Mine water inrush disaster early warning system based on acousto-optic fusion perception

Abstract

The invention discloses a mine water inrush disaster early warning system based on acousto-optic fusion perception, and belongs to the technical field of mine safety management. According to the mine water inrush disaster early warning system based on acousto-optic fusion perception, the acoustic data, the optical data and the environmental parameters in the target mine are acquired through the preset acquisition equipment end, and according to the acoustic data, the optical data and the environmental parameters, preliminary risk judgment is performed on the water inrush disaster in the target mine; if the result of the preliminary risk judgment is that the risk exists, the acoustic data, the optical data and the environmental parameters are uploaded to an edge computing node; the risk value of the water inrush disaster in the target mine is predicted according to the acoustic data, the optical data and the environmental parameters through a risk prediction model pre-arranged at the edge calculation node, and the risk value is fed back to a risk early warning end; and sending out corresponding early warning information according to the risk value so as to prompt related personnel of the current risk condition of the target mine.

Description

Technical Field

[0001] The present application relates to the technical field of mine safety management, and in particular to a mine water inrush disaster early warning system based on acoustic and optical fusion perception. Background Art

[0002] Mine flooding is a major safety threat faced by underground mining operations, such as coal and metal mines. It is characterized by suddenness, destructive power, and high difficulty in prediction. Once it occurs, it often causes tunnel flooding, equipment damage, and even significant casualties and economic losses.

[0003] Currently, monitoring and early warning of mine water inrush hazards primarily rely on monitoring hydrogeological parameters, surrounding rock stress and strain, geophysical exploration, and manual inspections. However, these traditional monitoring methods often focus on changes in a single or a few physical quantities. Before a water inrush hazard occurs, its precursor information is often multi-source, coupled, and complex. Relying solely on single-dimensional information makes it difficult to fully and accurately capture early signs of a hazard, which can easily lead to missed or false alarms. Summary of the Invention

[0004] The main purpose of this application is to provide a mine water inrush disaster warning system based on sound and light fusion perception, aiming to solve the technical problem that the current single-dimensional information in the mine is difficult to fully and accurately capture the early signals of disaster incubation, which easily leads to missed reports or false reports.

[0005] To achieve the above objectives, the present application provides a mine water inrush disaster early warning system based on acoustic and optical fusion perception, the mine water inrush disaster early warning system based on acoustic and optical fusion perception includes: a data monitoring terminal, an edge computing node and a risk warning terminal; The data monitoring terminal is used to collect acoustic data, optical data, and environmental parameters in the target mine through a preset collection device, and to make a preliminary risk assessment of water inrush disasters in the target mine based on the acoustic data, the optical data, and the environmental parameters. If the result of the preliminary risk assessment is that there is a risk, the acoustic data, the optical data, and the environmental parameters are uploaded to the edge computing node; The edge computing node is configured to, after receiving the acoustic data, optical data, and environmental parameters sent by the data monitoring terminal, predict a risk value of a water inrush disaster in the target mine based on the acoustic data, the optical data, and the environmental parameters using a risk prediction model pre-deployed at the edge computing node, and feed the risk value back to the risk warning terminal; The risk warning terminal is used to receive the risk value sent by the edge computing node, and issue corresponding warning information according to the size of the risk value to remind relevant personnel of the current risk situation of the target mine.

[0006] In one embodiment, when the edge computing node predicts the risk value of water inrush disaster in the target mine based on the acoustic data, the optical data, and the environmental parameters using a risk prediction model pre-deployed at the edge computing node: determining a dynamic weight corresponding to a water inrush stage of a water inrush disaster in the target mine according to the acoustic data, the optical data, and the environmental parameters; Through the risk prediction model pre-deployed at the edge computing node, the risk value of the water inrush disaster in the target mine is predicted based on the acoustic data, the optical data and the dynamic weight, wherein the dynamic weight is used to influence the tendency proportion of the acoustic data and the optical data when the risk prediction model performs risk prediction.

[0007] In one embodiment, when the edge computing node determines the dynamic weight corresponding to the water inrush stage of the water inrush disaster in the target mine based on the acoustic data, the optical data, and the environmental parameters: Obtaining first rule standards corresponding to different water inrush stages of the target mine, and obtaining second rule standards corresponding to different interference environments in the target mine; determining a first confidence parameter for the acoustic data and the optical data according to the first rule criterion; Determining a second confidence parameter corresponding to the environmental data according to the second rule standard; The dynamic weight corresponding to the water inrush stage of the water inrush disaster in the target mine is calculated according to the first confidence parameter, the second confidence parameter and the preset basic weights corresponding to the different confidence parameters.

[0008] In one embodiment, after the edge computing node predicts the risk value of water inrush disaster in the target mine based on the acoustic data, the optical data, and the environmental parameters: Obtaining a predicted historical risk value and label information corresponding to the historical risk value, wherein the label information is obtained by relevant operation and maintenance personnel based on the historical risk value and the actual environmental conditions of the target mine corresponding to the historical risk value; According to the label information, the preset basic weights corresponding to the different confidence parameters are optimized.

[0009] In one embodiment, when the edge computing node determines the dynamic weight corresponding to the water inrush stage of the water inrush disaster in the target mine based on the acoustic data, the optical data, and the environmental parameters, the method further: The acoustic data includes the count of vocalization events and the energy of the main frequency band of the sound; If the sound event count is greater than a preset number and / or the change corresponding to the energy of the main frequency band of the sound is greater than a preset energy change threshold, the first preset weight group is selected as the dynamic weight, wherein the first preset weight group includes acoustic weight and optical weight, and the acoustic weight is greater than the optical weight.

