A Dynamic Risk Classification and Assessment Method for Regulatory Data of Non-Coal Underground Mines

Abstract

This invention provides a method for dynamic risk classification and assessment of non-coal underground mine regulatory data. The method includes: constructing risk assessment indicators for non-coal mine monitoring and early warning; configuring a risk assessment model based on the different characteristics of non-coal mines, combined with the business needs of safety production risk early warning and prevention and the risk assessment indicators; outputting the dynamic risk classification and assessment results of non-coal mines according to the risk assessment model; and statistically analyzing and displaying the dynamic risk classification and assessment results. This invention's method for dynamic risk classification and assessment of non-coal underground mine regulatory data, based on the different characteristics, business needs, and risk assessment indicators of non-coal mines, configures a risk assessment model, uses the model to output the dynamic risk classification and assessment results of non-coal mines and performs statistical analysis, and displays the dynamic risk classification and assessment results and statistical analysis results, helping inspectors to comprehensively assess the risk situation of non-coal mines and improving inspection efficiency.

Description

Technical Field

[0001] This invention relates to the field of mine risk assessment technology, and in particular to a method for dynamic risk classification and assessment of non-coal underground mine regulatory data. Background Technology

[0002] Non-coal mines are mining sites for all metallic and non-metallic mineral resources other than coal, such as iron ore and phosphate mines. The risk characteristics of non-coal mines differ from those of traditional coal mines; therefore, risk assessment models used for coal mines cannot be directly applied to non-coal mines. Furthermore, the risks in non-coal mines are more complex, and the types of data sources are more diverse; therefore, risk assessments need to be conducted based on regulatory data for non-coal mines.

[0003] In the current technology for supervising non-coal mines, fixed risk indicators and risk warning thresholds are often set for risk monitoring. This static monitoring method cannot cope with complex risk situations and can only present a single risk indicator warning result. Furthermore, when multiple risk indicator warnings are triggered, manual statistics of these warning indicators are still required for further analysis, which is not user-friendly and reduces the efficiency of supervision.

[0004] In view of this, there is an urgent need for a dynamic risk classification and assessment method for non-coal underground mine regulatory data to at least address the above-mentioned shortcomings. Summary of the Invention

[0005] One of the objectives of this invention is to provide a dynamic risk classification and assessment method for non-coal underground mine supervision data. Based on the different characteristics of non-coal mines, combined with the business needs of safety production risk early warning and prevention and control and risk assessment indicators, a risk assessment model is configured. The model is used to output the dynamic risk classification and assessment results of non-coal mines in real time and perform statistical analysis. The dynamic risk classification and assessment results and statistical analysis results are displayed to help inspectors make a comprehensive judgment on the risk situation of non-coal mines and improve the efficiency of inspection.

[0006] This invention provides a method for dynamic risk classification and assessment of non-coal underground mine regulatory data, comprising:

[0007] Construct risk assessment indicators for monitoring and early warning of non-coal mines;

[0008] Based on the different characteristics of non-coal mines, and combined with the business needs of safety production risk early warning and prevention and risk assessment indicators, a risk assessment model is configured.

[0009] Based on the risk assessment model, output the dynamic risk classification assessment results for non-coal mines;

[0010] The results of the dynamic risk grading assessment are statistically analyzed and displayed accordingly.

[0011] Preferably, the configured risk assessment model includes:

[0012] Risk assessment models based on the analytic hierarchy process (AHP), the entropy weight method, the combined weighting method, and extension theory.

[0013] The preferred method for constructing a risk assessment model based on combined weighting is as follows:

[0014] ;

[0015] in, For the first The first sample The evaluation value of each indicator, The first subjective evaluation The weight of each indicator, The first determined for the combined evaluation The weight of each indicator, For the number of samples, For the number of indicators, For the first The weighted comprehensive evaluation of each sample is based on the preset normalized benchmark value.

[0016] The preferred method for constructing a risk assessment model based on extension theory is as follows:

[0017] Each indicator system has a membership function constructed, and the indicator system is divided into: positive indicators and negative indicators;

[0018] Each indicator system is divided into four levels: red, orange, yellow, and blue, with each level corresponding to its own range.

[0019] Perform inverse normalization on the inverse values ​​within the interval;

[0020] For each normalized value, correlation analysis using correlation functions in extension theory is performed.

