Fire warning method and system for coal mines, and storage medium

Abstract

The present invention discloses a fire warning method and system, and a storage medium for coal mines. The present invention deploys a multimodal sensor network in a mine; after data cleaning, time synchronization and fusion processing, multimodal feature data is extracted, and a fire risk score is calculated through a fire risk assessment model; the fire risk feature data is uploaded to a blockchain network for encrypted storage, and is monitored in real time through a smart contract, and a warning signal is automatically triggered and an emergency response program is started; virtual reality and augmented reality technologies are combined to build a virtual model of the mine, and the fire risk area and the flame spread path are dynamically displayed; the control center starts the emergency response program according to the warning signal, and feeds back the emergency handling situation in real time through virtual reality and augmented reality equipment, and implements fire emergency drills and capacity improvement; this scheme effectively improves the accuracy of fire warning and the efficiency of emergency response, and reduces casualties and property losses caused by fire.

Description

Technical Field

[0001] The present invention relates to the technical field of coal mine fire safety, and in particular to a fire early warning method and system, and a storage medium for coal mines. Background Art

[0002] Coal mine fires are a major hidden danger in coal mine safety production. Coal mine fires not only cause a large amount of economic losses, but also lead to serious casualties and environmental pollution. Coal mine fires are usually accompanied by the rapid spread of high temperatures, smoke and toxic gases, which can easily cause gas and coal dust explosions, further aggravating the severity of the disaster. Establishing an effective fire early warning system that can detect fires in time at the early stage of a fire and quickly initiate emergency response procedures is an important measure to ensure coal mine safety production and protect the lives and property of miners.

[0003] The existing coal mine fire early warning system has the following main problems:

[0004] The monitoring data of a single type of sensor is limited, and environmental interference and equipment failure can easily lead to false alarms and missed alarms; relying solely on the monitoring of temperature and gas concentration, it is difficult to achieve early warning of fire, and the fire can often only be detected when it develops to a more serious stage; the monitoring range of a single sensor is limited, and it is impossible to achieve comprehensive monitoring of the entire mine; the data processing capacity of the existing system is limited, and real-time data fusion and analysis cannot be achieved, delaying the early warning response; the early warning information transmission and emergency response program initiation are not timely enough, increasing the risk of fire losses and casualties.

[0005] The existing technology (Chinese invention patent, publication number: CN116537881A) has the following deficiencies when facing the above problems:

[0006] The lack of comprehensive monitoring of multiple types of sensors results in single data and high false alarm and missed alarm rates; the data preprocessing is simple and cannot achieve fusion analysis of multimodal data, which affects the accuracy and timeliness of the early warning; data storage and processing are centralized, and there are risks of data tampering and single point failure, which affects the reliability of the system; monitoring mainly relies on sensor data and lacks an intuitive fire monitoring interface, which affects emergency response capabilities; the early warning information is transmitted slowly, and the emergency response procedure is not initiated in time, affecting the emergency handling effect. Summary of the invention

[0007] In view of the many problems existing in the above-mentioned prior art, the present invention provides a fire warning method and system and storage medium for coal mines. The present invention deploys a multimodal sensor network to collect temperature, gas concentration, thermal imaging images, sound waves and vibration data in the mine in real time; uses smart contracts and blockchain technology to encrypt, store and monitor fire risk characteristic data, and automatically triggers a warning signal and starts an emergency response program when the data reaches a preset threshold; combines virtual reality and augmented reality technology to build a virtual model of the mine, dynamically displaying the fire risk area and the flame spread path; and continuously improves emergency response capabilities through regular emergency drills and feedback mechanisms. Early warning, precise positioning and rapid response to coal mine fires are achieved, the accuracy of fire monitoring and the efficiency of emergency handling are improved, and casualties and property losses caused by fires are reduced.

[0008] A fire early warning method for coal mines comprises the following steps:

[0009] Deploy a multimodal sensor network in the mine to collect temperature data, gas concentration data, thermal imaging image data, sound wave data and vibration data in the mine in real time, and preliminarily preprocess and format the data through an underground exchange device to generate a formatted sensor data set;

[0010] The formatted sensor data set is subjected to data cleaning, time synchronization and fusion processing, multimodal feature data is extracted, and a fire risk score is calculated through feature screening and a fire risk assessment model to generate fire risk feature data;

[0011] The fire risk characteristic data is uploaded to the blockchain network for encrypted storage, and the fire risk characteristic data is monitored in real time through smart contracts. When an abnormal situation is detected, an early warning signal is automatically triggered and an emergency response program is initiated;

[0012] Using the geographic information data of the mine to build a virtual model, and mapping the fire risk characteristic data to the virtual model, displaying the fire risk area, flame spread path and multi-dimensional data through virtual reality and augmented reality technology, and generating virtual monitoring view data;

[0013] The virtual monitoring view data is displayed in real time on the virtual fire monitoring interface. The control center initiates the emergency response procedure according to the received warning signal, and provides real-time feedback on the emergency handling situation through virtual reality and augmented reality devices, thereby implementing fire emergency drills and capacity improvement.

[0014] Preferably, the multimodal sensor network includes self-organization and self-repairing functions to ensure the stability and continuity of data transmission. When a sensor node fails, other nodes automatically reorganize the network and continue to transmit data through redundant paths.

[0015] Preferably, the preliminary pretreatment comprises the following steps:

[0016] Noise filtering of the formatted sensor data set includes using a filtering algorithm to remove high-frequency noise from the sensor data;

[0017] Interpolation processing is performed on missing data, including using linear interpolation or spline interpolation methods to fill in missing values ​​in sensor data and generate a cleaned sensor data set.

[0018] Preferably, the time synchronization and fusion processing comprises the following steps:

[0019] Time synchronization of data from different sensors, including using timestamp alignment methods to ensure that all sensor data are aligned on the same time axis;

[0020] The multimodal data fusion algorithm is used to fuse the time-synchronized data and extract multimodal feature data.

[0021] Preferably, the fire risk assessment model is generated by the following steps:

[0022] Use feature screening algorithms to evaluate and screen out the features that have the greatest impact on fire risk and generate a key feature dataset;

[0023] Train and validate key feature datasets through machine learning or deep learning models to generate fire risk assessment models;

[0024] The real-time key feature data set is input into the fire risk assessment model to calculate the fire risk score and generate fire risk feature data.

[0025] Preferably, the fire risk characteristic data is encrypted and stored through a blockchain network and monitored in real time through a smart contract, including the following steps:

[0026] Encrypting fire risk characteristic data to generate encrypted data;

[0027] Upload the encrypted data to the blockchain network, where it will be stored and verified by each node;

[0028] Smart contracts automatically monitor fire risk characteristic data and automatically trigger warning signals when they detect that temperature data exceeds a preset threshold or the rate of change exceeds a preset rate threshold;

[0029] Smart contracts automatically monitor fire risk characteristic data, automatically trigger warning signals, and initiate emergency response procedures;

[0030] The response mechanism of the smart contract in fire warning includes:

[0031] Numerical thresholds, including triggering an alarm when the set temperature exceeds a specific value and triggering an alarm when the set key gas concentration exceeds a specific value;

[0032] Value change rate thresholds, including triggering an early warning when the temperature is detected to rise rapidly in a short period of time and triggering an early warning when the concentration of key gases is detected to increase rapidly in a short period of time;

[0033] Anomaly detection where values ​​influence each other, including triggering an alert when the concentration of key gases increases significantly while the temperature rises, and conducting a comprehensive analysis based on multiple parameters, triggering an alert when these parameters exceed preset thresholds at the same time;

[0034] Trend analysis, including long-term trend detection, monitoring the long-term trend of temperature and gas concentration, triggering an alert when the trend continues to rise and reaches a preset threshold; and short-term fluctuation detection, monitoring the short-term fluctuation of data, triggering an alert when the data fluctuates abnormally in the short term;

[0035] Multi-point anomaly detection, including spatial distribution detection, monitoring data from multiple sensor nodes and triggering an alert when multiple nodes detect anomalies at the same time; and regional diffusion detection, monitoring the spatial diffusion of fire risk and triggering an alert when the fire risk spreads from one node to multiple adjacent nodes;

[0036] Complex event processing, including compound event detection, combining multiple data sources and multiple threshold conditions to detect compound events, triggering warnings when multiple conditions are met at the same time; and dynamic threshold adjustment, where smart contracts dynamically adjust thresholds based on actual conditions.