[0010] In one embodiment, when the edge computing node determines the dynamic weight corresponding to the water inrush stage of the water inrush disaster in the target mine based on the acoustic data, the optical data, and the environmental parameters, the method further: The optical data include the maximum temperature difference in the infrared region and the crack length in the visible light; If the maximum temperature difference in the infrared region is greater than the preset temperature difference and / or the change in the visible light crack length is greater than the preset length change threshold, a second preset weight group is selected as the dynamic weight, wherein the second preset weight group includes acoustic weight and optical weight, and the optical weight is greater than the acoustic weight.

[0011] In one embodiment, when the data monitoring end performs a preliminary risk assessment of the water inrush disaster in the target mine based on the acoustic data, the optical data, and the environmental parameters: performing time synchronization processing on the acoustic data and the optical data according to the influence of the environmental data on the acoustic-optical time difference; Based on the synchronously processed data and preset judgment rules, a preliminary risk judgment is made on the water inrush disaster in the target mine.

[0012] In one embodiment, after the data monitoring end performs a preliminary risk assessment of the water inrush disaster in the target mine based on the acoustic data, the optical data, and the environmental parameters, the method further includes: The acoustic data includes the sound event count and the sound main frequency band energy, and the optical data includes the maximum temperature difference in the infrared region and the visible light crack length; If any one of the sound event count and the visible light crack length is greater than the corresponding standard value, it is determined that there is an area to be inspected in the target mine, and a corresponding report is generated according to the corresponding situation of the area to be inspected; If the energy of the main frequency band of the sound and the maximum temperature difference in the infrared area are both greater than the corresponding standard values, it is determined that the result of the preliminary risk judgment of the target mine is that there is a risk.

[0013] In one embodiment, when the risk warning terminal issues corresponding warning information according to the size of the risk value: Determine a range threshold interval corresponding to the risk value within a preset risk range value; Corresponding warning information is generated according to the range threshold interval and the preset template information corresponding to the range threshold interval.

[0014] In one embodiment, the data monitoring terminals are distributed in the target mine, wherein the data monitoring terminals are distributed according to a preset distance between adjacent data monitoring terminals within the spatial range according to the spatial range of the actual channel scene in the target mine.

[0015] One or more technical solutions proposed in this application have at least the following technical effects: the mine water inrush disaster warning system based on acoustic and optical fusion perception includes: a data monitoring terminal, an edge computing node and a risk warning terminal; the data monitoring terminal is used to collect acoustic data, optical data and environmental parameters in the target mine through a preset collection device terminal, and make a preliminary risk judgment on the water inrush disaster in the target mine based on the acoustic data, the optical data and the environmental parameters. If the result of the preliminary risk judgment is that there is a risk, the acoustic data, the optical data and the environmental parameters are uploaded to the edge computing node; the edge computing node is used to, after receiving the acoustic data, optical data and environmental parameters sent by the data monitoring terminal, predict the risk value of the water inrush disaster in the target mine based on the acoustic data, the optical data and the environmental parameters through a risk prediction model pre-deployed at the edge computing node, and feed the risk value back to the risk warning terminal; the risk warning terminal is used to receive the acoustic data, optical data and environmental parameters sent by the edge computing terminal. The risk value sent by the computing node is calculated, and corresponding warning information is issued according to the size of the risk value to remind relevant personnel of the current risk situation of the target mine. That is, in this application, a three-end interactive system consisting of a data monitoring end, an edge computing node and a risk warning end is constructed. Through the data monitoring end, while collecting acoustic data, optical data and environmental data, a preliminary risk judgment is made for the three types of data. After judging that there is a risk, the above-mentioned multi-dimensional data is further sent to the edge computing node to make an accurate risk prediction through the risk prediction model arranged by the edge computing node, thereby realizing multi-level risk supervision and prediction of multi-dimensional data in the mine by constructing a preliminary risk judgment of the data monitoring end and an accurate prediction of the risk prediction model of the edge computing node. This not only ensures the reasonable practical application of multi-dimensional data, but also achieves the effect of accurate risk prediction through multi-dimensional data, thereby solving the technical problem that the single-dimensional information in the current mine is difficult to comprehensively and accurately capture the early signals of disaster incubation, which is prone to omissions or false alarms. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] 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.

[0017] 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.

[0018] Figure 1 This is a schematic diagram of the architecture of the mine water inrush disaster early warning system based on sound and light fusion perception in this application; Figure 2 This is a simplified flowchart of the risk prediction process using acoustic-optical fusion data for this application.

[0019] 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

[0020] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application. In one embodiment of the present application, the mine water inrush disaster warning system based on acoustic and optical fusion perception includes: a data monitoring terminal, an edge computing node and a risk warning terminal; The data monitoring terminal is used to collect acoustic data, optical data, and environmental parameters in the target mine through a preset collection device, and to make a preliminary risk assessment of water inrush disasters in the target mine based on the acoustic data, the optical data, and the environmental parameters. If the result of the preliminary risk assessment is that there is a risk, the acoustic data, the optical data, and the environmental parameters are uploaded to the edge computing node; The edge computing node is configured to, after receiving the acoustic data, optical data, and environmental parameters sent by the data monitoring terminal, predict a risk value of a water inrush disaster in the target mine based on the acoustic data, the optical data, and the environmental parameters using a risk prediction model pre-deployed at the edge computing node, and feed the risk value back to the risk warning terminal; The risk warning terminal is used to receive the risk value sent by the edge computing node, and issue corresponding warning information according to the size of the risk value to remind relevant personnel of the current risk situation of the target mine.

