Mining safety informatization management system and method based on big data

Abstract

The invention relates to the technical field of data processing, in particular to a mining safety informatization management system and method based on big data. Comprising the following steps: collecting original data of a mining area, preprocessing the original data to obtain preprocessed data, and performing enhancement processing by using a convolution enhancement algorithm based on a dynamic weighted graph to obtain enhanced data; analyzing through a dynamic nonlinear coupling weighted collaborative risk assessment algorithm to obtain an analysis result, obtaining a quantified potential safety risk based on the analysis result, and performing mining area potential safety hazard prediction analysis assessment in combination with mining historical data to obtain a prediction result; and carrying out real-time judgment on the quantized potential safety risk and the prediction result to obtain a judgment result, automatically triggering an alarm mechanism, and sending out alarm information. The technical problems of inaccurate data processing and difficulty in capturing a complex relationship and a nonlinear synergistic effect between data in a mining area safety monitoring management process are solved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a mining safety information management system and method based on big data. Background Art

[0002] Mine safety monitoring is one of the core tasks to ensure production safety and employee life safety in mines. In recent years, with the expansion of mine production scale and the continuous advancement of technical means, the design and application of safety monitoring systems have gradually developed towards intelligence and data. Traditional mine safety monitoring systems mainly rely on a single type of sensor equipment, such as gas sensors, temperature and humidity sensors, seismic sensors, etc., which are used to monitor air quality, equipment operation status and environmental changes in mines. These single-type sensor devices have high data acquisition accuracy, but due to the lack of comprehensive analysis and data integration capabilities, it is difficult to fully and accurately identify and assess safety risks.

[0003] With the advancement of technology, various types of sensors are integrated into the mine safety monitoring system. The diversification and distributed deployment of sensors have led to a substantial increase in the amount of data. Data from different sensors have different formats and resolutions, and there are also differences in acquisition frequency and measurement range, which makes it difficult to effectively fuse and process multi-source data. Traditional data processing methods usually adopt independent data processing modes, ignoring the synergy between different data sources, resulting in potential information loss.

[0004] However, the existing mining safety information management methods have the following technical problems: inaccurate data processing during mine safety monitoring and management, difficulty in capturing the complex relationships and nonlinear synergistic effects between data, and inaccurate assessment of mine safety status. Summary of the invention

[0005] The present invention provides a mining safety information management system and method based on big data to solve the technical problems of inaccurate data processing, difficulty in capturing complex relationships and nonlinear synergistic effects between data, and inaccurate assessment of the safety status of mining areas during mine safety monitoring and management.

[0006] The mining safety information management system and method based on big data of the present invention specifically include the following technical solutions: The mining safety information management method based on big data includes the following steps: S1. Collect the raw data of the mining area in real time, pre-process the raw data to obtain pre-processed data, and enhance the pre-processed data using a dynamic weighted graph convolution enhancement algorithm to obtain enhanced data; S2. Analyze the enhanced data using a dynamic nonlinear coupling weighted collaborative risk assessment algorithm to obtain analysis results, obtain quantified potential safety risks based on the analysis results, and simultaneously perform prediction analysis and assessment of potential safety hazards in the mining area in combination with historical mining data to obtain prediction results; S3. Make real-time judgments on the quantified potential safety risks and prediction results, obtain judgment results, automatically trigger the alarm mechanism based on the judgment results, and issue alarm information; based on the alarm information, adjust the mining operation plan and resource allocation, and provide suggestions for emergency response measures.

[0007] Preferably, the S1 specifically includes: In the implementation process of the dynamic weighted graph convolution enhancement algorithm, features are extracted from the preprocessed data to obtain feature data; based on the feature data, the relationship between data acquisition devices is expressed in the form of a graph. When constructing the graph, the strength of the relationship between each pair of data acquisition devices is evaluated, and a dynamic weighting coefficient is assigned to each edge of the graph.

[0008] Preferably, the S1 specifically includes: In the implementation process of the dynamic weighted graph convolution enhancement algorithm, the dynamic weighted graph convolution operation is introduced to fuse the features of each node with the information of its neighboring nodes, and each layer of graph convolution operation is weighted using the weighted dynamic weight coefficient to obtain the feature representation of the node updated by the graph convolution operation. The specific implementation formula is:

[0009] in, Is a node In the Feature representation of the layer; is the activation function; Is a node The set of all neighbor nodes of is the dynamic weighting coefficient; Is a node In the Feature representation of the layer; is the bias term.

