Mine safety intelligent early warning system based on multi-source data analysis

Abstract

The invention discloses a mine safety intelligent early warning system based on multi-source data analysis, and relates to the technical field of intelligent early warning, during the operation of the system, data is collected in real time from an environment sensor, an equipment sensor, a video monitoring system and a personnel positioning system, an environment data set, an equipment state data set and a personnel safety data set are obtained, and the data set is stored in a database; carrying out modeling after preprocessing and feature extraction, generating a mine safety intelligent early-warning model, analyzing and calculating a mine safety intelligent early-warning index Ksyj, generating multi-stage early-warning information based on an analysis and calculation result, predicting future states of equipment and environment, and generating corresponding early-warning information according to different stages of the early-warning index. The security data and the early warning information are visually displayed to an operator by using an augmented reality technology, an interface for interaction between a user and the system is provided, an analysis result and the early warning information of the system are displayed, and the user is supported to operate and manage.

Description

Technical Field

[0001] The present invention relates to the field of intelligent early warning technology, and in particular to a mine safety intelligent early warning system based on multi-source data analysis. Background Art

[0002] Mine safety is an important area to ensure the stability of mine workers, equipment and environment. In the process of mine operation, due to the complexity and variability of the operating environment, there are many safety risks, including equipment failure, environmental changes and personnel operation errors. If these risk factors are not monitored and managed in a timely and effective manner, they may lead to serious safety accidents and cause huge losses to mine operations. Therefore, a timely and accurate mine safety early warning system has become necessary so that effective measures can be taken when safety hazards first appear to reduce the possibility of accidents.

[0003] However, existing mine safety early warning systems often have certain deficiencies in practical applications. The single data source that traditional systems rely on often cannot fully reflect the actual safety status of the mine, resulting in incomplete or inaccurate early warning information. In addition, the limitations of data processing and analysis methods may also affect the effectiveness of the early warning model, making the system unable to effectively respond to complex safety risks. In addition, the system's response speed and processing capabilities in the face of emergencies may also be insufficient, and effective measures may not be taken in a timely manner. Early warning systems that lack comprehensive data analysis often cannot accurately predict the potential risks of equipment failures, environmental changes, and personnel behavior, thereby increasing the possibility of mine accidents and affecting the safety and stability of mine operations. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a mine safety intelligent early warning system based on multi-source data analysis, which solves the problems mentioned in the background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a mine safety intelligent early warning system based on multi-source data analysis, including a data acquisition module, a feature extraction and fusion module, a data analysis and modeling module, a multi-level early warning generation module, an augmented reality auxiliary module and a user visualization module;

[0006] The data acquisition module is used to collect data from environmental sensors, equipment sensors, video monitoring systems and personnel positioning systems in real time, obtain environmental data, equipment status data and personnel safety data, and establish environmental data groups, equipment status data groups and personnel safety data groups;

[0007] The feature extraction and fusion module is used to perform preprocessing and feature extraction on the environment data group, the equipment status data group and the personnel safety data group, fuse various data features into a unified format, form a comprehensive feature vector and upload it to the data analysis and modeling module;

[0008] The data analysis and modeling module is used to receive comprehensive feature vector data, use machine learning and statistical analysis methods to model the processed data, generate a mine safety intelligent early warning model, and analyze and calculate the mine safety intelligent early warning index Ksyj;

[0009] The multi-level warning generation module is used to generate multi-level warning information based on the results of analysis and calculation, predict the future status of the equipment and environment, generate corresponding warning information according to different levels of warning index, and send notifications to relevant personnel;

[0010] The augmented reality auxiliary module is used to use augmented reality technology to intuitively display safety data and warning information to operators, and display real-time safety data and warning information on AR devices, including smart glasses and helmets;

[0011] The user visualization module is used to provide an interface for users to interact with the system, support users to interact with the AR interface, display the analysis results and warning information of the system, and support users to operate and manage.

[0012] Preferably, the data acquisition module includes an environmental data acquisition unit, an equipment status data acquisition unit and a personnel safety data acquisition unit;

[0013] The environmental data acquisition unit is used to collect gas concentration, microclimate, soil gas, air quality and spectrum data in the mine environment in real time through gas sensors, microclimate sensors, soil gas sensors and spectrum analyzers, and obtain: gas concentration change rate P1, microclimate index P2, soil gas concentration P3, air quality index P4 and spectrum anomaly index P5 to form an environmental data group;

[0014] The equipment status data acquisition unit is used to monitor the operating status, vibration data, sound data, temperature data, power consumption and load data of the mining equipment through a vibration spectrum analyzer, an acoustic sensor, a thermal imager, a power sensor and a load sensor, and obtain: vibration spectrum amplitude Q1, sound abnormality index Q2, equipment thermal abnormality index Q3, power consumption change rate Q4 and load fluctuation index Q5, to form an equipment status data group;

[0015] The personnel safety data acquisition unit is used to monitor the personnel's position, safety equipment wearing status and behavior patterns in real time through GPS, biometric system, intelligent safety equipment sensor, behavior recognition system and mental health monitoring equipment, including personnel position, movement trajectory, safety equipment wearing status, biometric data, and behavior analysis data, and obtain: biometric stability SB, behavior abnormality index YC, environmental adaptability index HJ, mental health fluctuation JK and safety equipment wearing integrity PD, to form a personnel safety data group.