[0037] Preferably, the virtual model comprises the following steps:

[0038] Collect geographic information data of the mine, including the mine layout, passages and equipment locations, and build a virtual geographic model;

[0039] Fire risk characteristic data are mapped into a virtual geographic model, and fire risk areas, flame spread paths and multi-dimensional data are dynamically displayed through virtual reality and augmented reality technologies to generate virtual monitoring view data.

[0040] Preferably, the emergency response procedure comprises the following steps:

[0041] The control center determines the emergency response level based on the received warning signal and initiates the corresponding emergency response procedures, including evacuation, fire extinguishing and equipment protection;

[0042] Dispatch emergency resources, including fire brigade, medical team and emergency equipment, to ensure efficient execution of emergency response;

[0043] Real-time feedback of emergency response status through virtual reality and augmented reality devices, updating information on the virtual fire monitoring interface;

[0044] Conduct fire emergency drills regularly, evaluate drill effectiveness, and formulate improvement measures based on the evaluation results to enhance emergency response capabilities.

[0045] A fire early warning system for a coal mine, comprising:

[0046] A multimodal sensor network is deployed inside the mine to collect temperature data, gas concentration data, thermal imaging image data, sound wave data, and vibration data in real time.

[0047] A downhole exchange device connected to the multimodal sensor network and used to preliminarily preprocess and format the data to generate a formatted sensor data set;

[0048] A data processing module, used to perform data cleaning, time synchronization and fusion processing on the formatted sensor data set, extract multimodal feature data, calculate the fire risk score through feature screening and fire risk assessment model, and generate fire risk feature data;

[0049] A blockchain storage module is used to upload the fire risk characteristic data to the blockchain network for encrypted storage, and to monitor the fire risk characteristic data in real time through smart contracts, and to automatically trigger an early warning signal and start an emergency response program when an abnormal situation is detected;

[0050] A virtual model building module is used to build a virtual model using the geographic information data of the mine, and map the fire risk characteristic data to the virtual model, display the fire risk area, flame spread path and multi-dimensional data through virtual reality and augmented reality technology, and generate virtual monitoring view data;

[0051] The virtual fire monitoring interface is used to display the virtual monitoring view data in real time on the virtual fire monitoring interface. The control center starts the emergency response program according to the received early warning signal, and provides real-time feedback on the emergency handling situation through virtual reality and augmented reality equipment, so as to implement fire emergency drills and capacity improvement.

[0052] A storage medium stores a computer program, which, when executed by a processor, implements the steps of the fire warning method for coal mines.

[0053] Compared with the prior art, the advantages and beneficial effects of the present invention are:

[0054] The present invention uses a multimodal sensor network, including temperature, gas, infrared camera, sound wave and vibration sensors, with comprehensive coverage and diversified data. The network has self-organization and self-repair capabilities, which improves the reliability and accuracy of monitoring;

[0055] The present invention pre-processes the data through the downhole exchange device, uses the deep learning algorithm and data fusion technology to comprehensively analyze the data of multiple sensors, extracts the fire risk characteristics, generates fire risk characteristic data, improves the detection accuracy, and reduces false alarms and missed alarms;

[0056] Fire risk characteristic data is uploaded to the blockchain network to ensure data transparency, immutability and traceability; the present invention realizes automated early warning and emergency response through smart contracts, thus improving data credibility and system reliability;

[0057] VR and AR technologies are used to build a virtual model of the mine, superimpose fire risk characteristic data in real time, and provide an intuitive fire monitoring interface. The present invention displays the fire risk area and flame spread through a virtual fire monitoring interface, thereby improving emergency response capabilities;

[0058] The virtual fire monitoring interface displays the fire situation in real time. Once the warning signal is triggered, the control center and on-site staff are notified immediately. The present invention uses VR and AR technology to conduct fire emergency drills, thereby improving the speed and efficiency of emergency response.

[0059] After the emergency response is completed, the actual fire data and emergency response data are uploaded to the blockchain network to generate fire response data. The present invention optimizes and improves the system by analyzing the fire response data, thereby improving the early warning and response capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a schematic diagram of the process of the present invention;

[0061] Figure 2 It is a schematic diagram of data processing in the present invention;

[0062] Figure 3 This is a diagram of the smart contract mechanism in the present invention;

[0063] Figure 4 It is a structural block diagram of the system of the present invention. DETAILED DESCRIPTION

[0064] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present disclosure. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present disclosure. However, it is obvious that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.

[0065] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "comprise", "include", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0066] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification, and should not be interpreted in an idealized or overly rigid manner.

[0067] In the case of using expressions such as "at least one of A, B, and C, etc.", it should generally be interpreted in accordance with the meaning of the expression generally understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.). In the case of using expressions such as "at least one of A, B, or C, etc.", it should generally be interpreted in accordance with the meaning of the expression generally understood by those skilled in the art (for example, "a system having at least one of A, B, or C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0068] Some block diagrams and / or flow charts are shown in the accompanying drawings. It should be understood that some boxes or combinations thereof in the block diagrams and / or flow charts can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that these instructions can create a device for implementing the functions / operations described in these block diagrams and / or flow charts when executed by the processor. The technology of the present disclosure can be implemented in the form of hardware and / or software (including firmware, microcode, etc.). In addition, the technology of the present disclosure can take the form of a computer program product on a computer-readable storage medium storing instructions, which can be used by an instruction execution system or used in conjunction with an instruction execution system.

[0069] like Figure 1-Figure 2 As shown, a fire early warning method for a coal mine comprises the following steps:

[0070] Deploy a multimodal sensor network in the mine to collect temperature data, gas concentration data, thermal imaging image data, sound wave data and vibration data in the mine in real time, and preliminarily preprocess and format the data through an underground exchange device to generate a formatted sensor data set;

[0071] Preferably, the multimodal sensor network includes self-organization and self-repairing functions to ensure the stability and continuity of data transmission. When a sensor node fails, other nodes automatically reorganize the network and continue to transmit data through redundant paths.

[0072] The self-organizing function allows sensor nodes to automatically establish a network topology after deployment without human intervention. Each sensor node has the ability to discover and connect to each other, and can automatically identify surrounding nodes and establish communication links. Each sensor node performs a handshake protocol with neighboring nodes through broadcast signals to exchange location information and communication parameters. Based on the received signal strength indication (RSSI) and the energy status of the node, the nodes dynamically select the best communication path. A stable network topology is automatically established to ensure the reliability and effectiveness of data transmission.

[0073] The self-organizing function also includes a load balancing mechanism to ensure that no single node in the network is overloaded by too much data transmission. The sensor node dynamically adjusts the communication path according to its own working status and network traffic. When the traffic of a node is too large, other nodes will share part of the traffic. This improves the overall performance and stability of the network and extends the service life of the nodes.

[0074] The self-healing function allows the sensor network to automatically detect faults and isolate faulty nodes when nodes fail, ensuring the normal operation of the network. Each node sends health reports to neighboring nodes regularly. Once a node fails to send a report within the specified time, its neighboring nodes will identify the node as a faulty node and update the network topology to isolate the faulty node. Quickly identify and isolate faulty nodes, reduce the impact of faults on the network, and ensure the continuity of data transmission.

[0075] After a node fails, other nodes can automatically reorganize the network and reestablish communication paths to maintain network connectivity. When a node failure is detected, neighboring nodes rebroadcast handshake signals to find new communication partners and update their routing tables. The self-healing algorithm selects a new communication path based on the shortest path and energy optimization principles. Even when multiple nodes fail, the network can still maintain connectivity, ensuring that data can continue to be transmitted by bypassing the failed node.

[0076] The self-healing function also includes using redundant paths to continue transmitting data to ensure the stability of data transmission. Each node in the network maintains multiple backup paths. When the main path fails, the node can immediately switch to the backup path for data transmission. The selection of the backup path is based on the length of the path, the node energy and the stability of the path. Reducing the interruption time of data transmission, improving the robustness of the network and the reliability of data transmission.

[0077] Through the self-organization and self-repairing functions of the present invention, the sensor network can automatically adapt to environmental changes and maintain the stability and continuity of data transmission. Even in the case of node failure or complex network environment, data transmission can still proceed smoothly. The self-organization and self-repairing functions enable the network to quickly adjust and recover in the face of single or multiple node failures, ensuring the stability and reliability of the entire system. Even in a complex environment such as a mine, the high availability of the network can be maintained. Through load balancing and path optimization, the self-organization and self-repairing functions can effectively distribute network traffic and power consumption, reduce excessive consumption of a single node, thereby extending the service life of the sensor node and improving the sustainability of the entire network.