[0021] It should be noted that the mine water inrush disaster warning system based on sound and light fusion perception involved in this embodiment includes: data monitoring terminal, edge computing node and risk warning terminal. The specific interaction between the terminals in the system and their layout in the target mine can be referred to. Figure 1 .

[0022] Among them, the data monitoring end can be equipped with a corresponding acoustic data monitoring end, which is responsible for collecting data such as sound signals or mechanical vibration signals in the target mine. An optical data monitoring end can also be deployed accordingly, which is responsible for collecting data such as cracks and air turbidity in the corresponding corridors of the target mine at a fixed point and range. At the same time, an environmental monitoring end is also set up to collect data such as temperature and humidity in the target mine.

[0023] Among them, the data monitoring end is responsible for collecting acoustic data, optical data and environmental data in the target mine, and is responsible for performing simple logical judgments based on the above data to preliminarily determine whether there are corresponding risks in the target mine described by the above data. After determining that there is a risk, the data monitoring end will transmit the above three data to the edge computing node, and the edge computing node will perform accurate risk prediction based on the above three data.

[0024] It should be noted that the above-mentioned preliminary risk judgment process is mainly a simple logical judgment, setting the corresponding multi-dimensional data judgment threshold. For example, when any two of the acoustic data and optical data exceed the corresponding threshold standards at the same time, it can be judged that there is a risk, but the degree of risk is not within the judgment range of the data monitoring end. This ensures that the data monitoring end only needs a simple hardware chip to perform the corresponding data collection and logical judgment effects, avoiding the deployment of high-cost risk prediction models.

[0025] In addition, it should be noted that the scope of the target mine is relatively large. Therefore, when deploying data monitoring terminals, a large number of monitoring terminals need to be deployed according to the spatial location of the actual mine to ensure that all scopes of the target mine can be monitored in place. However, the span of its spatial location is large. Therefore, a fixed number of data monitoring terminals can be set to match an edge computing node to ensure the timeliness of the edge computing node's risk prediction.

[0026] Among them, a risk prediction model is pre-deployed at the edge computing node. The model can use acoustic data, optical data and environmental data as input data, and risk value as output data. After receiving the corresponding data from the data monitoring end, the risk prediction model will predict the risk of the environment at the current location of the data monitoring end, and generate the corresponding risk value, which will be sent to the corresponding risk warning end to ensure the overall interactive action.

[0027] Among them, the risk prediction model is mainly a neural network model constructed based on the actual sample data in the target mine. Its essence is to integrate the physical meanings represented by acoustic data, optical data and environmental data, and deeply learn the correlation between the above three data. By setting rules, the risk level of water inrush disasters represented by the three data is determined, and the acoustic data, optical data and environmental data are feature extracted, and the risk value is obtained through corresponding weight calculation. The risk value reflects the risk situation of the actual environment composed of the above data in the current target mine. Among them, the larger the risk value, the greater the corresponding risk.

[0028] Among them, the risk warning end mainly includes a communication module or an alarm module. The risk warning end can be deployed in the target mine. After receiving the corresponding risk value, it can issue corresponding warning information according to the size of the corresponding risk value to remind relevant personnel of the current risk situation of the target mine. For example, communication can be carried out between the risk warning end and the terminal equipped with the relevant personnel to inform the relevant personnel (including personnel in the target mine and personnel outside the mine) in the form of text warning information, or different colors of sound and light information can be emitted through the alarm model to remind relevant personnel that they should focus on inspection or directly evacuate.

[0029] In summary, the simplified process diagram of the above data interaction risk prediction can be referred to Figure 2 .

[0030] It should be noted that the above-mentioned risk prediction process mainly includes two parts. The data monitoring end independently initiates a simple logical judgment of the fusion of acoustic data, optical data and environmental data, and realizes simple logical operations by setting corresponding thresholds. On the one hand, it reduces the hardware cost of the data monitoring end. On the other hand, as data screening, since the target mine is in a low-risk state most of the time under normal circumstances, there is no need to incur excessive computing costs at this time, that is, there is no need to use a risk prediction model. The data with low risk conditions can be filtered out through the data monitoring end, and there is no need to send such data to the edge computing node, which directly reduces the frequency of using high-cost risk prediction models. In addition, the edge node performs accurate judgment on the three-in-one fusion of high-risk data that may exist under logical judgment, thereby ensuring timely risk prediction and the accuracy of risk prediction. That is, the entire risk prediction process is divided into simple logical judgment and precise logical analysis, which not only reduces the high cost investment of the risk prediction model, but also ensures the timeliness and accuracy of the entire risk prediction.

[0031] In this embodiment, when the edge computing node predicts the risk value of water inrush disaster in the target mine based on the acoustic data, the optical data, and the environmental parameters using the risk prediction model pre-deployed at the edge computing node: determining a dynamic weight corresponding to a water inrush stage of a water inrush disaster in the target mine according to the acoustic data, the optical data, and the environmental parameters; Through the risk prediction model pre-deployed at the edge computing node, the risk value of the water inrush disaster in the target mine is predicted based on the acoustic data, the optical data and the dynamic weight, wherein the dynamic weight is used to influence the tendency proportion of the acoustic data and the optical data when the risk prediction model performs risk prediction.