[0016] Preferably, the feature extraction and fusion module includes a data preprocessing unit, a feature extraction unit and a feature fusion unit;

[0017] The data preprocessing unit is used to clean, normalize and format the raw data, process missing values, outliers and noise data, scale the data to a uniform range through Min-Max normalization to eliminate the dimensional effects of different features, and convert the data into a consistent format;

[0018] The feature extraction unit is used to extract useful features from the preprocessed data and perform feature engineering: extract gas concentration change trends, microclimate fluctuation characteristics and air quality fluctuation characteristics from environmental data; extract characteristics such as vibration spectrum characteristics, energy consumption change rate and load change characteristics from equipment status data; extract characteristics such as positioning accuracy, behavior pattern characteristics and safety equipment wearing status from personnel safety data;

[0019] The feature fusion unit is used to fuse the extracted features into a unified feature vector to form input data for data analysis and modeling, and to combine features from different sources using feature-level fusion technology, and to perform feature mapping and reduction.

[0020] Preferably, the data analysis modeling module includes a model training unit and an early warning calculation unit;

[0021] The model training unit is used to model the comprehensive feature vector using a machine learning algorithm neural network (NN), generate a mine safety intelligent early warning model, and use cross-validation and hyperparameter optimization techniques, grid search and random search to optimize the performance of the model;

[0022] The early warning calculation unit is used to predict new data using the trained model, and generate the mine safety intelligent early warning index Ksyj, environmental safety factor Hjaq, equipment status factor Sbzt and personnel safety factor Ryaq after calculation;

[0023] The mine safety intelligent early warning index Ksyj is obtained by calculating the following formula:

[0024]

[0025] Wherein, Ksyj represents the mine safety intelligent early warning index, Hjaq represents the environmental safety factor, Sbzt represents the equipment status factor, Ryaq represents the personnel safety factor, μH represents the expected value of the environmental safety factor, σH represents the standard deviation of the environmental safety factor, μS represents the expected value of the equipment status factor, γS represents the standardized coefficient of the equipment status factor, μR represents the expected value of the personnel safety factor, σR represents the standard deviation of the personnel safety factor, α, β and γ represent the weight coefficients of the environmental safety factor Hjaq, the equipment status factor Sbzt and the personnel safety factor Ryaq respectively;

[0026] Among them, 0≤α≤1, 0≤β≤1, 0≤γ≤1, and α+β+γ≤1.0;

[0027] The environmental safety factor Hjaq is obtained by calculating the following formula:

[0028]

[0029] Where P1 represents the rate of change of gas concentration, P2 represents the microclimate index, P3 represents the soil gas concentration, P4 represents the air quality index, P5 represents the spectral anomaly index, μ i Indicates i The expected value of an environmental parameter, σ i Indicates i The standard deviation of the environmental parameters, w i Indicates i The weight of each environmental parameter, α represents the compression factor, which is used to adjust the sensitivity of the parameter to the safety factor, P i，th Indicates i The threshold value of an environmental parameter.

[0030] Preferably, the equipment state coefficient Sbzt is obtained by calculating through the following formula:

[0031]

[0032] In the formula, Q1 represents the vibration spectrum amplitude, Q2 represents the sound abnormality index, Q3 represents the equipment thermal abnormality index, Q4 represents the power consumption change rate, Q5 represents the load fluctuation index, and Norm(X) represents the normalized i The value of the parameter X, F i Indicates i The weights of the equipment status parameters, w1, w2, w3, w4 and w5 represent the proportional coefficients of the vibration spectrum amplitude Q1, the sound abnormality index Q2, the equipment thermal abnormality index Q3, the power consumption change rate Q4 and the load fluctuation index Q5 respectively;

[0033] In the formula, 0≤w1≤1, 0≤w2≤1, 0≤w3≤1, 0≤w4≤1, 0≤w5≤1, and w1+w2+w3+w4+w5≤1.0.

[0034] Preferably, the personnel safety factor Ryaq is obtained by calculating through the following formula:

[0035]

[0036] In the formula, SB represents biometric stability, YC represents behavioral abnormality index, HJ represents environmental adaptability index, JK represents mental health fluctuation, PD represents safety equipment wearing integrity, It is used to measure the biometric stability SB relative to its normal value, μSB represents the mean value of biometric stability, σSB represents the standard deviation of biometric stability, and YC max Indicates the maximum value of the abnormal behavior index, HJ max represents the maximum value of the environmental adaptability index, σJK represents the mean value of mental health fluctuation, PD max Indicates the maximum value of the safety equipment wearing integrity.