[0078] In one embodiment, the processing of node failure, scenario: in a certain area of ​​a mine, sensor node A fails due to battery exhaustion.

[0079] Neighboring nodes B and C fail to receive the health report of node A within the specified time and determine that A has failed.

[0080] Nodes B and C rebroadcast the handshake signal to find a new communication partner.

[0081] Nodes B and C update their routing tables and select a new path for data transmission, bypassing node A.

[0082] In the event of node A failure, the network can quickly adjust the communication path to ensure uninterrupted data transmission.

[0083] In another embodiment, the application of load balancing is in the following scenario: the traffic of sensor node D in a certain area is too large, resulting in an overload situation.

[0084] Node D detects that it is overloaded and broadcasts its load status through the network.

[0085] After receiving the load status, neighboring nodes E and F adjust part of the data traffic to themselves to share the pressure of D.

[0086] Nodes D, E, and F update the routing table to optimize path selection.

[0087] By dynamically adjusting the communication path, the load of node D is balanced and the overall network performance is improved.

[0088] The self-organization and self-repairing functions of the multimodal sensor network ensure the data transmission stability, continuity and robustness of the coal mine fire early warning system by automatically establishing and adjusting the network topology, detecting and isolating faulty nodes, and continuing to transmit data through redundant paths.

[0089] Preferably, the preliminary pretreatment comprises the following steps:

[0090] Noise filtering of the formatted sensor data set includes using a filtering algorithm to remove high-frequency noise from the sensor data;

[0091] Interpolation processing is performed on missing data, including using linear interpolation or spline interpolation methods to fill in missing values ​​in sensor data and generate a cleaned sensor data set.

[0092] In coal mine environments, sensor data is often interfered by high-frequency noise, which may come from electromagnetic interference, mechanical vibration, environmental background noise, etc. High-frequency noise usually manifests itself as dramatic fluctuations in data within a short period of time, which can mask the true sensor signal and affect the accuracy and stability of the data.

[0093] Commonly used filtering algorithms include low-pass filter, Kalman filter and median filter. Low-pass filter can effectively filter out high-frequency components and retain low-frequency signals. Its working principle is to smooth the data curve by calculating the moving average of the signal. Kalman filter is a recursive filter that filters out noise by estimating the current state and predicting the future state. Its advantage is that it can adjust the filtering parameters in real time in a dynamic environment. Median filter effectively filters out sharp impulse noise by taking the median of the neighborhood data instead of the center value.

[0094] The filtering algorithm can significantly reduce high-frequency noise in the data, improve the accuracy and stability of the data, and lay the foundation for subsequent processing and analysis.

[0095] The formatted sensor data set is subjected to data cleaning, time synchronization and fusion processing, multimodal feature data is extracted, and a fire risk score is calculated through feature screening and a fire risk assessment model to generate fire risk feature data;

[0096] Preferably, the time synchronization and fusion processing comprises the following steps:

[0097] Time synchronization of data from different sensors, including using timestamp alignment methods to ensure that all sensor data are aligned on the same time axis;

[0098] The multimodal data fusion algorithm is used to fuse the time-synchronized data and extract multimodal feature data.

[0099] In the coal mine fire warning system, data collected by different sensors have different timestamps, and the accuracy of fire warning depends on the synchronous analysis of multimodal data. Ensuring that all sensor data are aligned on the same time axis is the basis for data fusion and feature extraction. Time synchronization applies a unified timestamp standard to all sensor data, aligning the data of each sensor on the same time axis and eliminating data deviations caused by time differences.

[0100] Timestamp alignment methods include hardware synchronization and software synchronization.

[0101] Hardware synchronization: Ensure that the internal clocks of all sensors are consistent through precise clock synchronization mechanisms, such as GPS time synchronization. Software synchronization: Apply a unified timestamp to the collected data through post-processing technology. The specific steps include:

[0102] Collect timestamped data: When each sensor collects data, its timestamp is recorded.

[0103] Calculate time deviation: Calculate the time deviation of each sensor by comparing the timestamps of each sensor.

[0104] Adjust timestamps: Apply a time offset to correct the data timestamps and align all data to a unified timeline.

[0105] Improves the accuracy and consistency of data synchronization by ensuring that all sensor data is aligned on the same time axis.

[0106] Preferably, the fire risk assessment model is generated by the following steps:

[0107] Use feature screening algorithms to evaluate and screen out the features that have the greatest impact on fire risk and generate a key feature dataset;

[0108] Train and validate key feature datasets through machine learning or deep learning models to generate fire risk assessment models;

[0109] The real-time key feature data set is input into the fire risk assessment model to calculate the fire risk score and generate fire risk feature data.

[0110] In the coal mine fire early warning system, the large amount of data collected by sensors contains many different types of features. Screening out the features that have the greatest impact on fire risk is the basis for establishing an efficient assessment model. Feature screening evaluates the relationship between each feature and fire risk, screens out those features that have a significant impact on fire risk prediction, and forms a key feature data set.

[0111] Commonly used feature screening algorithms include Random Forest, LASSO regression and Principal Component Analysis (PCA).

[0112] Random Forest: Select features by building multiple decision trees and evaluating the importance of each feature in the decision tree. The principle is to evaluate the importance of features through the splitting nodes of the tree. The more important the feature, the greater its contribution to the model.

[0113] LASSO regression: By introducing the L1 regularization term, feature selection is performed simultaneously during the linear regression process. The principle is to reduce the regression coefficients of unimportant features to zero, thereby screening out important features.

[0114] Principal component analysis: The original features are converted into a small number of principal components through linear transformation. These principal components can retain the variance of the original data to the greatest extent. The principle is to use the covariance matrix between features to extract the main features.

[0115] Feature screening algorithms can effectively filter out the most important features from large amounts of data, reduce data dimensions, and improve model training efficiency and prediction performance.

[0116] In one embodiment, the scenario is: sensor data such as temperature, gas concentration, and thermal imaging images in a mine contain many features, and key features need to be screened out for fire risk assessment.

[0117] Collect sensor data to form an initial feature set.

[0118] The random forest algorithm is used to evaluate the importance of each feature and screen out important features.

[0119] Generate key feature datasets.

[0120] Filter out the most valuable features for fire risk prediction, reduce data dimensions, and improve model training efficiency.

[0121] The core of the fire warning system is to generate an assessment model that can accurately predict fire risks by training and validating key feature data sets. Machine learning or deep learning models learn the relationship between each feature and fire risk by training historical data, so as to accurately assess the risk of real-time data.

[0122] The core of the fire warning system is to generate an assessment model that can accurately predict fire risks by training and validating key feature data sets. Machine learning or deep learning models learn the relationship between each feature and fire risk by training historical data, so as to accurately assess the risk of real-time data.

[0123] Machine learning and deep learning models, commonly used models include decision trees, support vector machines (SVM), neural networks, etc.

[0124] Decision tree: Classify data through a tree structure and learn the impact of each feature on fire risk. Its principle is to classify data based on the split nodes of the feature and finally generate a decision path.

[0125] Support vector machine: Classify data by constructing a hyperplane in high-dimensional space. Its principle is to find the hyperplane that maximizes the classification interval, which can effectively process linearly inseparable data.

[0126] Neural network: Learn complex nonlinear relationships through the connection of multiple layers of neurons. Its principle is to adjust the weights of neurons through the back propagation algorithm and gradually optimize the model.

[0127] Through training and verification of key feature data sets, the generated fire risk assessment model can accurately predict fire risks and improve the timeliness and accuracy of early warnings.

[0128] like Figure 3 As shown, the fire risk characteristic data is uploaded to the blockchain network for encrypted storage, and the fire risk characteristic data is monitored in real time through smart contracts. When an abnormal situation is detected, an early warning signal is automatically triggered and an emergency response program is started;

[0129] Preferably, the fire risk characteristic data is encrypted and stored through a blockchain network and monitored in real time through a smart contract, including the following steps:

[0130] Encrypting fire risk characteristic data to generate encrypted data;

[0131] Upload the encrypted data to the blockchain network, where it will be stored and verified by each node;

[0132] In the coal mine fire early warning system, fire risk characteristic data involves real-time safety information in the mine, and it is crucial to ensure the confidentiality and integrity of this data during transmission and storage. Encryption processing converts fire risk characteristic data into difficult-to-crack ciphertext to prevent unauthorized access and tampering, thereby ensuring data security and privacy.

[0133] Commonly used encryption algorithms include symmetric encryption algorithms (such as AES) and asymmetric encryption algorithms (such as RSA).