[0010] Preferably, the S1 specifically includes: In the implementation process of the dynamic weighted graph convolution enhancement algorithm, after the graph convolution operation, the feature data is enhanced through a multi-layer perceptron to obtain the enhanced feature data.

[0011] Preferably, the S2 specifically includes: In the process of implementing the dynamic nonlinear coupling weighted collaborative risk assessment algorithm, the enhanced data is normalized to obtain normalized feature data; the normalized feature data is subjected to nonlinear synergistic effect extraction, and the nonlinear synergistic effect value is obtained by introducing a weighted synergistic effect model. The specific implementation formula is:

[0012] in, It is The feature data at time The nonlinear synergistic effect value of is the number of different nonlinear synergistic effects; It is The weight coefficients of the nonlinear transformation; is the index adjustment coefficient; is the logarithmic adjustment coefficient; is the power adjustment coefficient; Indicates Normalized feature data; is the parameter in the exponential transformation; is a small constant.

[0013] Preferably, the S2 specifically includes: In the implementation process of the dynamic nonlinear coupling weighted collaborative risk assessment algorithm, a dynamic coupling factor model is introduced to simulate the time-varying coupling relationship between characteristic data and describe the coupling risk. The specific implementation formula is:

[0014] in, It is The feature data at time The coupling risk; is the total number of feature data; It is The characteristic data and The feature data at time The coupling weight of It is The feature data at time The risk function of It is The feature data at time The normalization coefficient of is the time decay factor, which indicates the influence of time change on coupling risk.

[0015] Preferably, the S2 specifically includes: In the process of implementing the dynamic nonlinear coupling weighted collaborative risk assessment algorithm, the coupling risks of all characteristic data are processed to generate a global safety risk assessment and obtain the analysis results. Based on the analysis results, the weighted average method is used in combination with the coupling risks of the characteristic data to obtain the quantified potential safety risks.

[0016] Preferably, the S3 specifically includes: The quantified potential safety risks and prediction results are compared with the risk threshold to determine the current safety status of the mining area and obtain the judgment result. The specific judgment formula is as follows:

[0017]

[0018] in, and are the risk thresholds for low risk and high risk; It is based on the determination result of quantified potential safety risks; It is the judgment result based on the prediction result; It is the global risk quantification value, that is, the quantified potential security risk; Represents the prediction result.

[0019] The mining safety information management system based on big data includes the following parts: Data acquisition module, data processing and storage module, data analysis and prediction module, real-time alarm module, decision support and management module; The data acquisition module collects raw data, preprocesses the raw data, obtains preprocessed data, and sends the preprocessed data to the data processing and storage module; The data processing and storage module performs enhancement processing on the pre-processed data to obtain enhanced data, and sends the enhanced data to the data analysis and prediction module; The data analysis and prediction module conducts in-depth analysis on the enhanced data through a dynamic nonlinear coupling weighted collaborative risk assessment algorithm to obtain analysis results; based on the analysis results, quantified potential safety risks are obtained; at the same time, the enhanced data is combined with historical mining data to conduct prediction analysis and evaluation of potential safety hazards in the mining area, obtain prediction results, and send the quantified potential safety risks and prediction results to the real-time alarm module; The real-time alarm module makes real-time judgments on the quantified potential safety risks and prediction results, obtains the judgment results, automatically triggers the alarm mechanism based on the judgment results, issues alarm information, and sends the alarm information to the decision support and management module; The decision support and management module, after receiving the alarm information from the real-time alarm module, adjusts the mining operation plan and resource allocation, and provides suggestions for emergency response measures.

[0020] The beneficial effects of the technical solution of the present invention are: 1. Through processing based on the dynamic weighted graph convolution enhancement algorithm, the data processing capability of the mine safety monitoring system can be effectively improved, overcoming problems such as data noise, temporal and spatial unevenness, and insufficient sensor correlation; this algorithm can provide more accurate real-time data for mine safety management and provide a more solid foundation for the prediction and response of potential risks.