[0021] The higher the level to which the indicator data falls, the smaller the weight that indicator should be assigned.

[0022] Preferably, the statistical analysis of the dynamic risk grading assessment results and their corresponding display include:

[0023] Calculate the risk assessment level for each risk indicator and obtain a collaborative analysis strategy for risk assessment indicators;

[0024] The risk assessment classification of each risk indicator and the statistical analysis results determined by the collaborative analysis strategy based on the risk assessment indicators are visualized.

[0025] Preferably, the collaborative analysis strategy for obtaining risk assessment indicators includes:

[0026] The risk assessment indicators are grouped according to the same dynamic risk level, and the grouped risk indicators are sorted in descending order of their corresponding dynamic risk levels to obtain the first grouped indicator sequence.

[0027] The risk assessment indicators are grouped according to their correlation with other indicators, and the grouped risk indicators are sorted in descending order of the sum of their corresponding dynamic risk levels to obtain the second grouped indicator sequence.

[0028] Determine the positioning boxes that move from the first sequence position from the first aggregation index sequence and the second aggregation index sequence respectively. The selection content of the positioning boxes includes: the first aggregation index and the second aggregation index.

[0029] If the positioning box meets the positioning box stopping condition, extract the selection content features of the corresponding selected content;

[0030] Based on the characteristics of the selected content, determine the collaborative analysis strategy for risk assessment indicators;

[0031] The stopping condition for the positioning box is: the risk dynamic level corresponding to the first aggregation indicator is greater than the preset first threshold and / or the risk dynamic level corresponding to the second aggregation indicator is greater than the preset second threshold, and there is a target risk assessment indicator in the positioning box, and the target risk assessment indicator exists in both the first aggregation indicator and the second aggregation indicator selected by the positioning box.

[0032] The selected content features include: the difference between the dynamic risk level of the target risk assessment indicator and the first threshold, the correlation between the target risk assessment indicator and other risk assessment indicators in the corresponding second aggregation indicators, and the comparison relationship of the dynamic risk levels.

[0033] Preferably, the statistical analysis and display of the dynamic risk grading assessment results also includes:

[0034] Records are determined based on historical monitoring intentions, and multiple standard intention trend descriptions are identified.

[0035] Based on the locations where the inspectors conduct their inspections, determine the sequence of risk assessment items to be reviewed;

[0036] Based on the target project type switching position in the risk assessment item sequence, the risk assessment item sequence can be split and viewed to obtain the split sub-sequences;

[0037] Traverse the subsequences based on their relative sequence position to the sequence of risk assessment items being viewed, and extract the intent trend description of the subsequence being traversed.

[0038] Match the intent trend description with the standard intent trend description to obtain the matching rate fluctuation curve during the traversal process;

[0039] If the curve characteristics of the matching rate fluctuation curve conform to the characteristics of the standard curve, then obtain the target intent corresponding to the standard intent trend description of the corresponding fluctuation curve.

[0040] A quick way to view the target intent sent to the monitoring personnel.

[0041] This invention provides a dynamic risk classification and assessment system for non-coal underground mine supervision data, comprising:

[0042] The indicator construction module is used to construct risk assessment indicators for monitoring and early warning of non-coal mines.

[0043] The model configuration module is used to configure risk assessment models based on the different characteristics of non-coal mines, combined with the business needs and risk assessment indicators for safety production risk early warning and prevention.

[0044] The assessment results output module is used to output the dynamic risk classification assessment results of non-coal mines based on the risk assessment model.

[0045] The analysis and display module is used to statistically analyze the results of dynamic risk grading assessment and to display them accordingly.

[0046] Preferably, the risk assessment models configured by the model configuration module include:

[0047] Risk assessment models based on the analytic hierarchy process (AHP), the entropy weight method, the combined weighting method, and extension theory.

[0048] Preferably, the model configuration module constructs a risk assessment model based on combined weighting as follows:

[0049] ;

[0050] in, For the first The first sample The evaluation value of each indicator, The first subjective evaluation The weight of each indicator, The first determined for the combined evaluation The weight of each indicator, For the number of samples, For the number of indicators, For the first The weighted comprehensive evaluation of each sample is based on the preset normalized benchmark value.