[0037] Preferably, the multi-level warning generation module includes a warning level determination unit and a notification management unit;

[0038] The warning level determination unit is used to set warning thresholds M and N of different levels, and compare the mine safety intelligent warning index Ksyj with the preset warning threshold M and the preset warning threshold N to determine the warning level:

[0039] The mine safety intelligent early warning index Ksyj < the preset early warning threshold M, and the first early warning level evaluation is obtained. The mine safety status is normal, the equipment and environment are operating normally, and the mine equipment and environment are inspected and maintained at a fixed period. The vibration spectrum analyzer and acoustic sensor continue to be used to monitor the vibration spectrum amplitude and sound abnormalities of the equipment to ensure stable equipment operation;

[0040] The preset warning threshold M≤mine safety intelligent warning index Ksyj≤preset warning threshold N, obtain the second warning level evaluation, the mine safety status shows abnormality, and it is necessary to take moderate warning measures. Set up safety fences in the risk area to prevent irrelevant personnel from entering and reduce potential risks. When it is detected that the biometric stability is lower than the standard value or the behavior abnormality index is high, the system automatically triggers an alarm and notifies relevant personnel, triggers the system alarm, notifies relevant personnel, immediately checks the equipment for thermal abnormalities through a thermal imager, and monitors the power consumption change rate through a power sensor, evaluates whether there is equipment abnormality, sets warning signs, and restricts entry into this area;

[0041] The mine safety intelligent early warning index Ksyj> the preset early warning threshold N, and the third early warning level evaluation is obtained. The mine safety status is abnormal, and the emergency shutdown procedure is immediately started. According to the equipment status data, load fluctuation index and vibration spectrum amplitude, the emergency shutdown procedure is started, the relevant equipment is shut down, and personnel are organized for emergency evacuation to ensure that all personnel stay away from this high-risk area. The biometric system and mental health monitoring equipment are used to investigate the health status of personnel to ensure that there are no safety hazards caused by health problems.

[0042] Preferably, the notification management unit is used to send early warning notifications to equipment operators, maintenance personnel and security managers, and the notification methods include text messages, emails and in-application notifications, coordinate corresponding response measures according to the warning level and content, including arranging inspections, initiating emergency response procedures and adjusting operation plans, record the sending of early warning notifications and the execution of response measures, and collect feedback from relevant personnel.

[0043] Preferably, the augmented reality auxiliary module includes an augmented reality display unit;

[0044] The augmented reality display unit is used to display the mine's safety data, warning information, environmental status and equipment status to operators in a real-time visual form through augmented reality devices, smart glasses or helmets, superimpose real-time data in the operator's field of view, so that they can obtain key information without leaving their work stations and respond quickly, and provide interactive functions, allowing operators to interact with displayed data through gestures or voice commands: enlarge the data view, select specific data or switch display modes.

[0045] Preferably, the user visualization module includes a user interaction interface unit;

[0046] The user interaction interface unit is used to provide a user interface, display the system's analysis results, warning information and other relevant data, support users to interact with the system through a graphical interface, including executing operations, managing system settings, viewing data reports and adjusting warning parameters, allowing users to perform various management tasks, including configuring system settings, generating data reports and modifying warning thresholds.

[0047] The present invention provides a mine safety intelligent early warning system based on multi-source data analysis, which has the following beneficial effects:

[0048] (1) When the system is running, it collects data in real time from environmental sensors, equipment sensors, video surveillance systems, and personnel positioning systems to obtain environmental data groups, equipment status data groups, and personnel safety data groups. After preprocessing and feature extraction, it builds models to generate a mine safety intelligent early warning model, analyzes and calculates the mine safety intelligent early warning index Ksyj, generates multi-level early warning information based on the results of the analysis and calculation, predicts the future status of the equipment and environment, generates corresponding early warning information according to the different levels of the early warning index, and sends notifications to relevant personnel. It uses augmented reality technology to intuitively display safety data and early warning information to operators, provides an interface for users to interact with the system, displays the system's analysis results and early warning information, and supports users in operation and management.

[0049] (2) The data acquisition module integrates environmental sensors, equipment sensors, video monitoring systems and personnel positioning systems to achieve real-time data collection of the mine operating environment and personnel behavior. The comprehensiveness and real-time nature of this data collection lays a solid foundation for subsequent data processing and analysis, ensuring the timely update and accuracy of various types of safety data. The feature extraction and fusion module converts data from different sources into a unified format to form a comprehensive feature vector by preprocessing and extracting features from the collected data.

[0050] (3) The data analysis modeling module uses these feature vectors and advanced machine learning and statistical analysis methods to build a mine safety intelligent early warning model and calculate the mine safety intelligent early warning index Ksyj. The model can perform accurate early warning analysis based on comprehensive data, and predict the status of equipment and environment through a multi-level early warning generation module, thereby generating early warning information at different levels. This not only improves the accuracy and timeliness of early warnings, but also enables mine managers to identify and respond to potential risks earlier. Compared with traditional mine safety management methods, this system has significantly improved the efficiency and effectiveness of safety management in terms of improvement.