[0134] Symmetric encryption algorithm (AES): Encryption and decryption are performed using a single key, with high encryption performance. Its principle is to convert plaintext into ciphertext through a series of substitution and replacement operations.

[0135] Asymmetric encryption algorithm (RSA): uses a pair of keys for encryption and decryption, the public key is used for encryption, and the private key is used for decryption, ensuring the security of data during transmission. Its principle is to use the mathematical problem of large prime number decomposition to generate public and private keys.

[0136] Encryption processing can effectively protect the confidentiality and integrity of fire risk characteristic data, prevent data leakage and tampering, and improve the security of data transmission and storage.

[0137] Smart contracts automatically monitor fire risk characteristic data and automatically trigger warning signals when they detect that temperature data exceeds a preset threshold or the rate of change exceeds a preset rate threshold;

[0138] Smart contracts automatically monitor fire risk characteristic data, automatically trigger warning signals, and initiate emergency response procedures;

[0139] The response mechanism of the smart contract in fire warning includes:

[0140] Numerical thresholds, including triggering an alarm when the set temperature exceeds a specific value and triggering an alarm when the set key gas concentration exceeds a specific value;

[0141] Value change rate thresholds, including triggering an early warning when the temperature is detected to rise rapidly in a short period of time and triggering an early warning when the concentration of key gases is detected to increase rapidly in a short period of time;

[0142] Anomaly detection where values ​​influence each other, including triggering an alert when the concentration of key gases increases significantly while the temperature rises, and conducting a comprehensive analysis based on multiple parameters, triggering an alert when these parameters exceed preset thresholds at the same time;

[0143] Trend analysis, including long-term trend detection, monitoring the long-term trend of temperature and gas concentration, triggering an alert when the trend continues to rise and reaches a preset threshold; and short-term fluctuation detection, monitoring the short-term fluctuation of data, triggering an alert when the data fluctuates abnormally in the short term;

[0144] Multi-point anomaly detection, including spatial distribution detection, monitoring data from multiple sensor nodes and triggering an alert when multiple nodes detect anomalies at the same time; and regional diffusion detection, monitoring the spatial diffusion of fire risk and triggering an alert when the fire risk spreads from one node to multiple adjacent nodes;

[0145] Complex event processing, including compound event detection, combining multiple data sources and multiple threshold conditions to detect compound events, triggering warnings when multiple conditions are met at the same time; and dynamic threshold adjustment, where smart contracts dynamically adjust thresholds based on actual conditions.

[0146] In the coal mine fire early warning system, smart contracts can automatically monitor fire risk characteristic data, and automatically trigger early warning signals and start emergency response procedures when abnormal conditions are detected. This automated mechanism can significantly reduce human intervention and ensure the real-time and reliability of the early warning system. Smart contracts are automated programs running on the blockchain that monitor fire risk characteristic data through preset logic and conditions. When the monitored data exceeds the preset threshold or the rate of change exceeds the preset rate threshold, the smart contract will automatically execute the early warning operation and trigger the corresponding emergency response program. Real-time monitoring and early warning are achieved to ensure that fire risks are responded to in the first place, improve the accuracy and timeliness of the early warning system, and reduce losses caused by fire.

[0147] Set numerical thresholds for fire risk characteristic data, including temperature thresholds and key gas concentration thresholds. When the monitored data exceeds these thresholds, the smart contract will trigger an early warning signal. By analyzing historical data, determine the critical values ​​of temperature and gas concentration. These critical values ​​represent potential fire risks. The smart contract will continue to monitor these data, and when a certain data exceeds the preset threshold, the early warning signal will be triggered immediately. Real-time monitoring and early warning of key parameters are achieved to ensure rapid response when the fire risk increases and reduce the possibility of fire.

[0148] Smart contracts monitor the rate of change of fire risk characteristic data, and trigger warning signals when they detect a rapid rise in temperature or gas concentration in a short period of time. By analyzing the rate of change of data, smart contracts can detect abnormally rapid changes. These rapid changes may be early signs of fire. Smart contracts set thresholds for the rate of change of temperature and the rate of change of gas concentration, and trigger warning signals when the rate of change exceeds the preset value. Real-time monitoring and early warning of the rate of change of data can detect fire risks earlier and improve the sensitivity and response speed of the early warning system.

[0149] Smart contracts monitor the rate of change of fire risk characteristic data, and trigger warning signals when they detect a rapid rise in temperature or gas concentration in a short period of time. By analyzing the rate of change of data, smart contracts can detect abnormally rapid changes. These rapid changes may be early signs of fire. Smart contracts set thresholds for the rate of change of temperature and the rate of change of gas concentration, and trigger warning signals when the rate of change exceeds the preset value. Real-time monitoring and early warning of the rate of change of data can detect fire risks earlier and improve the sensitivity and response speed of the early warning system.

[0150] Smart contracts monitor the correlation between temperature and gas concentration, and trigger an early warning signal when the temperature rises and the gas concentration increases significantly. Through multivariate analysis, smart contracts can identify the correlation between temperature and gas concentration. When abnormal changes in both are detected at the same time, an early warning signal is triggered. This method can more accurately identify fire risks. Improve the accuracy of fire risk detection, ensure timely response when multivariate abnormal changes occur, and improve the reliability of the early warning system.

[0151] Smart contracts monitor the changing trends of temperature and gas concentrations through long-term trend detection and short-term fluctuation detection. When the trend continues to rise or the short-term fluctuation is abnormal, the warning signal is triggered. By analyzing historical data, smart contracts can identify the long-term trend and short-term fluctuation of temperature and gas concentration. When the trend continues to rise and reaches the preset threshold, or when the data fluctuates abnormally in the short term, the warning signal is triggered. Monitoring the trend of data changes can predict fire risks in advance and improve the foresight and response speed of the early warning system.

[0152] Smart contracts monitor the data of multiple sensor nodes through spatial distribution detection and regional diffusion detection. When multiple nodes detect anomalies at the same time or the fire risk spreads from one node to multiple nodes, an early warning signal is triggered. Through spatial distribution analysis, smart contracts can identify data anomalies of multiple sensor nodes. When multiple nodes detect anomalies at the same time, an early warning signal is triggered. In addition, through regional diffusion detection, smart contracts can identify the spatial diffusion of fire risks. When the fire risk spreads from one node to multiple adjacent nodes, an early warning signal is triggered. Joint monitoring of multiple sensor nodes is achieved to ensure timely early warning when multiple points are abnormal or regional diffusion occurs, and to improve the accuracy and coverage of the early warning system.

[0153] Smart contracts improve the flexibility and accuracy of early warnings through composite event detection and dynamic threshold adjustment, combined with multiple data sources and multiple threshold conditions for comprehensive analysis. Composite event detection combines multiple data sources (such as temperature, gas concentration, vibration data, etc.) and multiple threshold conditions for comprehensive analysis, and triggers early warning signals when multiple conditions are met at the same time. Dynamic threshold adjustment improves the flexibility and accuracy of early warnings by dynamically adjusting thresholds according to actual conditions (such as working conditions in the mine, seasonal changes, etc.). Improve the comprehensive analysis capabilities of the early warning system to ensure timely response when complex events occur, and improve the flexibility and accuracy of early warnings.

[0154] In the present invention, the smart contract realizes the real-time monitoring and early warning functions of the coal mine fire early warning system through numerical threshold monitoring, numerical change rate monitoring, abnormal detection of mutual influence between numerical values, trend analysis, multi-point anomaly detection and complex event processing, improves the accuracy, timeliness and flexibility of the early warning, and ensures the safe operation of the mine.

[0155] Using the geographic information data of the mine to build a virtual model, and mapping the fire risk characteristic data to the virtual model, displaying the fire risk area, flame spread path and multi-dimensional data through virtual reality and augmented reality technology, and generating virtual monitoring view data;

[0156] In order to build an accurate virtual model of a mine, it is necessary to collect detailed geographic information data of the mine. This data includes the layout, passages, equipment locations, etc. of the mine. By using laser scanning, total station measurement and 3D modeling technology, accurate geographic information data of the mine is obtained. This data is input into a computer and processed by modeling software to generate a 3D virtual model of the mine.

[0157] Laser scanning: Use laser scanning equipment to perform three-dimensional scanning of the mine to obtain detailed information about the internal structure of the mine. The laser scanning data is converted into a visual three-dimensional model through point cloud processing technology.

