current moment, is the humidity parameter of the signal node at the current moment, is the second learnable attenuation coefficient, is a learnable parameter, is the correction factor, The threshold for humidity to take effect; Determine a combined probability vector corresponding to each of the sensors based on the temperature attenuation coefficient and the humidity attenuation coefficient using a preset probability formula; Among them, the preset probability formula is: ； Where, is a signal node, j For sensor nodes, t For the current moment, is the combined probability vector of the sensor node at the current moment, is the temperature attenuation coefficient of the signal node at the current moment, is the humidity attenuation coefficient of the signal node at the current moment, is the original waveform probability vector of the sensor node at the current moment, is the original audio probability vector of the sensor node at the current moment.

3. The microseismic event identification method according to claim 1, wherein: The step of aggregating the combined probability vectors according to the air pressure parameter and the regional maximum principal stress value to obtain a target probability vector includes: Determining the air pressure attenuation coefficient based on the air pressure parameter using a preset air pressure attenuation coefficient formula; The preset pressure attenuation coefficient formula is: ； Where, is a signal node, t For the current moment, is the air pressure attenuation coefficient of the signal node at the current moment, is the air pressure parameter of the signal node at the current moment, is standard atmospheric pressure, is the sensitivity of the sensor node under standard atmospheric pressure, is the pressure influence coefficient; Determining the maximum principal stress value of the sensor corresponding to each of the sensor nodes; Determining the stress propagation attenuation coefficient corresponding to each sensor node based on the maximum principal stress value of the region and the maximum principal stress value of each sensor by using a preset stress propagation attenuation coefficient formula; The preset stress propagation attenuation coefficient formula is: ； Where, is a signal node, j For sensor nodes, t For the current moment, is the stress propagation attenuation coefficient between the sensor node and the signal node at the current moment, is the maximum principal stress value in the region, is the maximum principal stress value of the sensor, is the angle between the direction of the regional maximum principal stress of the signal node and the line connecting the signal node and the sensor node, is the material attenuation coefficient of stress propagation, is the distance between the signal node and the sensor node; Determining an attention weight between each of the sensor nodes and the signal node according to the air pressure attenuation coefficient and each of the stress propagation attenuation coefficients; The combined probability vectors are aggregated according to the attention weights to obtain a target probability vector.

4. The microseismic event identification method according to claim 3, wherein: The step of determining the attention weight between each of the sensor nodes and the signal node according to the air pressure attenuation coefficient and each of the stress propagation attenuation coefficients includes: Determining an attention coefficient between each of the sensor nodes and the signal node based on the air pressure attenuation coefficient and each of the stress propagation attenuation coefficients using a preset attention coefficient formula; The preset attention coefficient formula is: ； Where, is a signal node, j For sensor nodes, t For the current moment, is the attention coefficient between the sensor node and the signal node at the current moment, is the air pressure attenuation coefficient of the signal node at the current moment, is the stress propagation attenuation coefficient between the sensor node and the signal node at the current moment, is a nonlinear activation function, is the learnable parameter vector, T is the matrix transpose operation, is the learnable weight matrix of the shared linear transformation, is the environmental feature vector of the signal node’s environment, is the environmental feature vector of the environment where the sensor node is located; The attention coefficients are normalized by a preset attention weight formula to obtain the attention weight corresponding to each sensor node; Among them, the preset attention weight formula is: ； Where, is a signal node, j For sensor nodes, k For any sensor node that receives a stress signal, t For the current moment, is the attention weight between the sensor node and the signal node at the current moment, is the attention coefficient between the sensor node and the signal node at the current moment, is the environmental feature vector of the signal node’s environment, is the environmental feature vector of the sensor node’s environment, is the learnable weight matrix of the shared linear transformation, is the set of all sensor nodes that receive stress signals, is the air pressure attenuation coefficient of the signal node at the current moment, is the stress propagation attenuation coefficient between the sensor node and the signal node at the current moment, is the attention coefficient between any sensor node that receives the stress signal and the signal node at the current moment, is the propagation attenuation coefficient between any sensor node that receives the stress signal and the signal node at the current moment.

5. The microseismic event identification method according to claim 4, characterized in that: The step of aggregating the combined probability vectors according to the attention weights to obtain a target probability vector includes: Aggregating the combined probability vectors based on the attention weights using a first preset output formula to obtain a target probability vector; Among them, the first preset output formula is: ； Where, is a signal node, j For sensor nodes, t For the current moment, is the target probability vector of the signal node at the current moment, is the activation function, is the attention weight between the sensor node and the signal node at the current moment, is the learnable weight matrix of the shared linear transformation, is the combined probability vector of the sensor node at the current moment.

6. The microseismic event identification method according to claim 4, wherein: The step of aggregating the combined probability vectors according to the attention weights to obtain a target probability vector further includes: Aggregating the combined probability vectors based on the attention weights using a second preset output formula to obtain a target probability vector; Wherein, the second preset output formula is: ； Where, is a signal node, j is the sensor node, t is the current time, is the target probability vector of the signal node at the current moment, is the activation function, is the attention weight between the sensor node and the signal node at the current moment, is the learnable weight matrix of the shared linear transformation, is the combined probability vector of the sensor node at the current moment, Z is the number of attention heads.

7. A microseismic event identification device, characterized in that: The device comprises: A data acquisition module, configured to obtain a response result of each sensor node by responding to the stress signal of the signal node, wherein the response result includes at least waveform data and audio data; A probability prediction module is used to input each of the waveform data into a preset waveform prediction model to obtain an original waveform probability vector corresponding to each of the sensor nodes, wherein the original waveform probability vector represents the probability that the stress signal is triggered by each preset event; The probability prediction module is further configured to input each of the audio data into a preset audio prediction model to obtain an original audio probability vector corresponding to each of the sensor nodes, wherein the original audio probability vector represents the probability that the stress signal is triggered by each of the preset events; an event recognition module, configured to determine, based on each of the original waveform probability vectors and each of the original audio probability vectors, a probability that the stress signal is triggered by a preset microseismic event; The event identification module is further configured to: Acquire environmental parameters of the signal node at a current moment, wherein the environmental parameters include at least a temperature parameter, a humidity parameter, an air pressure parameter, and a regional maximum principal stress value; Combining the original waveform probability vector and the original audio probability vector of the same sensor node according to the temperature parameter and the humidity parameter to obtain a combined probability vector corresponding to each sensor node; Aggregating the combined probability vectors according to the air pressure parameter and the maximum principal stress value of the region to obtain a target probability vector; The probability that the stress signal is triggered by a preset microseismic event is determined based on the target probability vector.

8. A microseismic event identification device, characterized in that: The microseismic event identification device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the microseismic event identification method according to any one of claims 1 to 6.

9. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the microseismic event identification method according to any one of claims 1 to 6 are implemented.

Images