[0051] The beneficial effects of this invention are as follows:

[0052] Based on the different characteristics of non-coal mines, this invention combines the business needs of safety production risk early warning and prevention with risk assessment indicators, configures a risk assessment model, and uses the model to output the dynamic risk classification assessment results of non-coal mines in real time and perform statistical analysis. The dynamic risk classification assessment results and statistical analysis results are displayed to help inspectors make a comprehensive judgment on the risk situation of non-coal mines and improve the efficiency of inspection.

[0053] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in this application.

[0054] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0055] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0056] Figure 1 This is a schematic diagram of a dynamic risk classification and assessment method for non-coal underground mine supervision data in an embodiment of the present invention;

[0057] Figure 2 This is a schematic diagram of a dynamic risk classification and assessment system for non-coal underground mine supervision data in an embodiment of the present invention. Detailed Implementation

[0058] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0059] This invention provides a method for dynamic risk classification and assessment of non-coal underground mine regulatory data, such as... Figure 1 As shown, it includes:

[0060] Construct risk assessment indicators for monitoring and early warning of non-coal mines;

[0061] Based on the different characteristics of non-coal mines, and combined with the business needs of safety production risk early warning and prevention and risk assessment indicators, a risk assessment model is configured.

[0062] The configured risk assessment model includes:

[0063] Risk assessment models based on the analytic hierarchy process, risk assessment models based on the entropy weight method, risk assessment models based on combined weighting, and risk assessment models based on extension theory;

[0064] Based on the risk assessment model, output the dynamic risk classification assessment results for non-coal mines;

[0065] The results of the dynamic risk grading assessment are statistically analyzed and displayed accordingly.

[0066] The working principle and beneficial effects of the above technical solution are as follows:

[0067] Non-coal mine regulatory data includes monitoring and early warning data and basic data from mining enterprises. Risk assessment indicators are constructed based on the data types, monitoring areas, monitoring location relationships, and data change patterns of the non-coal mine monitoring network. These indicators include: indicator names and evaluation logic. Based on the constructed risk assessment indicators and the actual business needs of safety production risk early warning and prevention, a suitable risk assessment model is built. These models include: risk assessment models based on the Analytic Hierarchy Process (AHP), risk assessment models based on the entropy weight method, risk assessment models based on combined weighting, and risk assessment models based on extension theory. Model configuration includes: model parameter configuration, monitoring scope configuration, early warning level configuration, personnel-level early warning configuration, message management, and model start / stop control. After configuration, the dynamic risk grading assessment results are output. The following is a specific example illustrating the process of outputting the dynamic risk grading assessment results.

[0068] The risk assessment model is for suspected overcapacity production and is applied to underground mines. Data sources include power monitoring data, personnel undergrounding data, and enterprise production status data. Red Alert: In normal production mines, the number of personnel going underground is more than 30% higher than the historical average for the past 6 months, the monthly electricity consumption is more than 30% higher than the historical average for the past 6 months, and it is also more than 30% higher than the unit capacity electricity consumption of similar underground mines. Unit capacity electricity consumption = Total electricity consumption / Design capacity / 12. Yellow Alert: In normal production mines, the number of personnel going underground is more than 20% higher than the historical average for the past 6 months, the monthly electricity consumption is more than 20% higher than the historical average for the past 6 months, and it is also more than 20% higher than the unit capacity electricity consumption of similar underground mines.

[0069] The risk assessment model is for suspected illegal mining and is applied to underground mines. Data sources include electronic seal data and personnel underground data. Red Alert: ① For mines that have been shut down for a long time, the number of personnel underground monitored by electronic seals exceeds 30% of the historical average; ② For mines that have been shut down, the number of personnel underground exceeds 30% of the approved number, and the daily electricity consumption exceeds 30% of the historical average for the past month. Yellow Alert: ① For mines that have been shut down for a long time, the number of personnel underground monitored by electronic seals exceeds 20% of the historical average; ② For mines that have been shut down, the number of personnel underground exceeds 20% of the approved number, and the daily electricity consumption exceeds 20% of the historical average for the past month.

[0070] The risk assessment model is for suspected over-production and is applied in underground mines. Data sources include data on personnel exceeding permitted working hours. Red Alert: The number of personnel exceeding permitted working hours exceeds 15% of the total number of personnel entering the mine that day, and the average overtime exceeds 4 hours. Orange Alert: The number of personnel exceeding permitted working hours exceeds 10% of the total number of personnel entering the mine that day, and the average overtime exceeds 3 hours. Yellow Alert: The number of personnel exceeding permitted working hours exceeds 10% of the total number of personnel entering the mine that day.