[0051] (4) Traditional methods rely on regular inspections and manual monitoring, with slow response speed and incomplete information. This system realizes comprehensive monitoring and real-time warning of mine safety conditions through real-time data collection and intelligent analysis. The introduction of augmented reality auxiliary module and user visualization module enables operators to intuitively obtain real-time safety data and warning information on AR devices, thereby improving the accuracy of decision-making and the timeliness of response. This series of improvements not only enhances the ability of mine safety management, but also significantly reduces the probability of accidents and improves the safety and efficiency of mine operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a block diagram of a mine safety intelligent early warning system based on multi-source data analysis of the present invention; DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] Example 1

[0055] The present invention provides a mine safety intelligent early warning system based on multi-source data analysis, please refer to Figure 1 , including data acquisition module, feature extraction and fusion module, data analysis and modeling module, multi-level warning generation module, augmented reality auxiliary module and user visualization module;

[0056] The data acquisition module is used to collect data from environmental sensors, equipment sensors, video monitoring systems and personnel positioning systems in real time, obtain environmental data, equipment status data and personnel safety data, and establish environmental data groups, equipment status data groups and personnel safety data groups;

[0057] The feature extraction and fusion module is used to perform preprocessing and feature extraction on the environment data group, the equipment status data group and the personnel safety data group, fuse various data features into a unified format, form a comprehensive feature vector and upload it to the data analysis and modeling module;

[0058] The data analysis and modeling module is used to receive comprehensive feature vector data, use machine learning and statistical analysis methods to model the processed data, generate a mine safety intelligent early warning model, and analyze and calculate the mine safety intelligent early warning index Ksyj;

[0059] The multi-level warning generation module is used to generate multi-level warning information based on the results of analysis and calculation, predict the future status of the equipment and environment, generate corresponding warning information according to different levels of warning index, and send notifications to relevant personnel;

[0060] The augmented reality auxiliary module is used to use augmented reality technology to intuitively display safety data and warning information to operators, and display real-time safety data and warning information on AR devices, including smart glasses and helmets;

[0061] The user visualization module is used to provide an interface for users to interact with the system, support users to interact with the AR interface, display the analysis results and warning information of the system, and support users to operate and manage.

[0062] In this embodiment, data is collected in real time from environmental sensors, equipment sensors, video monitoring systems and personnel positioning systems to obtain environmental data groups, equipment status data groups and personnel safety data groups. After preprocessing and feature extraction, modeling is performed to generate a mine safety intelligent early warning model, analyze and calculate the mine safety intelligent early warning index Ksyj, generate corresponding early warning information according to different levels of the early warning index, and send notifications to relevant personnel. Augmented reality technology is used to intuitively display safety data and early warning information to operators, support users to interact with the AR interface, and display the system's analysis results and early warning information.

[0063] Example 2

[0064] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the data acquisition module includes an environmental data acquisition unit, an equipment status data acquisition unit and a personnel safety data acquisition unit;

[0065] The environmental data acquisition unit is used to collect gas concentration, microclimate, soil gas, air quality and spectrum data in the mine environment in real time through gas sensors, microclimate sensors, soil gas sensors and spectrum analyzers, and obtain: gas concentration change rate P1, microclimate index P2, soil gas concentration P3, air quality index P4 and spectrum anomaly index P5 to form an environmental data group;

[0066] The equipment status data acquisition unit is used to monitor the operating status, vibration data, sound data, temperature data, power consumption and load data of the mining equipment through a vibration spectrum analyzer, an acoustic sensor, a thermal imager, a power sensor and a load sensor, and obtain: vibration spectrum amplitude Q1, sound abnormality index Q2, equipment thermal abnormality index Q3, power consumption change rate Q4 and load fluctuation index Q5, to form an equipment status data group;

[0067] The personnel safety data acquisition unit is used to monitor the personnel's position, safety equipment wearing status and behavior patterns in real time through GPS, biometric system, intelligent safety equipment sensor, behavior recognition system and mental health monitoring equipment, including personnel position, movement trajectory, safety equipment wearing status, biometric data, and behavior analysis data, and obtain: biometric stability SB, behavior abnormality index YC, environmental adaptability index HJ, mental health fluctuation JK and safety equipment wearing integrity PD, to form a personnel safety data group.

[0068] The feature extraction and fusion module includes a data preprocessing unit, a feature extraction unit and a feature fusion unit;

[0069] The data preprocessing unit is used to clean, normalize and format the raw data, process missing values, outliers and noise data, scale the data to a uniform range through Min-Max normalization to eliminate the dimensional effects of different features, and convert the data into a consistent format;

[0070] The feature extraction unit is used to extract useful features from the preprocessed data and perform feature engineering: extract gas concentration change trends, microclimate fluctuation characteristics and air quality fluctuation characteristics from environmental data; extract characteristics such as vibration spectrum characteristics, energy consumption change rate and load change characteristics from equipment status data; extract characteristics such as positioning accuracy, behavior pattern characteristics and safety equipment wearing status from personnel safety data;

[0071] The feature fusion unit is used to fuse the extracted features into a unified feature vector to form input data for data analysis and modeling, and to combine features from different sources using feature-level fusion technology, and to perform feature mapping and reduction.

[0072] In this embodiment, the comprehensiveness and accuracy of mine safety management are significantly improved by comprehensively processing the environmental data, equipment status data and personnel safety data in the data acquisition module. The integration of the environmental data acquisition unit, the equipment status data acquisition unit and the personnel safety data acquisition unit ensures real-time monitoring of all aspects of mine information. The feature extraction and fusion module eliminates noise and inconsistency in the data through systematic data preprocessing, feature extraction and fusion, and converts multi-source data into a unified feature vector. This method not only improves the reliability and effectiveness of the data, but also enhances the ability to analyze the complex mine safety environment, providing a solid foundation for subsequent data analysis modeling. On the whole, this integration and fusion method improves the sensitivity and accuracy of the early warning system, making mine safety management more efficient and intelligent.