[0032] It should be noted that before using the risk prediction model to predict the risk of water inrush disasters in the target mine, it is necessary to consider that when a water inrush disaster occurs, the phenomenon characteristics represented by different stages of the water inrush disaster are different. For example, there will be large sound fluctuations in the early stage of the water inrush disaster, and there may be obvious changes in cracks and flowing water on the monitoring image in the middle stage of the water inrush disaster. Therefore, corresponding to different periods and stages of the water inrush disaster, there will be different tendencies of acoustic data and optical data. Therefore, in this embodiment, before the risk prediction model is used, a dynamic weight will be calculated based on the actually collected acoustic data, optical data and environmental data, and the dynamic weight will be used to influence the tendency proportion of acoustic data and optical data in the risk prediction model, thereby ensuring the accuracy of the risk results predicted by the risk prediction model.

[0033] The dynamic weight includes an acoustic weight and an optical weight that can be characterized according to actual data.

[0034] In this embodiment, when the edge computing node determines the dynamic weight corresponding to the water inrush stage of the water inrush disaster in the target mine based on the acoustic data, the optical data, and the environmental parameters: Obtaining first rule standards corresponding to different water inrush stages of the target mine, and obtaining second rule standards corresponding to different interference environments in the target mine; determining a first confidence parameter for the acoustic data and the optical data according to the first rule criterion; Determining a second confidence parameter corresponding to the environmental data according to the second rule standard; The dynamic weight corresponding to the water inrush stage of the water inrush disaster in the target mine is calculated according to the first confidence parameter, the second confidence parameter and the preset basic weights corresponding to the different confidence parameters.

[0035] It is understandable that when calculating the dynamic weight, it is necessary to consider the corresponding states represented by the acoustic data, optical data and environmental data. For example, the sound energy frequency band corresponding to the acoustic data is very high, or the optical data monitors the growth of cracks in the tunnel wall in the target mine, etc., and simulation verification is required for different application scenarios. In addition, there will be certain data influence tendencies for different application scenarios. For example, high-frequency abnormal sounds will be generated in the early stage of water inrush, obvious crack changes will appear in the middle stage of water inrush, etc., or during any water inrush disaster, the turbid air inside the tunnel will affect the judgment of optical data, etc.

[0036] Therefore, in this embodiment, different rule standards are designed, and different confidence parameters are assigned to different rule standards. Corresponding basic weights are pre-set for different rule standards. For specific setting examples, please refer to the following text. After designing the corresponding rule standards, a calculation formula for the corresponding dynamic weight is designed.

[0037] Specifically, the calculation formula for dynamic weight is: in, The final weight of the acoustic data (0~1); The weight of the optical data (complementary to the weight of the acoustic data, summing to 1); The confidence level of the i-th rule standard (indicating the credibility of the rule, 0-1); The preset basic weight of the i-th rule (0-1).

[0038] The physical meaning of the numerator in the above formula is the weighted acoustic weights of all rules and the credibility of each rule standard.

[0039] The sum of the credibility of all first rule criteria and second rule criteria in the above formula is used for normalization.

[0040] Among them, the acoustic weight in the dynamic weight finally calculated is the fuzzy weighted average of the weights of each rule, and the optical weight in the dynamic weight is its complement.

[0041] For example, assume that the system calculates the dynamic adjustment weight according to the following three rules: Rule 1: Sudden increase in high-frequency acoustic emission energy (μ=0.9, w=0.8); Rule 2, infrared local cooling (μ=0.8, w=0.3); Rule 3: Camera is blocked by dust (μ=0.9, w=0.1).

[0042] The calculation steps are as follows: Numerator (weighted sum): 0.9 × 0.8 + 0.8 × 0.3 + 0.9 × 0.1 = 0.72 + 0.24 + 0.09 = 1.05; Denominator (sum of membership): 0.9+0.8+0.9=2.6; Acoustic weight in dynamic weight: =1.05 / 2.6≈0.4; Optical weight in dynamic weight: =1−0.4=0.6.

[0043] It should be noted that although the acoustic energy suddenly increases after the above triggering rule criteria (Rule 1 recommends a weight of 0.8), the camera is blocked by dust (Rule 3 strongly suppresses optical data), and the final optical weight is still high (0.6) because infrared cooling (Rule 2) provides a reliable supplement.

[0044] In addition, corresponding examples are given for the first rule criterion, the second rule criterion, the first confidence parameter, and the second confidence parameter, as follows: (1) Acoustic-dominated scenario: Rule 1: A sudden increase in high-frequency acoustic emission energy, triggered by a 20dB increase in energy in the 50-80kHz band, with a confidence parameter of 0.9 and a corresponding preset basic weight of 0.8; Rule 2: The frequency of acoustic event counts suddenly increases. The trigger condition is that the acoustic emission count is greater than 5 times / second for 30 seconds. The confidence parameter is 0.7, and the corresponding preset basic weight is 0.7. Rule 3: The sound source is located close to the aquifer, and its trigger condition is that the distance to the water inflow point is less than 10m. Its confidence parameter is 0.6, and its corresponding preset basic weight is 0.6; The corresponding scenario is the early stage of water inrush, when rock fractures generate high-frequency sound waves, but no obvious seepage has been detected optically. At this time, it is necessary to rely on acoustic data first (the dynamic weight corresponding to the acoustic data is expected to be greater than 0.7).