[0158] Total station survey: Use total stations to conduct precise surveys to obtain information about the mine’s layout and passageways. These survey data are combined with laser scanning data to further refine the 3D model.

[0159] 3D modeling software: All collected geographic information data is input into 3D modeling software for data integration and model generation. Commonly used software includes AutoCAD, Revit and 3ds Max.

[0160] The generated three-dimensional virtual model can accurately reflect the internal structure and layout of the mine, providing a basis for subsequent fire risk data mapping and virtual reality display.

[0161] Mapping fire risk characteristic data to the virtual model can realize the visualization of fire risk and help monitoring personnel intuitively understand the fire risk area and flame spread path. Through data mapping technology, real-time monitoring data (such as temperature, gas concentration, etc.) are associated with the virtual model, and these data are displayed in the virtual model in a visual way.

[0162] Real-time monitoring data is transferred to the virtual model through programming interfaces (such as API). Fire risk characteristic data is intuitively displayed in the model using visualization techniques such as color, shape, and animation. The virtual model needs to be synchronized with real-time data to ensure that the data displayed in the model can reflect the actual situation in real time. Use data streaming technology to achieve real-time data update and synchronization.

[0163] The present invention realizes the visual display of fire risk characteristic data, enables monitoring personnel to intuitively understand the fire risk area and flame spread path, and improves the accuracy and timeliness of risk monitoring and emergency decision-making.

[0164] Through virtual reality technology, the three-dimensional virtual model of the mine and the fire risk characteristic data are presented in the virtual reality equipment, so that the monitoring personnel can immersively observe and analyze the fire risk. Using VR head display equipment (such as Oculus Rift, HTC Vive) and VR software, the virtual model and real-time data are integrated to generate an immersive virtual monitoring view.

[0165] Through the programming interface, the virtual model and real-time data are transmitted to the VR device. The VR device provides an immersive monitoring experience through three-dimensional display and motion tracking. Monitoring personnel can observe the fire risk area and the flame spread path through the VR device.

[0166] Through augmented reality technology, virtual models and fire risk characteristic data are superimposed on real scenes, allowing monitoring personnel to observe and analyze fire risks in real environments. Using AR devices (such as Microsoft HoloLens, MagicLeap) and AR software, virtual models and real-time data are superimposed on real scenes to generate augmented reality views.

[0167] Through the programming interface, the virtual model and real-time data are transmitted to the AR device. The AR device superimposes the virtual model and data on the real scene through the camera and display screen to provide an augmented reality experience. Monitoring personnel can observe the fire risk area and the flame spread path through the AR device.

[0168] The application of VR and AR technologies has improved the intuitiveness and immersion of fire risk monitoring, enabling monitoring personnel to understand and analyze fire risks more comprehensively and improve the accuracy and timeliness of emergency response.

[0169] Example 1: Constructing a virtual model of a mine. Scenario: Using laser scanning and total station measurement technology, a three-dimensional virtual model of a mine is constructed.

[0170] Use laser scanning equipment to perform three-dimensional scanning of the mine and obtain point cloud data of the mine's internal structure.

[0171] Use total stations to conduct precise surveys and obtain information on mine layout and access.

[0172] The laser scanning data and total station measurement data are input into the 3D modeling software to generate a 3D virtual model of the mine.

[0173] The generated 3D virtual model accurately reflects the internal structure and layout of the mine, providing a basis for subsequent data mapping and display.

[0174] Embodiment 2: Mapping of fire risk characteristic data, scenario: Mapping fire risk characteristic data into a virtual model for visual display.

[0175] Collect real-time monitoring data, including fire risk characteristic data such as temperature and gas concentration.

[0176] Through the programming interface, real-time data is transferred to the virtual model for data mapping.

[0177] Using color, shape and animation techniques, fire risk characteristic data is visualized in a virtual model.

[0178] Realize visual display of fire risk characteristic data, so that monitoring personnel can intuitively understand the fire risk area and flame spread path.

[0179] Example 3: Application of virtual reality and augmented reality technologies, scenario: Using virtual reality and augmented reality technologies to display fire risk areas and flame spread paths.

[0180] The virtual model and real-time data are transmitted to the VR head display device to generate an immersive virtual monitoring view.

[0181] The virtual model and real-time data are transmitted to the AR device to generate an augmented reality view, which overlays the virtual data on the real scene.

[0182] Through VR and AR technology, an immersive and augmented reality fire risk monitoring experience is provided to improve the accuracy and timeliness of monitoring and emergency response.

[0183] The present invention constructs a virtual model by utilizing the geographic information data of the mine and maps the fire risk characteristic data into the virtual model. By combining virtual reality and augmented reality technologies, the present invention can realize the intuitive display and monitoring of coal mine fire risks and improve the accuracy and timeliness of early warning and emergency response.

[0184] Preferably, the virtual model comprises the following steps:

[0185] Collect geographic information data of the mine, including the mine layout, passages and equipment locations, and build a virtual geographic model;

[0186] Fire risk characteristic data are mapped into a virtual geographic model, and fire risk areas, flame spread paths and multi-dimensional data are dynamically displayed through virtual reality and augmented reality technologies to generate virtual monitoring view data.

[0187] In order to build an accurate virtual model of a mine, it is necessary to collect detailed geographic information data of the mine, including the mine's layout, channels, and equipment locations. These data provide the basis for the construction of the virtual model. Accurate geographic information data of the mine is obtained by using a variety of measurement technologies, such as laser scanning, total station measurement, and geographic information system (GIS). These data are processed and integrated to form a complete three-dimensional model of the mine.

[0188] Laser scanning: Use laser scanning equipment to scan the interior of the mine to obtain detailed 3D point cloud data. Laser scanning can quickly and efficiently capture the complex structure inside the mine.

[0189] Total station survey: Use a total station to conduct precise surveys to obtain detailed information about the mine’s layout, passageways, and equipment locations. The total station survey data is combined with laser scanning data to further improve the accuracy of the model.

[0190] Geographic Information System (GIS): GIS technology is used to manage and analyze geographic information data, integrating the spatial data and attribute data of the mine to form a comprehensive geographic information database.

[0191] The generated geographic information data can accurately reflect the internal structure and layout of the mine, and provide reliable data support for building a virtual geographic model.

[0192] In one embodiment, scenario: constructing a three-dimensional virtual model of a mine.

[0193] Use laser scanning equipment to perform three-dimensional scanning of the mine and obtain point cloud data inside the mine.

[0194] Use total stations to measure mine layouts, passages and equipment locations to obtain precise data.

[0195] The laser scanning data and total station data are input into the GIS system for data integration and processing to generate a three-dimensional virtual model of the mine.

[0196] The generated virtual model can accurately reflect the internal structure of the mine and provide a basis for the mapping and display of fire risk characteristic data.

[0197] Based on the collected geographic information data, a virtual geographic model of the mine is constructed, including the mine's three-dimensional structure, layout, channels, and equipment locations. The geographic information data is converted into a three-dimensional virtual model using three-dimensional modeling software (such as AutoCAD, Revit, and 3ds Max). Three-dimensional modeling software can process complex geometric data and attribute data to generate realistic virtual models.

[0198] 3D modeling software: Generates virtual geographic models by inputting geographic information data, performing geometric modeling, texture mapping and attribute setting.

[0199] Model optimization: Optimize the generated 3D model, including simplifying the number of polygons, optimizing texture mapping, etc., to improve the real-time rendering performance of the model.

[0200] The generated virtual geographic model can intuitively display the internal structure and layout of the mine, providing an accurate spatial framework for mapping fire risk characteristic data.

[0201] Map fire risk characteristic data (such as temperature, gas concentration, etc.) to the virtual geographic model to achieve visual display of fire risk. Through the programming interface (API), the real-time monitoring data is transmitted to the virtual geographic model, and the fire risk characteristic data is dynamically displayed in the virtual model using visualization technologies such as color, shape and animation.

[0202] Data mapping: associate real-time monitoring data (such as temperature and gas concentration) with corresponding locations in the virtual model, and display the data through color changes, shape changes, etc.

[0203] Real-time update: The virtual model needs to be synchronized with the real-time data to ensure that the data displayed in the model can reflect the actual situation in real time. Use data streaming technology to achieve real-time update and synchronization of data.

[0204] It realizes the dynamic display of fire risk characteristic data, so that monitoring personnel can intuitively understand the fire risk area, flame spread path and multi-dimensional data, and improve the accuracy and timeliness of monitoring.