[0073] Example 3

[0074] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the data analysis and modeling module includes a model training unit and an early warning calculation unit;

[0075] The model training unit is used to model the comprehensive feature vector using a machine learning algorithm neural network (NN), generate a mine safety intelligent early warning model, and use cross-validation and hyperparameter optimization techniques, grid search and random search to optimize the performance of the model;

[0076] The early warning calculation unit is used to predict new data using the trained model, and generate the mine safety intelligent early warning index Ksyj, environmental safety factor Hjaq, equipment status factor Sbzt and personnel safety factor Ryaq after calculation;

[0077] The mine safety intelligent early warning index Ksyj is obtained by calculating the following formula:

[0078]

[0079] Wherein, Ksyj represents the mine safety intelligent early warning index, Hjaq represents the environmental safety factor, Sbzt represents the equipment status factor, Ryaq represents the personnel safety factor, μH represents the expected value of the environmental safety factor, σH represents the standard deviation of the environmental safety factor, μS represents the expected value of the equipment status factor, γS represents the standardized coefficient of the equipment status factor, μR represents the expected value of the personnel safety factor, σR represents the standard deviation of the personnel safety factor, α, β and γ represent the weight coefficients of the environmental safety factor Hjaq, the equipment status factor Sbzt and the personnel safety factor Ryaq respectively;

[0080] Among them, 0≤α≤1, 0≤β≤1, 0≤γ≤1, and α+β+γ≤1.0;

[0081] The environmental safety factor Hjaq is obtained by calculating the following formula:

[0082]

[0083] Where P1 represents the rate of change of gas concentration, P2 represents the microclimate index, P3 represents the soil gas concentration, P4 represents the air quality index, P5 represents the spectral anomaly index, and μi represents the i The expected value of an environmental parameter, σ i Indicates i The standard deviation of the environmental parameters, w i Indicates i The weight of each environmental parameter, α represents the compression factor, which is used to adjust the sensitivity of the parameter to the safety factor, P i，th Indicates i The threshold value of an environmental parameter.

[0084] Used to measure environmental parameters P i Relative to the normal value μ i This function can measure the severity of the deviation of the parameter value from the normal range. It is used to standardize the impact of environmental parameters and adjust the influence of parameter values ​​on the safety factor. This ensures that the contribution of parameter values ​​to the safety factor gradually decreases after exceeding the threshold.

[0085] The equipment state coefficient Sbzt is obtained by calculating through the following formula:

[0086]

[0087] In the formula, Q1 represents the vibration spectrum amplitude, Q2 represents the sound abnormality index, Q3 represents the equipment thermal abnormality index, Q4 represents the power consumption change rate, Q5 represents the load fluctuation index, and Norm(X) represents the normalized i The value of the parameter X, F i Indicates i The weights of the equipment status parameters, w1, w2, w3, w4 and w5 represent the proportional coefficients of the vibration spectrum amplitude Q1, the sound abnormality index Q2, the equipment thermal abnormality index Q3, the power consumption change rate Q4 and the load fluctuation index Q5 respectively;

[0088] in, X min Indicates the minimum value of parameter X, X max Indicates the maximum value of parameter X;

[0089] Where, 0≤w1≤1, 0≤w2≤1, 0≤w3≤1, 0≤w4≤1, 0≤w5≤1, and w1+w2+w3+w4+w5≤1.0

[0090] The personnel safety factor Ryaq is obtained by calculating the following formula:

[0091]

[0092] In the formula, SB represents biometric stability, YC represents behavioral abnormality index, HJ represents environmental adaptability index, JK represents mental health fluctuation, PD represents safety equipment wearing integrity, It is used to measure the biometric stability SB relative to its normal value, μSB represents the mean value of biometric stability, σSB represents the standard deviation of biometric stability, and YC max Indicates the maximum value of the abnormal behavior index, HJ max represents the maximum value of the environmental adaptability index, σJK represents the mean value of mental health fluctuation, PD max Indicates the maximum value of the safety equipment wearing integrity.

[0093] The behavior abnormality index YC, the lower the better. By normalizing its value to the range of [0,1], it reflects the degree of behavioral abnormality. The closer the index value is to 1, the less abnormal it is.

[0094] In this embodiment, the data analysis and modeling module significantly improves the prediction accuracy and response speed of the early warning system by applying advanced machine learning technology to the analysis and modeling of mine safety data. The model training unit uses a neural network (NN) algorithm, combined with cross-validation and hyperparameter optimization technology, to ensure the high performance and high reliability of the mine safety intelligent early warning model. The early warning calculation unit accurately calculates the mine safety intelligent early warning index Ksyj, environmental safety factor Hjaq, equipment status factor Sbzt and personnel safety factor Ryaq by predicting new data for the trained model. The calculation of these indicators not only reflects the current safety situation, but also provides a detailed analysis of the environment, equipment and personnel safety, which helps to identify potential risk factors and abnormal situations in a timely manner. Compared with traditional early warning methods, this data-driven method significantly improves the accuracy and real-time nature of the early warning, thereby effectively enhancing the scientific nature and responsiveness of mine safety management.