[0021] 2. Through the dynamic nonlinear coupling weighted collaborative risk assessment algorithm, the enhanced data is further analyzed to quantify the potential safety risks of the mining area. The dynamic nonlinear coupling weighted collaborative risk assessment algorithm combines the nonlinear synergistic effect and dynamic coupling relationship between the feature data. By normalizing the enhanced data, the normalized feature data is obtained, and the nonlinear synergistic effect of the normalized feature data is extracted, and the dynamic coupling factor model is introduced to deeply explore the potential risks in the feature data. Through this process, the safety risks of the mining area can be quantified more accurately, and the potential safety hazards can be predicted and evaluated in combination with the historical mining data, ultimately providing decision support for the safety management of the mining area. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a structural diagram of the mining safety information management system based on big data described in the present invention; Figure 2 This is a flow chart of the mining safety information management method based on big data described in the present invention. DETAILED DESCRIPTION

[0023] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all 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.

[0024] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0025] The specific scheme of the mining safety information management system and method based on big data provided by the present invention is described in detail below with reference to the accompanying drawings.

[0026] Refer to the attached Figure 1 , which shows a structural diagram of a mining safety information management system based on big data provided by an embodiment of the present invention, the system includes the following parts: Data acquisition module, data processing and storage module, data analysis and prediction module, real-time alarm module, decision support and management module; The data acquisition module is responsible for collecting the raw data of the mining area in real time, preprocessing the raw data to obtain the preprocessed data, and sending the preprocessed data to the data processing and storage module for further processing and storage; The original data includes mining area environmental data (such as temperature, humidity, gas concentration, etc.), equipment status data (such as operating status, vibration, pressure, etc.), and personnel location data; The data processing and storage module performs enhancement processing on the pre-processed data to obtain enhanced data, sends the enhanced data to the data analysis and prediction module, and stores it in the existing database; The data analysis and prediction module conducts in-depth analysis on the enhanced data through a dynamic nonlinear coupling weighted collaborative risk assessment algorithm to obtain analysis results; based on the analysis results, quantified potential safety risks (such as equipment failure, gas leakage, etc.) are obtained; at the same time, the enhanced data is combined with historical mining data (such as equipment failure records, accident history, etc.) to conduct prediction analysis and evaluation of potential safety hazards in the mining area, obtain prediction results, and send the quantified potential safety risks and prediction results to the real-time alarm module to trigger the alarm mechanism; The real-time alarm module makes real-time judgments on the quantified potential safety risks and prediction results, obtains the judgment results, automatically triggers the alarm mechanism based on the judgment results, issues alarm information, and sends the alarm information to the decision support and management module to help the management of the mine area make quick responses and decisions. At the same time, the alarm information also helps to update the safety status of the mine area in real time to ensure timely response to potential dangers; After receiving the alarm information from the real-time alarm module, the decision support and management module adjusts the mining operation plan and resource allocation, and provides suggestions for emergency response measures to realize the information management of mining safety.

[0027] Refer to the attached Figure 2 , which shows a flow chart of a mining safety information management method based on big data provided by an embodiment of the present invention, the method comprising the following steps: S1. Collect the raw data of the mining area in real time, pre-process the raw data to obtain pre-processed data, and enhance the pre-processed data using a dynamic weighted graph convolution enhancement algorithm to obtain enhanced data; The raw data of the mining area is collected in real time through data acquisition equipment such as sensors, including mining area environmental data (such as temperature, humidity, gas concentration, etc.), equipment status data (such as operating status, vibration, pressure, etc.), and personnel positioning data, etc. The raw data is pre-processed by data cleaning, format unification, standardization and normalization to obtain pre-processed data. The technical means adopted in the above pre-processing process are existing technologies and will not be elaborated here.

[0028] The preprocessed data is enhanced using a dynamic weighted graph convolution enhancement algorithm to obtain enhanced data. The dynamic weighted graph convolution enhancement algorithm aims to combine the spatiotemporal relationship between mining area sensors, historical feature data, and real-time data streams to enhance the quality of preprocessed data, thereby improving the accuracy of mining area safety monitoring and alarm. The dynamic weighted graph convolution enhancement algorithm models the relationship between mining area data collection equipment, and introduces dynamic weighted graph convolution operations on this basis to gradually enhance data features and ultimately improve the quality of mining area safety monitoring data. The specific implementation process is as follows: The preprocessed data is recorded as , where each Indicates The pre-processed data corresponding to each data acquisition device, is the total number of data collection devices. Then, the preprocessed data is subjected to feature extraction using existing feature engineering techniques to obtain feature data. Existing feature engineering technologies include: statistical feature extraction, frequency domain analysis, Fourier transform, sliding window technology, etc. For example, the vibration data of the data acquisition equipment can be extracted through Fourier transform to extract frequency domain features, and the gas concentration data can be calculated through sliding window technology to calculate the rate of change, peak value and other features.