[0205] Through virtual reality technology, virtual geographic models and fire risk characteristic data are presented in VR devices to provide an immersive monitoring experience. Using VR devices (such as Oculus Rift, HTC Vive), virtual models and real-time data are integrated to generate an immersive virtual monitoring view.

[0206] Through programming interfaces, virtual geographic models and real-time data are transmitted to VR devices. VR devices provide an immersive monitoring experience through three-dimensional display and motion tracking, allowing monitoring personnel to fully observe fire risk areas and flame spread paths.

[0207] Through augmented reality technology, virtual geographic models and fire risk characteristic data are superimposed on real scenes to provide an augmented reality monitoring experience. Using AR devices (such as Microsoft HoloLens, Magic Leap), virtual models and real-time data are superimposed on real scenes to generate augmented reality views.

[0208] Through programming interfaces, virtual geographic models and real-time data are transmitted to AR devices. AR devices overlay virtual models and data on real scenes through cameras and displays, providing an augmented reality experience that enables monitoring personnel to observe fire risk areas and flame spread paths in real environments.

[0209] The application of VR and AR technologies has improved the intuitiveness and immersion of fire risk monitoring, enabling monitoring personnel to understand and analyze fire risks more comprehensively and improve the accuracy and timeliness of emergency response.

[0210] The virtual monitoring view data is displayed in real time on the virtual fire monitoring interface. The control center initiates the emergency response procedure according to the received warning signal, and provides real-time feedback on the emergency handling situation through virtual reality and augmented reality devices, thereby implementing fire emergency drills and capacity improvement.

[0211] In the coal mine fire early warning system, real-time display of virtual monitoring view data enables monitoring personnel to promptly understand the fire risk situation inside the mine and provide a basis for rapid emergency measures. By combining virtual models with real-time data, dynamic monitoring views are generated. These views are displayed in real time through a dedicated monitoring interface, allowing monitoring personnel to intuitively observe the fire risk area and the flame spread path.

[0212] Using data integration technology, real-time data from different sensors (such as temperature, gas concentration, etc.) are integrated into the virtual model to generate a dynamic virtual monitoring view. Using real-time rendering technology, the virtual monitoring view data is displayed in real time in the monitoring interface to ensure the timeliness and accuracy of the data.

[0213] Real-time visualization of fire risk areas and flame spread paths can be achieved to help monitoring personnel quickly and accurately assess fire risk situations and improve the response speed of the early warning system.

[0214] In one embodiment, scenario: In a mine monitoring center, fire risk data is displayed in real time through a virtual fire monitoring interface.

[0215] Collect and integrate real-time monitoring data within the mine, including temperature, gas concentration, etc.

[0216] Map real-time data into a virtual geographic model to generate a dynamic virtual monitoring view.

[0217] Real-time rendering technology is used to display dynamic monitoring view data in the virtual fire monitoring interface.

[0218] Monitoring personnel can view fire risk areas and flame spread paths in real time through the monitoring interface, improving the efficiency of monitoring and early warning.

[0219] When the virtual fire monitoring interface shows abnormal fire risk characteristic data and the smart contract triggers an early warning signal, the control center needs to quickly start the emergency response program to minimize the losses caused by the fire. The emergency response program includes a series of preset emergency measures, such as evacuation, fire extinguishing, equipment protection, etc. When receiving the early warning signal, the control center quickly organizes and dispatches emergency resources according to the predetermined emergency response program.

[0220] When the smart contract triggers an early warning signal, the system automatically notifies the control center and starts the emergency response procedure. The control center dispatches the fire brigade, medical team and emergency equipment based on the fire risk assessment results to ensure the rapid and efficient execution of the emergency response.

[0221] By quickly launching emergency response procedures, timely organizing and dispatching emergency resources, we can effectively reduce the losses caused by fire and ensure the safety of mine personnel and equipment.

[0222] Through virtual reality (VR) and augmented reality (AR) devices, monitoring personnel and emergency response teams can obtain more intuitive fire risk displays and emergency response guidance, improving the efficiency and accuracy of emergency response. VR and AR devices can integrate virtual models and real-time data to provide immersive and augmented reality monitoring and emergency response experience.

[0223] VR headsets (such as Oculus Rift and HTC Vive) are used to display virtual models of mines and real-time fire risk data, providing immersive fire monitoring and emergency response guidance. AR devices (such as Microsoft HoloLens and MagicLeap) are used to overlay virtual models and fire risk data on real scenes, providing augmented reality emergency response guidance and real-time feedback.

[0224] Provide intuitive fire risk display and emergency response guidance, improve the response speed and accuracy of monitoring personnel and emergency teams, and enhance emergency response capabilities.

[0225] Regular fire emergency drills can improve the response capability and collaboration efficiency of the emergency team, ensuring that emergency response can be carried out quickly and effectively when an actual fire occurs. Virtual reality and augmented reality technologies are used to simulate fire scenes and emergency response processes, and virtual emergency drills are conducted. During the drill, the system records and analyzes the performance of the emergency team in real time, providing feedback and improvement suggestions.

[0226] Using VR and AR technology, fire scenes and emergency response processes are simulated, allowing emergency teams to conduct drills in a virtual environment. The system records the response time, processing steps, and collaboration of the emergency team in real time, analyzes them, and provides suggestions for improvement.

[0227] Improve the response speed and handling capabilities of the emergency team to ensure that emergency response can be carried out quickly and effectively when an actual fire occurs, thereby minimizing fire losses.

[0228] The present invention can greatly improve the accuracy, timeliness and effectiveness of coal mine fire warning and emergency response through real-time display of a virtual fire monitoring interface, rapid initiation of emergency response procedures, use of VR and AR equipment for emergency handling guidance and regular fire emergency drills.

[0229] Preferably, the emergency response procedure comprises the following steps:

[0230] The control center determines the emergency response level based on the received warning signal and initiates the corresponding emergency response procedures, including evacuation, fire extinguishing and equipment protection;

[0231] Dispatch emergency resources, including fire brigade, medical team and emergency equipment, to ensure efficient execution of emergency response;

[0232] Real-time feedback of emergency response status through virtual reality and augmented reality devices, updating information on the virtual fire monitoring interface;

[0233] Conduct fire emergency drills regularly, evaluate drill effectiveness, and formulate improvement measures based on the evaluation results to enhance emergency response capabilities.

[0234] In the coal mine fire early warning system, determining the emergency response level based on the received early warning signal and initiating the corresponding emergency response procedure is an important part of ensuring timely and effective response to fires. Through smart contracts and fire risk assessment models, the system can assess the severity of fire risks and automatically determine the emergency response level based on the risk score. Emergency response levels are usually divided into three levels: low, medium, and high, and each level corresponds to different emergency measures and resource scheduling plans.

[0235] Low-level response: Applicable to minor fires or situations with low risks, taking local evacuation and equipment protection measures. Intermediate response: Applicable to moderate fire risks, taking extensive evacuation, local fire extinguishing and key equipment protection measures. High-level response: Applicable to serious fire risks, taking comprehensive evacuation, comprehensive fire extinguishing and all-round equipment protection measures.

[0236] According to the severity of the fire risk, the emergency response level is quickly determined and the corresponding emergency response procedures are initiated to ensure that the fire risk can be controlled in a timely and effective manner.

[0237] According to the determined emergency response level, emergency resources, including fire brigade, medical team and emergency equipment, are quickly dispatched to ensure efficient execution of emergency response. The system automatically dispatches corresponding emergency resources according to the emergency response level to ensure that resources are quickly in place and used efficiently. The dispatch process includes notifying relevant emergency personnel, allocating emergency equipment and arranging emergency vehicles.

[0238] Fire brigade dispatch: Dispatch the appropriate number of fire brigades and equipment based on the fire risk level to ensure that the fire can be brought under control quickly.

[0239] Medical team dispatch: When a fire occurs, the medical team will be dispatched in a timely manner to prepare necessary medical equipment and medicines to ensure that the wounded can receive timely treatment.

[0240] Emergency equipment dispatch: including fire extinguishers, respirators, fire-resistant clothing and other equipment, which are allocated according to the fire risk level to ensure the efficient execution of on-site emergency response.

[0241] Rapidly dispatch emergency resources to ensure efficient execution of fire emergency response and minimize casualties and property losses caused by fire.

[0242] During the emergency response process, virtual reality (VR) and augmented reality (AR) devices are used to provide real-time feedback on the emergency response and update the information on the virtual fire monitoring interface to ensure that the control center and the emergency team can simultaneously understand the on-site situation. VR and AR devices can integrate virtual models with real-time data to display the progress and effects of emergency response in real time. Through these devices, the control center and on-site emergency personnel can obtain intuitive and detailed feedback information to improve the coordination and efficiency of emergency response.