[0095] Example 4

[0096] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the multi-level warning generation module includes a warning level determination unit and a notification management unit;

[0097] The warning level determination unit is used to set warning thresholds M and N of different levels, and compare the mine safety intelligent warning index Ksyj with the preset warning threshold M and the preset warning threshold N to determine the warning level:

[0098] The mine safety intelligent early warning index Ksyj < the preset early warning threshold M, and the first early warning level evaluation is obtained. The mine safety status is normal, the equipment and environment are operating normally, and the mine equipment and environment are inspected and maintained at a fixed period. The vibration spectrum analyzer and acoustic sensor continue to be used to monitor the vibration spectrum amplitude and sound abnormalities of the equipment to ensure stable equipment operation;

[0099] The preset warning threshold M≤mine safety intelligent warning index Ksyj≤preset warning threshold N, obtain the second warning level evaluation, the mine safety status shows abnormality, and it is necessary to take moderate warning measures. Set up safety fences in the risk area to prevent irrelevant personnel from entering and reduce potential risks. When it is detected that the biometric stability is lower than the standard value or the behavior abnormality index is high, the system automatically triggers an alarm and notifies relevant personnel, triggers the system alarm, notifies relevant personnel, immediately checks the equipment for thermal abnormalities through a thermal imager, and monitors the power consumption change rate through a power sensor, evaluates whether there is equipment abnormality, sets warning signs, and restricts entry into this area;

[0100] The mine safety intelligent early warning index Ksyj> the preset early warning threshold N, and the third early warning level evaluation is obtained. The mine safety status is abnormal, and the emergency shutdown procedure is immediately started. According to the equipment status data, load fluctuation index and vibration spectrum amplitude, the emergency shutdown procedure is started, the relevant equipment is shut down, and personnel are organized for emergency evacuation to ensure that all personnel stay away from this high-risk area. The biometric system and mental health monitoring equipment are used to investigate the health status of personnel to ensure that there are no safety hazards caused by health problems.

[0101] The notification management unit is used to send early warning notifications to equipment operators, maintenance personnel and security managers, including SMS, email and in-application notifications, coordinate corresponding response measures according to the warning level and content, including arranging inspections, initiating emergency response procedures and adjusting operation plans, record the sending of early warning notifications and the execution of response measures, and collect feedback from relevant personnel.

[0102] In this embodiment, the multi-level warning generation module greatly improves the emergency response capability of the mine safety warning system through accurate warning level determination and effective notification management. The warning level determination unit divides the mine safety intelligent warning index Ksyj into different warning levels according to the set thresholds M and N, ensuring the accurate identification and classification of different safety states. At the first warning level, the system maintains normal operation and regularly inspects and maintains the equipment; at the second warning level, the system automatically triggers the alarm and takes moderate warning measures, such as setting up safety fences and signs; at the third warning level, the system quickly starts the emergency shutdown procedure to shut down the equipment and evacuate personnel. The notification management unit ensures that all relevant personnel receive the warning notification in time and take appropriate response measures according to the warning level. This module significantly improves the efficiency of mine safety management through a comprehensive warning and notification mechanism, ensures rapid response and effective handling in various safety incidents, and greatly improves the safety and stability of mine operations.

[0103] Example 5

[0104] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the augmented reality auxiliary module includes an augmented reality display unit;

[0105] The augmented reality display unit is used to display the mine's safety data, warning information, environmental status and equipment status to operators in a real-time visual form through augmented reality devices, smart glasses or helmets, superimpose real-time data in the operator's field of view, so that they can obtain key information without leaving their work stations and respond quickly, and provide interactive functions, allowing operators to interact with displayed data through gestures or voice commands: enlarge the data view, select specific data or switch display modes.

[0106] The user visualization module includes a user interaction interface unit;

[0107] The user interaction interface unit is used to provide a user interface, display the system's analysis results, warning information and other relevant data, support users to interact with the system through a graphical interface, including executing operations, managing system settings, viewing data reports and adjusting warning parameters, allowing users to perform various management tasks, including configuring system settings, generating data reports and modifying warning thresholds.

[0108] In this embodiment, the integration of the augmented reality auxiliary module and the user visualization module significantly improves the operational convenience and data interactivity of the mine safety early warning system. The augmented reality display unit superimposes safety data, early warning information and environmental status in the operator's field of view in real time through smart glasses or helmets, and can quickly obtain key information and make immediate responses without interrupting work. Operators can interact with data through gestures or voice commands, such as zooming in on the view, selecting specific data or switching display modes, which improves the efficiency of information acquisition and the flexibility of operation. At the same time, the user interaction interface unit provides an intuitive graphical interface that allows users to easily manage system settings, view data reports and adjust early warning parameters. The combination of these two modules not only optimizes the real-time display of mine safety data and the operator's interactive experience, but also improves the overall management efficiency and response speed of the system, ultimately achieving a higher level of mine safety assurance.