[0045] (2) Optical-dominated scenarios: Rule 1: Local cooling in infrared images, triggered by a temperature drop of more than 2°C on the wellbore wall, with a confidence parameter of 0.8 and a corresponding preset basic weight of 0.3. Rule 2: Sudden increase in water turbidity, triggered by a laser scattering detection of suspended solids concentration greater than 50, with a confidence parameter of 0.7 and a corresponding preset basic weight of 0.2; Rule 3: Visible light crack expansion, whose trigger condition is to identify the crack growth rate greater than 0.1mm / s, its confidence parameter is 0.6, and its corresponding preset basic weight is 0.4; The corresponding scenario is the middle stage of water inrush, when water seepage causes the temperature to drop and the turbidity to increase, but the acoustic signal is stable. At this time, it is necessary to rely on optical data first (the dynamic weight corresponding to optical data is expected to increase to 0.6-0.8).

[0046] (3) Environmental interference scenarios: Rule 1: The camera is blocked by dust. The trigger condition is that the image clarity is less than the preset clarity threshold. The confidence parameter is 0.9, and the corresponding preset basic weight is 0.1. Rule 2: Mining machinery noise interference, its triggering condition is that the low-frequency energy accounts for more than 60%, its confidence parameter is 0.8, and its corresponding preset basic weight is 0.3; Rule 3: Infrared thermal imager fogging, its triggering condition is that the standard deviation of the temperature field is less than 0.5°C, its confidence parameter is 0.7, and its corresponding preset basic weight is 0.2; The corresponding scenario is that dust or mechanical noise makes the data of a certain mode unreliable. In this case, the weight of the affected mode is reduced (the dynamic weight corresponding to the optical data is expected to be reduced to 0.2).

[0047] (4) Balanced weight scenario: Rule 1: Abnormal synchronization of acoustic and optical signals. The trigger condition is that the difference between acoustic and optical delay is greater than 10ms. The confidence parameter is 0.5, and the corresponding preset basic weight is 0.5. Rule 2: Multimodal weak correlation, its triggering condition is that the correlation coefficient of the acoustic and optical features is less than 0.3, its confidence parameter is 0.4, and its corresponding preset basic weight is 0.5; Rule 3: Infrared thermal imager fogging, its triggering condition is that the standard deviation of the temperature field is less than 0.5°C, its confidence parameter is 0.7, and its corresponding preset basic weight is 0.2; The corresponding scenario is that the acoustic and optical data conflict, and the dominant mode cannot be clearly identified (the dynamic weight corresponding to the optical data is expected to be 0.5, and the dynamic weight corresponding to the acoustic data is expected to be 0.5).

[0048] In this embodiment, when the edge computing node determines the dynamic weight corresponding to the water inrush stage of the water inrush disaster in the target mine based on the acoustic data, the optical data, and the environmental parameters, the method further: If the sound event count is greater than a preset number and / or the change corresponding to the energy of the main frequency band of the sound is greater than a preset energy change threshold, the first preset weight group is selected as the dynamic weight, wherein the first preset weight group includes acoustic weight and optical weight, and the acoustic weight is greater than the optical weight.

[0049] If the maximum temperature difference in the infrared region is greater than the preset temperature difference and / or the change in the visible light crack length is greater than the preset length change threshold, a second preset weight group is selected as the dynamic weight, wherein the second preset weight group includes acoustic weight and optical weight, and the optical weight is greater than the acoustic weight.

[0050] It should be noted that, according to the example of the above-mentioned rule standard, when calculating the dynamic weight, the corresponding preset basic weight will be set in advance, and the dynamic weight actually used for the risk prediction model will be calculated according to the calculation formula of the preset basic weight and the corresponding dynamic weight. However, this process is also an approximate calculation, and its main calculation principle is still to limit the actual effect according to the rules. Therefore, the process of calculating the dynamic weight can be further simplified, and the mapping connection can be directly performed according to the parameters mentioned in the above-mentioned rules. It is only necessary to set the corresponding judgment conditions. When the event count is greater than the preset number of times and / or the change corresponding to the energy of the main frequency band of the sound is greater than the preset energy change threshold, the first preset weight group is directly selected as the dynamic weight. At this time, the acoustic weight in the first preset weight group is greater than the optical weight, which can be set to 0.7 and 0.3; when the temperature difference in the infrared region is greater than the preset temperature difference and / or the change in the visible light crack length is greater than the preset length change threshold, the second preset weight group can be directly selected as the dynamic weight, where the acoustic weight in the second preset weight group is less than the optical weight, which can be set to 0.3 and 0.7.

[0051] Among them, the first preset weight group corresponds to the early stage of the water inrush disaster, and the second preset weight group corresponds to the middle and late stages of the water inrush disaster.

[0052] In this embodiment, after the edge computing node predicts the risk value of water inrush disaster in the target mine based on the acoustic data, the optical data, and the environmental parameters: Obtain the predicted historical risk value and label information corresponding to the historical risk value, wherein the label information is obtained by relevant operation and maintenance personnel based on the historical risk value and the actual environmental conditions of the target mine corresponding to the historical risk value; based on the label information, optimize the preset basic weights corresponding to the different confidence parameters.