[0243] Monitoring personnel use VR equipment to observe the virtual model of the fire scene and understand the progress of emergency response in real time. On-site emergency personnel use AR equipment to superimpose virtual information on the actual scene and obtain real-time processing guidance and feedback.

[0244] Provide real-time and intuitive emergency handling feedback to improve the coordination and efficiency of emergency response and ensure that fires can be quickly brought under control.

[0245] Regularly conducting fire emergency drills, evaluating the effectiveness of the drills, and formulating improvement measures based on the evaluation results to enhance emergency response capabilities are important measures to ensure that the emergency team always maintains a high level of emergency response capabilities. Through virtual reality and augmented reality technology, fire scenes and emergency response processes are simulated to conduct virtual emergency drills. During the drill, the system records and analyzes the performance of the emergency team in real time, providing feedback and improvement suggestions.

[0246] Using VR and AR technology, fire scenes and emergency response processes are simulated, allowing emergency teams to conduct drills in a virtual environment. The system records the response time, processing steps, and collaboration of the emergency team in real time, analyzes them, and provides suggestions for improvement.

[0247] Improve the response speed and handling capabilities of the emergency team to ensure that emergency response can be carried out quickly and effectively when an actual fire occurs, thereby minimizing fire losses.

[0248] The present invention can significantly improve the accuracy, timeliness and effectiveness of coal mine fire emergency response by determining the emergency response level, dispatching emergency resources, using VR and AR devices to provide real-time feedback on emergency handling situations, and regularly conducting fire emergency drills.

[0249] like Figure 4 As shown, a fire warning system for a coal mine comprises:

[0250] Multimodal sensor networks are deployed inside mines to collect temperature data, gas concentration data, thermal imaging image data, acoustic wave data, and vibration data in real time. Multimodal sensor networks refer to the deployment of multiple types of sensors inside mines to collect various environmental data in real time. These sensors include temperature sensors, gas sensors (such as CO, CH4, etc.), infrared cameras, acoustic wave sensors, and vibration sensors. These sensors form a stable and continuous data collection network through self-organization and self-repair functions, ensuring that when a sensor node fails, other nodes can automatically reorganize the network and continue to transmit data through redundant paths.

[0251] Through the multimodal sensor network, the environmental data in the mine can be collected comprehensively and in real time, providing a diverse data source to ensure the accuracy and reliability of the data.

[0252] Temperature sensors: deployed at key points in the mine to monitor temperature changes in real time.

[0253] Gas sensors: installed in ventilation openings and coal mining areas to monitor the concentration of harmful gases.

[0254] Infrared camera: installed in the main passage of the mine to provide thermal imaging images.

[0255] Acoustic sensor: installed near underground equipment to monitor sound changes.

[0256] Vibration sensor: installed on the mine wall to monitor vibration.

[0257] The downhole exchange device is connected to the multimodal sensor network and is used to preliminarily preprocess and format the data to generate a formatted sensor data set; the downhole exchange device is connected to the multimodal sensor network and is used to preliminarily preprocess and format the collected data. Preprocessing includes removing noise, filling missing data, etc., and formatting data is to unify the data format for subsequent processing.

[0258] Through the downhole exchange device, the quality and consistency of data can be improved, providing a reliable data basis for subsequent data processing.

[0259] Use filtering algorithms to remove high-frequency noise from sensor data. Unify data formats and generate formatted sensor data sets.

[0260] A data processing module, used to perform data cleaning, time synchronization and fusion processing on the formatted sensor data set, extract multimodal feature data, calculate the fire risk score through feature screening and fire risk assessment model, and generate fire risk feature data;

[0261] The data processing module is used to further process the formatted sensor data set, including data cleaning, time synchronization and fusion processing. Data cleaning includes noise filtering and missing data interpolation. Time synchronization ensures that the data of different sensors are aligned on the same time axis. Data fusion uses multimodal data fusion algorithms (such as deep learning models) to perform comprehensive analysis on the data and extract multimodal feature data. Fire risk scores are calculated through feature screening and fire risk assessment models to generate fire risk feature data.

[0262] Improve the accuracy and efficiency of data processing, accurately extract fire risk characteristics, reduce false alarms and missed alarms, and improve the accuracy of fire warnings.

[0263] Use filtering algorithms to remove noise, and use linear interpolation or spline interpolation methods to fill missing values. Use timestamp alignment methods to ensure that data is aligned on the same time axis. Use deep learning models to fuse time-synchronized data and extract multimodal feature data. Use random forest algorithms to filter features, and use neural network models to calculate fire risk scores.

[0264] The blockchain storage module is used to upload the fire risk characteristic data to the blockchain network for encrypted storage, and monitor the fire risk characteristic data in real time through smart contracts. When an abnormal situation is detected, the early warning signal is automatically triggered and the emergency response program is started; the blockchain storage module is used to upload the fire risk characteristic data to the blockchain network for encrypted storage. Blockchain technology ensures the transparency, immutability and traceability of data. The fire risk characteristic data is monitored in real time through smart contracts. When an abnormal situation is detected, the smart contract automatically triggers the early warning signal and starts the emergency response program.

[0265] Improve the security and credibility of data storage and ensure the reliability and real-time performance of the early warning system.

[0266] The fire risk characteristic data is encrypted using an encryption algorithm. The encrypted data is uploaded to the blockchain network and stored and verified by each node. The fire risk characteristic data is monitored in real time, and when the temperature or gas concentration exceeds the preset threshold, an early warning signal is automatically triggered.

[0267] A virtual model building module is used to build a virtual model using the geographic information data of the mine, and map the fire risk characteristic data to the virtual model, display the fire risk area, flame spread path and multi-dimensional data through virtual reality and augmented reality technology, and generate virtual monitoring view data;

[0268] The virtual model building module uses the mine's geographic information data to build a virtual model and maps the fire risk feature data into the virtual model. Through virtual reality (VR) and augmented reality (AR) technology, the fire risk area, flame spread path and multi-dimensional data are dynamically displayed to generate virtual monitoring view data.

[0269] Provides an intuitive fire monitoring interface to help monitoring personnel quickly and accurately assess fire risks and flame spread, and improve the accuracy and efficiency of emergency response.

[0270] Use 3D modeling software to build a 3D virtual model of the mine. Dynamically map fire risk feature data to the virtual model. Display fire risk areas and flame spread paths through VR headsets or AR devices.

[0271] The virtual fire monitoring interface is used to display the virtual monitoring view data in real time on the virtual fire monitoring interface. The control center starts the emergency response program according to the received early warning signal, and provides real-time feedback on the emergency handling situation through virtual reality and augmented reality equipment, so as to implement fire emergency drills and capacity improvement.

[0272] The virtual fire monitoring interface is used to display the virtual monitoring view data in real time on the virtual fire monitoring interface. The control center initiates the emergency response program according to the received warning signal, and provides real-time feedback on the emergency handling situation through virtual reality and augmented reality equipment, and implements fire emergency drills and capacity improvement.

[0273] Improve the intuitiveness and timeliness of fire monitoring, ensure the efficient execution of emergency response, and enhance emergency handling capabilities.

[0274] The virtual monitoring view data is displayed in real time on the virtual fire monitoring interface. The control center starts the emergency response program and dispatches emergency resources according to the early warning signal. The emergency handling situation is fed back in real time through VR / AR equipment, and the information on the virtual fire monitoring interface is updated. Fire emergency drills are carried out regularly, the drill effects are evaluated, and improvement measures are formulated based on the evaluation results to enhance emergency response capabilities.

[0275] Through the coordinated work of the above modules, the system of the present invention realizes early warning, precise positioning and rapid response of coal mine fires, greatly improves the accuracy of fire monitoring and the efficiency of emergency response, and reduces casualties and property losses caused by fires.

[0276] A storage medium stores a computer program, which, when executed by a processor, implements the steps of the fire warning method for coal mines.