[0071] The risk assessment model is for detecting intrusions into restricted areas and is applicable to underground mines and open-pit mines.

[0072] The risk assessment model is used for detecting speeding of transport vehicles, illegal passenger transport in mining trucks, and sleeping on duty or leaving the post. It is applicable to both open-pit and underground mines.

[0073] The risk assessment model is used for detecting personnel climbing over conveyor belts, identifying and detecting smoke and fire, and detecting violations such as "vehicles not allowing pedestrians to pass," and its application scope is underground mines;

[0074] Data sources include video AI data and video stream data.

[0075] By integrating with video AI systems in underground and open-pit mines, early warning data can be accessed and displayed. Alternatively, video models can be integrated for AI analysis to achieve image analysis and early warning.

[0076] The first risk assessment model involves the unplanned deletion of abnormal monitoring points, applied in underground mines, with data sources including safety monitoring data. Ground monitoring personnel delete monitoring points that have already exceeded limits (or are expected to exceed limits) and then redefine them on different ports or substations, preventing inspectors from finding the original monitoring points and thus shielding the alarm risk from regulatory inspection. The second risk assessment model involves abnormal sensor interruption, also applied in underground mines, with data sources including safety monitoring data. When sensor readings are rising, human intervention causes data interruption to prevent the system from issuing an alarm. Data transmission resumes normally after the readings return to normal, but this does not guarantee accurate monitoring of gas concentration and requires verification and correction.

[0077] The risk assessment model monitors non-natural fluctuations in the data, and its application scope is underground mines. The data source includes safety monitoring data. The fluctuations that appear after data modification exhibit non-natural data fluctuations. These fluctuations differ from normal data fluctuations, characterized by sudden spikes followed by a period of relatively stable overall data with rapid changes after the spikes.

[0078] The risk assessment model is based on leaderless supervision and is applied in underground mines. Data sources include personnel location monitoring data. h represents the leaderless supervision time, with blue alerts (h ≤ 1 hour), yellow alerts (1 hour < h ≤ 2 hours), orange alerts (2 hours < h ≤ 6 hours), and red alerts (h > 6 hours).

[0079] The risk assessment model for proxy clocking in / out is applied in underground mines, and the data source includes personnel location monitoring data. By comparing the underground trajectory maps of the shift leader and the underground workers, if there is a high degree of overlap in their trajectories, it is determined that the shift leader is suspected of having their clocking in / out done by a proxy.

[0080] The risk assessment model is a graded alarm system for CO monitoring exceeding limits, applied in underground mines, with data sources including safety monitoring data. Y represents CO concentration, with blue alerts (24ppm ≤ Y < 50ppm, duration less than 20 minutes); yellow alerts (50ppm ≤ Y < 100ppm); orange alerts (100ppm ≤ Y < 200ppm); and red alerts (① Y ≥ 200ppm; ② Two CO sensors simultaneously alarm, exceeding 100ppm, and duration exceeding 20 minutes; ③ Carbon monoxide concentration change exceeding 5ppm within one minute). The duration of any of these conditions exceeds one minute.

[0081] The risk assessment model is based on a temperature over-limit alarm, applied in underground mines, and the data source includes safety monitoring data. Red Alert: ① Ambient temperature sensor alarm value exceeds 30% of the threshold for more than 1 minute; ② Equipment temperature sensor alarms more than 10 times (including 10 times) in the past 30 days. Orange Alert: ① Ambient temperature sensor alarm value exceeds 20% of the threshold for more than 1 minute; ② Temperature sensor alarms more than 5 times in the past 30 days. Yellow Alert: Ambient temperature sensor alarm value exceeds 10% of the threshold for more than 1 minute.

[0082] The risk assessment model uses smoke sensor alarms and is applied in underground mines. Data sources include safety monitoring data. Red Alert: Smoke alarm lasting more than 60 minutes. Orange Alert: Smoke alarm lasting more than 30 minutes. Yellow Alert: Smoke alarm lasting more than 10 minutes. Blue Alert: Smoke alarm lasting less than 10 minutes.