[0053] It should be noted that the risk value predicted by the risk prediction model does not necessarily fully meet the actual needs of the changing situation in the target mine. Therefore, after the risk prediction model has been used for a period of time, the risk values ​​predicted during that period are collected and summarized into historical risk values. The historical risk values ​​and the label information corresponding to the historical risk values ​​are processed, and the parameters related to the dynamic weights required for the risk prediction model are further optimized, that is, different confidence parameters and their corresponding preset basic weights are optimized, and the confidence parameters of different rules and the size of the preset basic weights are adjusted to ensure the accuracy of subsequent calculations of dynamic weights.

[0054] Among them, the label information corresponding to the historical risk value is obtained by relevant operation and maintenance personnel based on the historical risk value and the actual environmental conditions in the target mine corresponding to the historical risk value, thereby establishing a corresponding relationship between the risk value and the actual environmental conditions to ensure the accuracy of subsequent model use.

[0055] In this embodiment, when the data monitoring end performs a preliminary risk assessment of the water inrush disaster in the target mine based on the acoustic data, the optical data, and the environmental parameters: According to the influence of the environmental data on the time difference between sound and light, the acoustic data and the optical data are time-synchronized; based on the synchronized data and preset judgment rules, a preliminary risk judgment of the water inrush disaster in the target mine is made.

[0056] It is understandable that environmental data includes the smoke and dust conditions, temperature and humidity conditions inside the target mine tunnel. Such environmental conditions will affect the situation where the collection time of the acoustic data and optical data collected by the data monitoring end cannot correspond. Therefore, before making a preliminary risk prediction for the acoustic data and optical data, these two types of data need to be time-synchronized to ensure that different data for the same event have the same timestamp, so as to facilitate subsequent logical reasoning and determine whether there is a risk in the current target mine.

[0057] Among them, the time synchronization processing for acoustic data and optical data can mainly include the following two solutions: One of them is: the deployment of the data monitoring end and data synchronization requires certain hardware configuration. The acoustic end uses a distributed high-frequency microphone array (20-200kHz), with a group deployed every 50 meters, and calculates the sound source position through the TDOA (time difference between localization and location) algorithm; the optical end uses an infrared thermal imager (to monitor the temperature field) and a visible light camera (to identify cracks), which are installed coaxially with the acoustic sensor.

[0058] The synchronization mechanism between the two can adopt any of the following schemes: Solution 1: Hardware-level PTP (Precision Time Protocol), with a synchronization error of less than 1μs (requires wired network support).

[0059] Solution 2: Software-level dynamic time warping to align non-uniformly sampled data (such as acoustic 100kHz or optical 30fps).

[0060] Another option is to select a reasonable time difference based on environmental data. This involves setting a fixed delay parameter for sound wave propagation, taking into account the differences in the propagation speeds of sound and light. This fixed delay parameter can then be adjusted based on different environmental scenarios. For example, in scenarios with high humidity or low visibility, the time to collect light will be slightly later than in scenarios with high visibility, and the fixed delay parameter for sound wave propagation needs to be lowered.

[0061] Among them, the preset judgment rules refer to the flexible judgment conditions that are customized by relevant personnel by linking multi-dimensional data. The judgment conditions need to be combined with the data in the acoustic data and optical data that have an influencing relationship with each other for joint judgment. For example, if a high-frequency sound segment is generated in the acoustic data and the optical data collects the enlargement of the cracks in the corridor wall in the target mine, it can be used as a basis for judgment to determine whether there is a risk in the target mine.

[0062] In this embodiment, after the data monitoring end performs a preliminary risk assessment of the water inrush disaster in the target mine based on the acoustic data, the optical data, and the environmental parameters, the method further includes: The acoustic data includes the sound event count and the sound main frequency band energy, and the optical data includes the maximum temperature difference in the infrared region and the visible light crack length; If any one of the sound event count and the visible light crack length is greater than the corresponding standard value, it is determined that there is an area to be inspected in the target mine, and a corresponding report is generated according to the corresponding situation of the area to be inspected; If the energy of the main frequency band of the sound and the maximum temperature difference in the infrared area are both greater than the corresponding standard values, it is determined that the result of the preliminary risk judgment of the target mine is that there is a risk.

[0063] It can be understood that the acoustic data at least includes the occurrence of the event and the energy of the main frequency band of the sound, and it may also include parameters such as the maximum decibel of the sound. The optical data at least includes the maximum temperature difference in the infrared region and the length of the visible light crack, and it may also include visibility caused by environmental dust in the space.

[0064] In this embodiment, the acoustic data includes the sound event count and the sound main frequency band energy, and the optical data includes the maximum temperature difference in the infrared region and the visible light crack length.

[0065] Among them, the corresponding preliminary risk judgment steps can be set in combination with the above-mentioned acoustic data and optical data. Specifically, when the energy of the main frequency band of sound and the maximum temperature difference in the infrared area are both greater than the corresponding standard values, the result of the preliminary risk judgment in the target mine is determined to be that there is a risk. If any data in the event count and the visible light crack length is greater than the corresponding standard value, it is determined that there is an area to be inspected in the target mine, and a corresponding report needs to be generated based on the area to be inspected.

[0066] That is, in the preliminary risk judgment process, the judgment results will be divided into two categories, including the existence of risks and the existence of areas to be inspected. When it is determined that there is a risk, risk prediction is required through the edge computing node. When it is determined that there is an area to be inspected, a corresponding report needs to be generated to prompt the corresponding personnel to conduct inspections.