[0277] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0278] The present application is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0279] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0280] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0281] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0282] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0283] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0284] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0285] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A fire warning method for coal mines, characterized in that: The following steps are involved: Deploy a multimodal sensor network in the mine to collect temperature data, gas concentration data, thermal imaging image data, sound wave data and vibration data in the mine in real time, and preliminarily preprocess and format the data through an underground exchange device to generate a formatted sensor data set; The formatted sensor data set is subjected to data cleaning, time synchronization and fusion processing, multimodal feature data is extracted, and a fire risk score is calculated through feature screening and a fire risk assessment model to generate fire risk feature data; The fire risk characteristic data is uploaded to the blockchain network for encrypted storage, and the fire risk characteristic data is monitored in real time through a smart contract. When an abnormal situation is detected, a warning signal is automatically triggered and an emergency response program is started. The fire risk characteristic data is encrypted and stored through the blockchain network, and is monitored in real time through a smart contract, including the following steps: Encrypting fire risk characteristic data to generate encrypted data; Upload the encrypted data to the blockchain network, where it will be stored and verified by each node; Smart contracts automatically monitor fire risk characteristic data and automatically trigger warning signals when they detect that temperature data exceeds a preset threshold or the rate of change exceeds a preset rate threshold; Smart contracts automatically monitor fire risk characteristic data, automatically trigger warning signals, and initiate emergency response procedures; The response mechanism of the smart contract in fire warning includes: Numerical thresholds, including triggering an alarm when the set temperature exceeds a specific value and triggering an alarm when the set key gas concentration exceeds a specific value; Value change rate thresholds, including triggering an early warning when the temperature is detected to rise rapidly in a short period of time and triggering an early warning when the concentration of key gases is detected to increase rapidly in a short period of time; Anomaly detection where values ​​influence each other, including triggering an alert when the concentration of key gases increases significantly while the temperature rises, and conducting a comprehensive analysis based on multiple parameters, triggering an alert when these parameters exceed preset thresholds at the same time; Trend analysis, including long-term trend detection, monitoring the long-term trend of temperature and gas concentration, triggering an alert when the trend continues to rise and reaches a preset threshold; and short-term fluctuation detection, monitoring the short-term fluctuation of data, triggering an alert when the data fluctuates abnormally in the short term; Multi-point anomaly detection, including spatial distribution detection, monitoring data from multiple sensor nodes and triggering an alert when multiple nodes detect anomalies at the same time; and regional diffusion detection, monitoring the spatial diffusion of fire risk and triggering an alert when the fire risk spreads from one node to multiple adjacent nodes; Complex event processing, including compound event detection, combining multiple data sources and multiple threshold conditions to detect compound events, triggering warnings when multiple conditions are met at the same time; and dynamic threshold adjustment, where smart contracts dynamically adjust thresholds based on actual conditions; Using the geographic information data of the mine to build a virtual model, and mapping the fire risk characteristic data to the virtual model, displaying the fire risk area, flame spread path and multi-dimensional data through virtual reality and augmented reality technology, and generating virtual monitoring view data; The virtual monitoring view data is displayed in real time on the virtual fire monitoring interface. The control center initiates the emergency response procedure according to the received warning signal, and provides real-time feedback on the emergency handling situation through virtual reality and augmented reality devices, thereby implementing fire emergency drills and capacity improvement.

2. The fire warning method for coal mine according to claim 1, characterized in that: The multimodal sensor network includes self-organization and self-repairing functions to ensure the stability and continuity of data transmission. When a sensor node fails, other nodes automatically reorganize the network and continue to transmit data through redundant paths.

3. The fire warning method for coal mine according to claim 1, characterized in that: The preliminary pretreatment comprises the following steps: Noise filtering of the formatted sensor data set includes using a filtering algorithm to remove high-frequency noise from the sensor data; Interpolation processing is performed on missing data, including using linear interpolation or spline interpolation methods to fill in missing values ​​in sensor data and generate a cleaned sensor data set.

4. The fire warning method for coal mine according to claim 1, characterized in that: The time synchronization and fusion process comprises the following steps: Time synchronization of data from different sensors, including using timestamp alignment methods to ensure that all sensor data are aligned on the same time axis; The multimodal data fusion algorithm is used to fuse the time-synchronized data and extract multimodal feature data.

5. The fire warning method for coal mine according to claim 1, characterized in that: The fire risk assessment model is generated by the following steps: Use feature screening algorithms to evaluate and screen out the features that have the greatest impact on fire risk and generate a key feature dataset; Train and validate key feature datasets through machine learning or deep learning models to generate fire risk assessment models; The real-time key feature data set is input into the fire risk assessment model to calculate the fire risk score and generate fire risk feature data.

6. The fire warning method for coal mine according to claim 1, characterized in that: The virtual model includes the following steps: Collect geographic information data of the mine, including the mine layout, passages and equipment locations, and build a virtual geographic model; Fire risk characteristic data are mapped into a virtual geographic model, and fire risk areas, flame spread paths and multi-dimensional data are dynamically displayed through virtual reality and augmented reality technologies to generate virtual monitoring view data.

7. The fire warning method for coal mine according to claim 1, characterized in that: The emergency response procedure includes the following steps: The control center determines the emergency response level based on the received warning signal and initiates the corresponding emergency response procedures, including evacuation, fire extinguishing and equipment protection; Dispatch emergency resources, including fire brigade, medical team and emergency equipment, to ensure efficient execution of emergency response; Real-time feedback of emergency response status through virtual reality and augmented reality devices, updating information on the virtual fire monitoring interface; Conduct fire emergency drills regularly, evaluate drill effectiveness, and formulate improvement measures based on the evaluation results to enhance emergency response capabilities.

8. A fire warning system for coal mines, characterized in that: include: A multimodal sensor network is deployed inside the mine to collect temperature data, gas concentration data, thermal imaging image data, sound wave data, and vibration data in real time. A downhole exchange device connected to the multimodal sensor network and used to preliminarily preprocess and format the data to generate a formatted sensor data set; A data processing module, used to perform data cleaning, time synchronization and fusion processing on the formatted sensor data set, extract multimodal feature data, and calculate the fire risk score through feature screening and a fire risk assessment model to generate fire risk feature data; The blockchain storage module is used to upload the fire risk characteristic data to the blockchain network for encrypted storage, and monitor the fire risk characteristic data in real time through smart contracts. When an abnormal situation is detected, an early warning signal is automatically triggered and an emergency response program is started. The fire risk characteristic data is encrypted and stored through the blockchain network and monitored in real time through smart contracts, including the following steps: Encrypting fire risk characteristic data to generate encrypted data; Upload the encrypted data to the blockchain network, where it will be stored and verified by each node; Smart contracts automatically monitor fire risk characteristic data and automatically trigger warning signals when they detect that temperature data exceeds a preset threshold or the rate of change exceeds a preset rate threshold; Smart contracts automatically monitor fire risk characteristic data, automatically trigger warning signals, and initiate emergency response procedures; The response mechanism of the smart contract in fire warning includes: Numerical thresholds, including triggering an alarm when the set temperature exceeds a specific value and triggering an alarm when the set key gas concentration exceeds a specific value; Value change rate thresholds, including triggering an early warning when the temperature is detected to rise rapidly in a short period of time and triggering an early warning when the concentration of key gases is detected to increase rapidly in a short period of time; Anomaly detection where values ​​influence each other, including triggering an alert when the concentration of key gases increases significantly while the temperature rises, and conducting a comprehensive analysis based on multiple parameters, triggering an alert when these parameters exceed preset thresholds at the same time; Trend analysis, including long-term trend detection, monitoring the long-term trend of temperature and gas concentration, triggering an alert when the trend continues to rise and reaches a preset threshold; and short-term fluctuation detection, monitoring the short-term fluctuation of data, triggering an alert when the data fluctuates abnormally in the short term; Multi-point anomaly detection, including spatial distribution detection, monitoring data from multiple sensor nodes and triggering an alert when multiple nodes detect anomalies at the same time; and regional diffusion detection, monitoring the spatial diffusion of fire risk and triggering an alert when the fire risk spreads from one node to multiple adjacent nodes; Complex event processing, including compound event detection, combining multiple data sources and multiple threshold conditions to detect compound events, triggering warnings when multiple conditions are met at the same time; and dynamic threshold adjustment, where smart contracts dynamically adjust thresholds based on actual conditions; A virtual model building module is used to build a virtual model using the geographic information data of the mine, and map the fire risk characteristic data to the virtual model, display the fire risk area, flame spread path and multi-dimensional data through virtual reality and augmented reality technology, and generate virtual monitoring view data; The virtual fire monitoring interface is used to display the virtual monitoring view data in real time on the virtual fire monitoring interface. The control center starts the emergency response program according to the received early warning signal, and provides real-time feedback on the emergency handling situation through virtual reality and augmented reality equipment, so as to implement fire emergency drills and capacity improvement.

9. A storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the fire warning method for coal mines described in any one of claims 1 to 7 are implemented.

Images