[0029] Since the characteristic data corresponding to each data acquisition device has different distribution in space and time, the relationship between the data acquisition devices is expressed in the form of a graph. Each data acquisition device is regarded as a node in the graph. , while the relationships between nodes (such as physical distance, data correlation, etc.) are connected through edges. To represent, that is, node and nodes The edge between.

[0030] The establishment of the graph structure depends on the physical connection between data acquisition devices and the correlation between feature data. When constructing the graph, the strength of the relationship between each pair of data acquisition devices is evaluated and a dynamic weight coefficient is assigned to each edge. , indicating that at a certain moment ,node With Node The weighted relationship between them. The calculation formula is as follows:

[0031] in, is the dynamic weighting coefficient, which means Moment A data acquisition device and The weighted relationship or correlation strength between the data acquisition devices, the dynamic weighting coefficient reflects the strength of the relationship between the two in data transmission or information dissemination; is a first weighting coefficient, used to control the effect of the physical distance between the data acquisition devices on the dynamic weighting coefficient; It is A data acquisition device and The physical distance between the data collection devices; is a second weighting coefficient, used to control the influence of the relevance of the characteristic data corresponding to the data acquisition device (e.g., data trend similarity) on the dynamic weighting coefficient; is Moment Feature data extracted from pre-processed data corresponding to each data acquisition device; is Moment Feature data extracted from pre-processed data corresponding to each data acquisition device; is the third weighting factor, used to control the impact of the dynamic weighting factor based on mining history data; is Moment A data acquisition device and The data collection equipment is based on the time series correlation of historical mining data to reflect the impact of equipment failure records, accident history, etc. on the current sensor association. It is calculated through the existing time series analysis method to describe the dependence between the two over a long period of time and improve the modeling ability of long-term trends. It is Mining history data corresponding to each data collection device; It is Mining historical data corresponding to each data collection device.

[0032] Furthermore, the dynamic weighted graph convolution operation is introduced. The dynamic weighted graph convolution operation realizes the propagation and update of information by fusing the features of each node with the information of its neighboring nodes, and uses the dynamic weighting coefficient calculated by weighted calculation. , weighted processing is performed on each layer of graph convolution operations. Assume that each node is at time The characteristics are , is the feature representation of the node updated by the graph convolution operation, and the update formula is:

[0033] in, Is a node In the The feature representation of the layer represents the nodes updated by the graph convolution operation. features, which combine the features of the nodes and all its neighbor nodes information; is the activation function; Is a node The set of all neighbor nodes of the node. The neighbor node is the node in the graph structure. directly connected nodes; Is a node In the Feature representation of the layer; is the bias term.

[0034] Through graph convolution operations, the features of neighboring nodes are gradually integrated to construct a high-level representation of the node. The specific number of layers is defined according to the specific scenario and will not be described here. As the number of graph convolution layers increases, the feature representation of the node becomes richer and richer, and the spatiotemporal dependencies between nodes can be effectively captured.

[0035] After the graph convolution operation, the node features obtained have already integrated the information of its neighboring nodes. However, in order to further improve the expressiveness of node features and the accuracy of feature data, the feature data is enhanced through a multi-layer perceptron (MLP). Specifically, for each node , applying the following enhancement formula:

[0036] in, Is a node The enhanced feature representation is obtained by and its neighbor nodes The final result is obtained after weighted processing, convolution operation and nonlinear activation of the features; It is a multi-layer perceptron, an existing feed-forward neural network composed of fully connected layer neurons, which is used to perform further nonlinear mapping of node features after graph convolution operation.

[0037] Through enhanced processing, the features of the nodes will become richer, further improving the quality of the pre-processed data and adapting to more complex security risk monitoring tasks.

[0038] After the data enhancement in the above steps, we get The feature representation is the data after enhancement processing.