[0067] It should be noted that through the above-mentioned preliminary risk judgment process, simple logical rules can be used to determine whether there is a risk in the current target mine, thereby reducing the cost of deploying a risk prediction model in the target mine. At the same time, this judgment method can rely solely on low-cost hardware equipment to perform the corresponding preliminary risk judgment, greatly reducing the cost requirements of risk prediction. At the same time, risk prediction is performed in a combination of this judgment method and the risk prediction model, which can reduce the amount of data required to be processed by the risk prediction model, and screen out most of the data through preliminary risk judgment, thereby ensuring the operating energy consumption cost under continuous monitoring conditions.

[0068] In this embodiment, when the risk warning terminal issues corresponding warning information according to the size of the risk value: Determine a range threshold interval corresponding to the risk value in a preset risk range value; and generate corresponding warning information according to the range threshold interval and preset template information corresponding to the range threshold interval.

[0069] It can be understood that the risk value predicted by the risk prediction model represents the probability of a water inrush risk. The larger the risk value, the greater the risk of a water inrush disaster under the current circumstances. The risk value is in the range of 0-1. Within this range, multiple value range intervals can be further divided according to the size of the risk value.

[0070] For example, there are three intervals of 0-0.4; 0.4-0.7; and 0.7-1, and each interval corresponds to a risk warning state. Specifically, the first interval 0-0.4 (including a risk value equal to 0) corresponds to a low-risk or no-risk interval. Within this interval, it is only necessary to remind relevant personnel to conduct normal daily inspections. The second interval 0.4-0.6 (including a risk value equal to 0.4) corresponds to a medium-risk interval. There is a certain tendency of sudden water disasters, but there is no risk of sudden water disasters for the time being. It is necessary to focus on defense and monitoring of this area. The third interval 0.7-1 (including risk values ​​equal to 0.7 and 1) corresponds to a high-risk interval. There is a risk of sudden water disasters, and the time of occurrence of the sudden water disaster risk is expected to be close to the current time. It is necessary to immediately disperse personnel and remove relevant equipment.

[0071] Among them, the first interval, the second interval and the third interval can be used as the range threshold intervals corresponding to the preset risk range values. When the risk value falls into the corresponding range threshold interval, the preset template information corresponding to the range threshold interval is selected, and the corresponding warning information is generated according to the preset template information.

[0072] In this embodiment, the data monitoring terminals are distributed in the target mine, wherein the data monitoring terminals are distributed according to the spatial range of the actual channel scene in the target mine and are arranged at a preset distance between adjacent data monitoring terminals within the spatial range.

[0073] It is understandable that in order to monitor the entire target mine in all directions, it is necessary to reasonably deploy data monitoring terminals according to the actual corridor direction of the target mine and the high-risk areas within the target mine. The main deployment method of the data monitoring terminals is distributed deployment, which will be based on the spatial range of the actual channel scene in the target mine and the preset distance between adjacent data monitoring terminals within the spatial range. Among them, the preset distance can be appropriately reduced in high-risk areas that require key monitoring.

[0074] The mine water inrush disaster warning system based on acoustic and optical fusion perception described in this embodiment includes: a data monitoring terminal, an edge computing node and a risk warning terminal; the data monitoring terminal is used to collect acoustic data, optical data and environmental parameters in the target mine through a preset collection device terminal, and make a preliminary risk judgment on the water inrush disaster in the target mine based on the acoustic data, the optical data and the environmental parameters. If the result of the preliminary risk judgment is that there is a risk, the acoustic data, the optical data and the environmental parameters are uploaded to the edge computing node; the edge computing node is used to, after receiving the acoustic data, optical data and environmental parameters sent by the data monitoring terminal, predict the risk value of the water inrush disaster in the target mine based on the acoustic data, the optical data and the environmental parameters through a risk prediction model pre-deployed at the edge computing node, and feed the risk value back to the risk warning terminal; the risk warning terminal is used to receive the risk value sent by the edge computing node, and According to the size of the risk value, corresponding early warning information is issued to remind relevant personnel of the current risk situation of the target mine. That is, in this application, a three-terminal interactive system consisting of a data monitoring end, an edge computing node and a risk early warning end is constructed. Through the data monitoring end, while collecting acoustic data, optical data and environmental data, a preliminary risk judgment is made for the three types of data. After judging that there is a risk, the above-mentioned multi-dimensional data is further sent to the edge computing node to make an accurate risk prediction through the risk prediction model arranged by the edge computing node, thereby realizing multi-level risk supervision and prediction of multi-dimensional data in the mine by constructing a preliminary risk judgment of the data monitoring end and an accurate prediction of the risk prediction model of the edge computing node. This not only ensures the reasonable practical application of multi-dimensional data, but also achieves the effect of accurate risk prediction through multi-dimensional data, thereby solving the technical problem that the single-dimensional information in the current mine is difficult to comprehensively and accurately capture the early signals of disaster incubation, which is prone to omissions or false alarms.

[0075] 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.

[0076] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or system 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 system. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not preclude the presence of other identical elements in the process, method, article, or system comprising the element.