[0109] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A mine safety intelligent early warning system based on multi-source data analysis, characterized by: It includes data acquisition module, feature extraction and fusion module, data analysis and modeling module, multi-level warning generation module, augmented reality auxiliary module and user visualization module; The data acquisition module is used to collect data from environmental sensors, equipment sensors, video monitoring systems and personnel positioning systems in real time, obtain environmental data, equipment status data and personnel safety data, and establish environmental data groups, equipment status data groups and personnel safety data groups; The feature extraction and fusion module is used to perform preprocessing and feature extraction on the environment data group, the equipment status data group and the personnel safety data group, fuse various data features into a unified format, form a comprehensive feature vector and upload it to the data analysis and modeling module; The data analysis and modeling module is used to receive comprehensive feature vector data, use machine learning and statistical analysis methods to model the processed data, generate a mine safety intelligent early warning model, and analyze and calculate the mine safety intelligent early warning index Ksyj; The multi-level warning generation module is used to generate multi-level warning information based on the results of analysis and calculation, predict the future status of the equipment and environment, generate corresponding warning information according to different levels of warning index, and send notifications to relevant personnel; The augmented reality auxiliary module is used to use augmented reality technology to intuitively display safety data and warning information to operators, and display real-time safety data and warning information on AR devices, including smart glasses and helmets; The user visualization module is used to provide an interface for users to interact with the system, support users to interact with the AR interface, display the analysis results and warning information of the system, and support users to operate and manage.

2. The mine safety intelligent early warning system based on multi-source data analysis according to claim 1 is characterized by: The data acquisition module includes an environmental data acquisition unit, an equipment status data acquisition unit and a personnel safety data acquisition unit; The environmental data acquisition unit is used to collect gas concentration, microclimate, soil gas, air quality and spectrum data in the mine environment in real time through gas sensors, microclimate sensors, soil gas sensors and spectrum analyzers, and obtain: gas concentration change rate P1, microclimate index P2, soil gas concentration P3, air quality index P4 and spectrum anomaly index P5 to form an environmental data group; The equipment status data acquisition unit is used to monitor the operating status, vibration data, sound data, temperature data, power consumption and load data of the mining equipment through a vibration spectrum analyzer, an acoustic sensor, a thermal imager, a power sensor and a load sensor, and obtain: vibration spectrum amplitude Q1, sound abnormality index Q2, equipment thermal abnormality index Q3, power consumption change rate Q4 and load fluctuation index Q5, to form an equipment status data group; The personnel safety data acquisition unit is used to monitor the personnel's position, safety equipment wearing status and behavior patterns in real time through GPS, biometric system, intelligent safety equipment sensor, behavior recognition system and mental health monitoring equipment, including personnel position, movement trajectory, safety equipment wearing status, biometric data, and behavior analysis data, and obtain: biometric stability SB, behavior abnormality index YC, environmental adaptability index HJ, mental health fluctuation JK and safety equipment wearing integrity PD, to form a personnel safety data group.

3. The mine safety intelligent early warning system based on multi-source data analysis according to claim 1 is characterized by: The feature extraction and fusion module includes a data preprocessing unit, a feature extraction unit and a feature fusion unit; The data preprocessing unit is used to clean, normalize and format the raw data, process missing values, outliers and noise data, scale the data to a uniform range through Min-Max normalization to eliminate the dimensional effects of different features, and convert the data into a consistent format; The feature extraction unit is used to extract useful features from the preprocessed data and perform feature engineering: extracting gas concentration variation trends, microclimate fluctuation characteristics and air quality fluctuation characteristics from the environmental data; Extract characteristics such as vibration spectrum, energy consumption change rate and load change characteristics from equipment status data; Extracting data such as location accuracy, behavior pattern characteristics, and safety equipment wearing status from personnel safety data; The feature fusion unit is used to fuse the extracted features into a unified feature vector to form input data for data analysis and modeling, and to combine features from different sources using feature-level fusion technology, and to perform feature mapping and reduction.

4. The mine safety intelligent early warning system based on multi-source data analysis according to claim 2 is characterized by: The data analysis and modeling module includes a model training unit and an early warning calculation unit; The model training unit is used to model the comprehensive feature vector using a machine learning algorithm neural network (NN), generate a mine safety intelligent early warning model, and use cross-validation and hyperparameter optimization techniques, grid search and random search to optimize the performance of the model; The early warning calculation unit is used to predict new data using the trained model, and generate the mine safety intelligent early warning index Ksyj, environmental safety factor Hjaq, equipment status factor Sbzt and personnel safety factor Ryaq after calculation; The mine safety intelligent early warning index Ksyj is obtained by calculating the following formula: Wherein, Ksyj represents the mine safety intelligent early warning index, Hjaq represents the environmental safety factor, Sbzt represents the equipment status factor, Ryaq represents the personnel safety factor, μH represents the expected value of the environmental safety factor, σH represents the standard deviation of the environmental safety factor, μS represents the expected value of the equipment status factor, γS represents the standardized coefficient of the equipment status factor, μR represents the expected value of the personnel safety factor, σR represents the standard deviation of the personnel safety factor, α, β and γ represent the weight coefficients of the environmental safety factor Hjaq, the equipment status factor Sbzt and the personnel safety factor Ryaq respectively; Among them, 0≤α≤1, 0≤β≤1, 0≤γ≤1, and α+β+γ≤1.0; The environmental safety factor Hjaq is obtained by calculating the following formula: Where P1 represents the rate of change of gas concentration, P2 represents the microclimate index, P3 represents the soil gas concentration, P4 represents the air quality index, P5 represents the spectral anomaly index, and μi represents the i The expected value of an environmental parameter, σ i Indicates i The standard deviation of the environmental parameters, wi represents the i The weight of the environmental parameters, α represents the compression factor, which is used to adjust the sensitivity of the parameters to the safety factor, and Pi,th represents the i The threshold value of an environmental parameter.