[0039] S2. Analyze the enhanced data using a dynamic nonlinear coupling weighted collaborative risk assessment algorithm to obtain analysis results, obtain quantified potential safety risks based on the analysis results, and simultaneously perform prediction analysis and assessment of potential safety hazards in the mining area in combination with historical mining data to obtain prediction results; The enhanced data is deeply analyzed through the dynamic nonlinear coupling weighted collaborative risk assessment algorithm to obtain the analysis results; the dynamic nonlinear coupling weighted collaborative risk assessment algorithm deeply explores the potential interactive relationship between the features by considering the complex nonlinear synergy, dynamic coupling relationship and time weighting mechanism among the characteristic data of the mining area, and finally quantifies the overall safety risk of the mining safety information management system. The specific implementation process is as follows: First, the enhanced data is normalized to eliminate the dimensional differences between different feature data and ensure that data from different data sources can be effectively integrated. The purpose of normalization is to map the enhanced data to the same scale to avoid the dimensional differences between different feature data affecting the subsequent calculation process and obtain the normalized feature data. , Indicates Normalized feature data; The normalized feature data then enters the nonlinear synergistic effect extraction stage. By introducing the weighted synergistic effect model, the nonlinear synergistic effect value is obtained, which is in the following form:

[0040] in, It is The feature data at time The nonlinear synergistic effect value of represents the impact of this feature on the safety risk of the mining safety information management system, reflecting the contribution of this feature to the safety risk in a time-varying environment; is the number of different nonlinear synergies, i.e., the number of different transformation types used in the weighted synergy model to capture the characteristic relationships, which is determined according to the specific scenario; It is The weight coefficients of the nonlinear transformation; is the exponential adjustment coefficient, which is used to control the influence of the exponential decay term; is the logarithmic adjustment coefficient, which is used to control the impact of logarithmic transformation and capture the risk effect of feature changes on a logarithmic scale; is the power adjustment coefficient, which is used to control the influence of quadratic changes and reflects the impact of the quadratic nonlinear synergistic effect of the normalized characteristic data on the risk of the mining safety information management system; It is a parameter in exponential transformation, which is used to control the rate of exponential change; is a small constant used to smooth out errors in the logarithmic function.

[0041] Furthermore, in order to capture the complex relationship between features that changes over time, a dynamic coupling factor model is introduced to simulate the time-varying coupling relationship between feature data. The dynamic coupling factor model considers the time-varying coupling relationship between feature data in different time periods, especially the time-varying coupling effect between feature data, and is used to describe the coupling risk, which is expressed by the following formula:

[0042] in, It is The feature data at time The coupling risk represents the Top, The impact of the interaction between individual characteristic data and other characteristic data on the overall safety risk of the mining safety information management system; is the total number of feature data; It is The characteristic data and The feature data at time The coupling weight of The characteristic data and The strength of the relationship between the feature data; It is The feature data at time The risk function is the contribution value to the safety risk of the mining safety information management system, which indicates the contribution of the characteristic data to the safety risk of the mining safety information management system. The risk function can quantify the safety impact of the characteristic data. It can be the direct value of the characteristic data or it can be determined according to the expert experience method through a linear, nonlinear or weighted transformation function. It is The feature data at time The normalization coefficient is used to adjust the influence weight of characteristic data on other characteristic data during the coupling process; is the time decay factor, which indicates the degree of influence of time changes on coupling risk; Further comprehensive processing of the coupling risk of all feature data generates a global security risk assessment to obtain the analysis results. In order to achieve this goal, the weighted average method is used to combine the coupling risks of all feature data to form a global risk quantification value. This process not only considers the importance of each feature data, but also considers the impact of historical feature data on the current assessment results by introducing the time decay effect. The specific mathematical formula is:

[0043] in, It is the global risk quantification value, that is, the quantified potential safety risk, which reflects the safety status of the entire mining area; It is The feature data at time The weighting coefficient is used to adjust the contribution of the feature in the global risk assessment and is determined based on expert experience; is the attenuation factor, which is used to control the influence of historical characteristic data on the current risk assessment; It is Feature data in the past The coupling risk; is the time-integrated variable; is the upper limit of the time window, indicating the length of the past period considered.

[0044] Through the above process, the coupling risk information of each feature data is integrated to obtain the quantified potential safety risk, providing a basis for subsequent safety warnings.

[0045] At the same time, combined with historical mining data, according to expert experience and combined with the usage scenario, appropriate existing machine learning models (such as decision trees, random forests, support vector machines, etc.) are selected to conduct prediction analysis and evaluation of potential safety hazards in mining areas to obtain prediction results. ; S3. Make real-time judgments on the quantified potential safety risks and prediction results, obtain judgment results, automatically trigger the alarm mechanism based on the judgment results, and issue alarm information; based on the alarm information, adjust the mining operation plan and resource allocation, and provide suggestions for emergency response measures.