[0077] The serial numbers of the above embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0078] The above are only preferred embodiments of the present application and do not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A mine water inrush disaster early warning system based on acoustic and optical fusion perception, characterized in that: The mine water inrush disaster warning system based on acoustic and optical fusion perception includes: a data monitoring end, an edge computing node and a risk warning end; the data monitoring end is used to collect acoustic data, optical data and environmental parameters in the target mine through a preset collection device end, and make a preliminary risk judgment on the water inrush disaster in the target mine based on the acoustic data, the optical data and the environmental parameters. If the result of the preliminary risk judgment is that there is a risk, the acoustic data, the optical data and the environmental parameters are uploaded to the edge computing node; the edge computing node is used to, after receiving the acoustic data, optical data and environmental parameters sent by the data monitoring end, predict the risk value of the water inrush disaster in the target mine based on the acoustic data, the optical data and the environmental parameters through a risk prediction model pre-deployed at the edge computing node, and feed the risk value back to the risk warning end; the risk warning end is used to receive the risk value sent by the edge computing node, and issue corresponding warning information based on the size of the risk value to remind relevant personnel of the current risk situation of the target mine.

2. The system according to claim 1, wherein When the edge computing node predicts the risk value of the water inrush disaster in the target mine based on the acoustic data, the optical data, and the environmental parameters using a risk prediction model pre-deployed at the edge computing node: determining a dynamic weight corresponding to a water inrush stage corresponding to the water inrush disaster in the target mine based on the acoustic data, the optical data, and the environmental parameters; Through the risk prediction model pre-deployed at the edge computing node, the risk value of the water inrush disaster in the target mine is predicted based on the acoustic data, the optical data and the dynamic weight, wherein the dynamic weight is used to influence the tendency proportion of the acoustic data and the optical data when the risk prediction model performs risk prediction.

3. The system according to claim 2, wherein: When the edge computing node determines the dynamic weight corresponding to the water inrush stage of the water inrush disaster in the target mine based on the acoustic data, the optical data, and the environmental parameters: obtaining a first rule standard corresponding to the water inrush disaster in the target mine at different water inrush stages, and obtaining a second rule standard corresponding to different interference environments in the target mine; determining a first confidence parameter of the acoustic data and the optical data based on the first rule standard; and determining a second confidence parameter corresponding to the environmental data based on the second rule standard; The dynamic weight corresponding to the water inrush stage of the water inrush disaster in the target mine is calculated according to the first confidence parameter, the second confidence parameter and the preset basic weights corresponding to the different confidence parameters.

4. The system according to claim 3, wherein: After the edge computing node predicts the risk value of the water inrush disaster in the target mine based on the acoustic data, the optical data and the environmental parameters: obtain the predicted historical risk value and label information corresponding to the historical risk value, wherein the label information is obtained by relevant operation and maintenance personnel based on the historical risk value and the actual environmental conditions of the target mine corresponding to the historical risk value; according to the label information, optimize the preset basic weights corresponding to the different confidence parameters.

5. The system according to claim 2, wherein: When the edge computing node determines the dynamic weight corresponding to the water inrush stage of the corresponding water inrush disaster in the target mine based on the acoustic data, the optical data and the environmental parameters, the method further: the acoustic data includes a sound event count and the energy of the main frequency band of the sound; if the sound event count is greater than a preset number of times and / or the change corresponding to the energy of the main frequency band of the sound is greater than a preset energy change threshold, then a first preset weight group is selected as the dynamic weight, wherein the first preset weight group includes an acoustic weight and an optical weight, and the acoustic weight is greater than the optical weight.

6. The system according to claim 2, wherein: When the edge computing node determines the dynamic weight corresponding to the water inrush stage of the corresponding water inrush disaster in the target mine based on the acoustic data, the optical data, and the environmental parameters, the method further comprises: the optical data including the maximum temperature difference in the infrared region and the visible light crack length; If the maximum temperature difference in the infrared region is greater than the preset temperature difference and / or the change in the visible light crack length is greater than the preset length change threshold, a second preset weight group is selected as the dynamic weight, wherein the second preset weight group includes acoustic weight and optical weight, and the optical weight is greater than the acoustic weight.

7. The system according to claim 1, wherein: When the data monitoring end makes a preliminary risk assessment of the water inrush disaster in the target mine based on the acoustic data, the optical data, and the environmental parameters: performing time synchronization processing on the acoustic data and the optical data based on the influence of the environmental data on the time difference between sound and light; Based on the synchronously processed data and preset judgment rules, a preliminary risk judgment is made on the water inrush disaster in the target mine.

8. The system according to claim 7, wherein: After the data monitoring end makes a preliminary risk assessment of the sudden water disaster in the target mine based on the acoustic data, the optical data and the environmental parameters, the method further includes: the acoustic data includes the sound event count and the sound main frequency band energy, and the optical data includes the maximum temperature difference in the infrared area and the visible light crack length; if any one of the sound event count and the visible light crack length is greater than the corresponding standard value, it is determined that there is an area to be inspected in the target mine, and a corresponding report is generated according to the corresponding situation of the area to be inspected; if the sound main frequency band energy and the maximum temperature difference in the infrared area are both greater than the corresponding standard values, it is determined that the result of the preliminary risk assessment of the target mine is that there is a risk.

9. The system according to claim 1, wherein: When the risk warning end issues corresponding warning information based on the size of the risk value: determine the range threshold interval corresponding to the risk value in the preset risk range value; generate corresponding warning information based on the range threshold interval and the preset template information corresponding to the range threshold interval.

10. The system according to claim 1, wherein: The data monitoring terminals are distributedly arranged in the target mine, wherein the data monitoring terminals are arranged at a preset distance between adjacent data monitoring terminals within the spatial range according to the spatial range of the actual channel scene in the target mine.

Images