5. The mine safety intelligent early warning system based on multi-source data analysis according to claim 2 is characterized by: The equipment state coefficient Sbzt is obtained by calculating through the following formula: In the formula, Q1 represents the vibration spectrum amplitude, Q2 represents the sound abnormality index, Q3 represents the equipment thermal abnormality index, Q4 represents the power consumption change rate, Q5 represents the load fluctuation index, and Norm(X) represents the normalized i The value of the parameter X, F i Indicates i The weights of the equipment status parameters, w1, w2, w3, w4 and w5 represent the proportional coefficients of the vibration spectrum amplitude Q1, the sound abnormality index Q2, the equipment thermal abnormality index Q3, the power consumption change rate Q4 and the load fluctuation index Q5 respectively; In the formula, 0≤w1≤1, 0≤w2≤1, 0≤w3≤1, 0≤w4≤1, 0≤w5≤1, and w1+w2+w3+w4+w5≤1.

0.

6. The mine safety intelligent early warning system based on multi-source data analysis according to claim 2 is characterized by: The personnel safety factor Ryaq is obtained by calculating the following formula: In the formula, SB represents biometric stability, YC represents behavioral abnormality index, HJ represents environmental adaptability index, JK represents mental health fluctuation, PD represents safety equipment wearing integrity, It is used to measure the biometric stability SB relative to its normal value, μSB represents the mean value of biometric stability, σSB represents the standard deviation of biometric stability, and YC max Indicates the maximum value of the abnormal behavior index, HJ max represents the maximum value of the environmental adaptability index, σJK represents the mean value of mental health fluctuation, PD max Indicates the maximum value of the safety equipment wearing integrity.

7. The mine safety intelligent early warning system based on multi-source data analysis according to claim 1 is characterized by: The multi-level warning generation module includes a warning level determination unit and a notification management unit; The warning level determination unit is used to set warning thresholds M and N of different levels, and compare the mine safety intelligent warning index Ksyj with the preset warning threshold M and the preset warning threshold N to determine the warning level: The mine safety intelligent early warning index Ksyj < the preset early warning threshold M, and the first early warning level evaluation is obtained. The mine safety status is normal, the equipment and environment are operating normally, and the mine equipment and environment are inspected and maintained at a fixed period. The vibration spectrum analyzer and acoustic sensor continue to be used to monitor the vibration spectrum amplitude and sound abnormalities of the equipment to ensure stable equipment operation; The preset warning threshold M≤mine safety intelligent warning index Ksyj≤preset warning threshold N, obtain the second warning level evaluation, the mine safety status shows abnormality, and it is necessary to take moderate warning measures. Set up safety fences in the risk area to prevent irrelevant personnel from entering and reduce potential risks. When it is detected that the biometric stability is lower than the standard value or the behavior abnormality index is high, the system automatically triggers an alarm and notifies relevant personnel, triggers the system alarm, notifies relevant personnel, immediately checks the equipment for thermal abnormalities through a thermal imager, and monitors the power consumption change rate through a power sensor, evaluates whether there is equipment abnormality, sets warning signs, and restricts entry into this area; The mine safety intelligent early warning index Ksyj> the preset early warning threshold N, and the third early warning level evaluation is obtained. The mine safety status is abnormal, and the emergency shutdown procedure is immediately started. According to the equipment status data, load fluctuation index and vibration spectrum amplitude, the emergency shutdown procedure is started, the relevant equipment is shut down, and personnel are organized for emergency evacuation to ensure that all personnel stay away from this high-risk area. The biometric system and mental health monitoring equipment are used to investigate the health status of personnel to ensure that there are no safety hazards caused by health problems.

8. The mine safety intelligent early warning system based on multi-source data analysis according to claim 7 is characterized by: The notification management unit is used to send early warning notifications to equipment operators, maintenance personnel and security managers, including SMS, email and in-application notifications, coordinate corresponding response measures according to the warning level and content, including arranging inspections, initiating emergency response procedures and adjusting operation plans, record the sending of early warning notifications and the execution of response measures, and collect feedback from relevant personnel.

9. The mine safety intelligent early warning system based on multi-source data analysis according to claim 1 is characterized by: The augmented reality auxiliary module includes an augmented reality display unit; The augmented reality display unit is used to display the mine's safety data, warning information, environmental status and equipment status to operators in a real-time visual form through augmented reality devices, smart glasses or helmets, superimpose real-time data in the operator's field of view, so that they can obtain key information without leaving their work stations and respond quickly, and provide interactive functions, allowing operators to interact with displayed data through gestures or voice commands: enlarge the data view, select specific data or switch display modes.

10. The mine safety intelligent early warning system based on multi-source data analysis according to claim 1 is characterized in that: The user visualization module includes a user interaction interface unit; The user interaction interface unit is used to provide a user interface, display the system's analysis results, warning information and other relevant data, support users to interact with the system through a graphical interface, including executing operations, managing system settings, viewing data reports and adjusting warning parameters, allowing users to perform various management tasks, including configuring system settings, generating data reports and modifying warning thresholds.

Images