[0046] Based on the quantified potential safety risks and prediction results, they are compared with the risk threshold preset according to the expert experience method to achieve real-time judgment, determine the current safety status of the mining area, obtain the judgment result, and automatically trigger the alarm mechanism based on the judgment result to issue an alarm message. The risk threshold divides the quantified potential safety risks and prediction results into three levels: low risk, medium risk, and high risk. Different levels of alarms will be triggered according to different safety statuses to promptly alert potential safety hazards. The specific judgment formula is as follows:

[0047]

[0048] in, and are the risk thresholds for low risk and high risk, which define the boundaries of different security states; It is based on the determination result of quantified potential safety risks; It is the judgment result based on the prediction results. According to the quantified potential safety risks and the size of the prediction results, the current safety status of the mining area is judged, and the alarm mechanism is triggered to issue an alarm message. For example, for the quantified potential safety risks, when the safety status is low risk, the mine area is considered to be in a low-risk state, no emergency alarm is issued, and it is only recommended to continue the existing operation plan; when the safety status is medium risk, the mine area is considered to be in a medium risk state, a "warning" message will be issued, and it is recommended to strengthen monitoring or increase safety inspections; when the safety status is high risk, the mine area is judged to be high risk, and a "danger" alarm will be issued, requiring immediate shutdown or emergency response measures; for the prediction results, when the safety status is low risk, the mine area is considered to have a low risk at the prediction time, and no emergency alarm will be triggered, but it is recommended to maintain regular operation safety inspections; when the safety status is medium risk, the mine area is considered to have a neutral risk at the prediction time, and a warning will be issued to remind the mine area that safety hazards may occur at some point in the future, and it is recommended to strengthen safety inspections and preventive measures; when the safety status is high risk, the mine area is considered to have a high risk at the prediction time, and a high-risk alarm will be issued to remind the mine area that it is about to enter a dangerous state, and it may be necessary to immediately adjust the operation plan, stop certain high-risk operations, or initiate an emergency response.

[0049] Furthermore, based on the alarm information, the mining operation plan and resource allocation are adjusted, and suggestions for emergency response measures are provided, such as: According to the quantified potential safety risk determination result S1: Low risk: Continue with current mining plan and maintain existing resource allocation.

[0050] Medium risk: Adjust the operation plan appropriately, such as reducing the operation depth or increasing safety monitoring.

[0051] High risk: Suspend operations, organize emergency drills, evacuate personnel and conduct emergency risk elimination.

[0052] According to the judgment result S2 of the prediction result: Low risk: Maintain current operating schedule and conduct routine safety inspections.

[0053] Medium risk: Strengthen monitoring, conduct partial equipment maintenance, and adjust operating procedures.

[0054] High risk: Immediately stop operations, activate the emergency plan, and dispatch emergency personnel for rescue.

[0055] In summary, the mining safety information management system and method based on big data has been completed.

[0056] The order of the embodiments of the invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0057] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0058] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A mining safety information management method based on big data, characterized in that: The following steps are involved: S1. Collect the raw data of the mining area in real time, pre-process the raw data to obtain pre-processed data, and enhance the pre-processed data using a dynamic weighted graph convolution enhancement algorithm to obtain enhanced data; S2. Analyze the enhanced data using a dynamic nonlinear coupling weighted collaborative risk assessment algorithm to obtain analysis results, obtain quantified potential safety risks based on the analysis results, and simultaneously perform prediction analysis and assessment of potential safety hazards in the mining area in combination with historical mining data to obtain prediction results; S3. Make real-time judgments on the quantified potential safety risks and prediction results, obtain judgment results, automatically trigger the alarm mechanism based on the judgment results, and issue alarm information; based on the alarm information, adjust the mining operation plan and resource allocation, and provide suggestions for emergency response measures.

2. The mining safety information management method based on big data according to claim 1 is characterized in that: The S1 specifically includes: In the implementation process of the dynamic weighted graph convolution enhancement algorithm, features are extracted from the preprocessed data to obtain feature data; based on the feature data, the relationship between data acquisition devices is expressed in the form of a graph. When constructing the graph, the strength of the relationship between each pair of data acquisition devices is evaluated, and a dynamic weighting coefficient is assigned to each edge of the graph.

3. The mining safety information management method based on big data according to claim 2 is characterized in that: The S1 specifically includes: In the implementation process of the dynamic weighted graph convolution enhancement algorithm, the dynamic weighted graph convolution operation is introduced to fuse the features of each node with the information of its neighboring nodes, and each layer of graph convolution operation is weighted using the weighted dynamic weight coefficient to obtain the feature representation of the node updated by the graph convolution operation. The specific implementation formula is: ； in, Is a node In the Feature representation of the layer; is the activation function; Is a node The set of all neighbor nodes of is the dynamic weighting coefficient; Is a node In the Feature representation of the layer; is the bias term.

4. The mining safety information management method based on big data according to claim 3 is characterized in that: The S1 specifically includes: In the implementation process of the dynamic weighted graph convolution enhancement algorithm, after the graph convolution operation, the feature data is enhanced through a multi-layer perceptron to obtain the enhanced feature data.

5. The mining safety information management method based on big data according to claim 1 is characterized in that: The S2 specifically includes: In the process of implementing the dynamic nonlinear coupling weighted collaborative risk assessment algorithm, the enhanced data is normalized to obtain normalized feature data; the normalized feature data is subjected to nonlinear synergistic effect extraction, and the nonlinear synergistic effect value is obtained by introducing a weighted synergistic effect model. The specific implementation formula is: ； in, It is The feature data at time The nonlinear synergistic effect value of is the number of different nonlinear synergistic effects; It is The weight coefficients of the nonlinear transformation; is the index adjustment coefficient; is the logarithmic adjustment coefficient; is the power adjustment coefficient; Indicates Normalized feature data; is the parameter in the exponential transformation; is a small constant.

6. The mining safety information management method based on big data according to claim 5 is characterized in that: The S2 specifically includes: In the implementation process of the dynamic nonlinear coupling weighted collaborative risk assessment algorithm, a dynamic coupling factor model is introduced to simulate the time-varying coupling relationship between characteristic data and describe the coupling risk. The specific implementation formula is: ； in, It is The feature data at time The coupling risk; is the total number of feature data; It is The characteristic data and The feature data at time The coupling weight of It is The feature data at time The risk function of It is The feature data at time The normalization coefficient of is the time decay factor, which indicates the influence of time change on coupling risk.

7. The mining safety information management method based on big data according to claim 6 is characterized in that: The S2 specifically includes: In the process of implementing the dynamic nonlinear coupling weighted collaborative risk assessment algorithm, the coupling risks of all characteristic data are processed to generate a global safety risk assessment and obtain the analysis results. Based on the analysis results, the weighted average method is used in combination with the coupling risks of the characteristic data to obtain the quantified potential safety risks.

8. The mining safety information management method based on big data according to claim 1 is characterized in that: The S3 specifically includes: The quantified potential safety risks and prediction results are compared with the risk threshold to determine the current safety status of the mining area and obtain the judgment result. The specific judgment formula is as follows: ； ； in, and are the risk thresholds for low risk and high risk; It is based on the determination result of quantified potential safety risks; It is the judgment result based on the prediction result; It is the global risk quantification value, that is, the quantified potential security risk; Indicates the prediction result.

9. A mining safety information management system based on big data, applied to the mining safety information management method based on big data according to claim 1, characterized in that: Includes the following parts: Data acquisition module, data processing and storage module, data analysis and prediction module, real-time alarm module, decision support and management module; The data acquisition module collects raw data, preprocesses the raw data, obtains preprocessed data, and sends the preprocessed data to the data processing and storage module; The data processing and storage module performs enhancement processing on the pre-processed data to obtain enhanced data, and sends the enhanced data to the data analysis and prediction module; The data analysis and prediction module conducts in-depth analysis on the enhanced data through a dynamic nonlinear coupling weighted collaborative risk assessment algorithm to obtain analysis results; Based on the analysis results, the potential safety risks are quantified. At the same time, the enhanced processed data is combined with the historical mining data to conduct prediction analysis and evaluation of potential safety hazards in the mining area, and the prediction results are obtained. The quantified potential safety risks and prediction results are sent to the real-time alarm module. The real-time alarm module makes real-time judgments on the quantified potential safety risks and prediction results, obtains the judgment results, automatically triggers the alarm mechanism based on the judgment results, issues alarm information, and sends the alarm information to the decision support and management module; The decision support and management module, after receiving the alarm information from the real-time alarm module, adjusts the mining operation plan and resource allocation, and provides suggestions for emergency response measures.

Images