[0083] The risk assessment model is based on NOx sensor over-limit alarms, applied in underground mines, with data sources including safety monitoring data. W represents the NO sensor monitoring value. Yellow alert: 2.5ppm ≤ W < 5ppm, duration greater than 1 minute. Orange alert: 5ppm ≤ W < 10ppm, duration greater than 1 minute. Red alert: W ≥ 10ppm, duration greater than 1 minute.

[0084] The risk assessment model is based on sulfur dioxide sensor over-limit alarms, applied in underground mines, with data sources including safety monitoring data. Z represents the sulfur dioxide sensor monitoring value: Blue alert: 3ppm ≤ Z < 5ppm (duration greater than 1 minute); Yellow alert: 5ppm ≤ Z ≤ 8ppm (duration greater than 1 minute); Orange alert: Z ≤ 10ppm (duration greater than 1 minute); Red alert: Z > 10ppm (duration greater than 1 minute).

[0085] The risk assessment model is based on an over-limit alarm from a hydrogen sulfide sensor, applied in underground mines, with data sources including safety monitoring data. E represents the hydrogen sulfide sensor reading: Blue alert: 4ppm ≤ E < 6.6ppm (duration greater than 1 minute); Yellow alert: E ≤ 10ppm (duration greater than 1 minute); Orange alert: E ≤ 15ppm (duration greater than 1 minute); Red alert: E > 20ppm (duration greater than 1 minute).

[0086] The risk assessment model is based on oxygen concentration alarms and is applied in underground mines. Data sources include safety monitoring data. R represents the oxygen sensor monitoring value. Blue alert: 19% ≤ R < 21% (duration greater than 1 minute). Yellow alert: 18% ≤ R < 19% (duration greater than 1 minute). Orange alert: 17% ≤ R < 18% (duration greater than 1 minute). Red alert: 16% ≤ R < 17% (duration greater than 1 minute).

[0087] The risk assessment model is for early warning of slope instability risk, applied in open-pit mines, and the data source includes slope monitoring data. Red Alert: ① Red alert issued by slope displacement monitoring; ② Sudden change of more than 5% in the monitoring values ​​of multiple displacement sensors in the same area; ③ Simultaneous alarm of displacement acceleration from two or more sensors. Orange Alert: ① Sudden increase of 10% in displacement monitoring value from a single sensor; ② Sensor orange alert. Yellow Alert: ① Sudden increase of 5% in displacement monitoring value from a single sensor; ② Sensor yellow alert.

[0088] The risk assessment model is for early warning of dam instability risk, applied to tailings dams, and data sources include environmental monitoring data. Red Alert: ① Displacement monitoring shows a red alert; ② Multiple displacement sensor readings change abruptly by more than 10%; ③ Two or more sensors simultaneously trigger displacement acceleration alarms. Orange Alert: ① Single sensor displacement monitoring value increases abruptly by 10%; ② Sensors trigger orange alerts. Yellow Alert: ① Single sensor displacement monitoring value increases abruptly by 5%; ② Sensors trigger yellow alerts. Blue Alert: Sensors trigger blue alerts.

[0089] The risk assessment model is for system stability analysis and is applied to tailings ponds, underground mines, and open-pit mines. Data sources include various monitoring systems. Orange Alert: System data upload is abnormal, with cumulative downtime exceeding 5 hours per day. Yellow Alert: ① The cumulative monthly network outages of the monitoring and early warning system exceed 10 times; the cumulative monthly network outages of the mine monitoring and early warning system exceed 5 times, ② and the cumulative downtime within a day is 1 hour ≤ downtime ≤ 5 hours. Blue Alert: The network outage time of the mine monitoring and early warning system is less than 1 hour.

[0090] The risk assessment model is based on meteorological early warning and is applied to tailings ponds, underground mines, and open-pit mines. Data sources include meteorological monitoring data. Red Alert: ① Red alert for the area where the enterprise is located (rainfall, high temperature, typhoon, etc.); ② Orange alert for the area where the enterprise is a key mine. Orange Alert: ① Orange alert for the area where the enterprise is located (rainfall, high temperature, typhoon, etc.); ② Yellow alert for the area where the enterprise is a key mine. Yellow Alert: ① Yellow alert for the area where the enterprise is located (rainfall, high temperature, typhoon, etc.); ② Blue alert for the area where the enterprise is a key mine. Blue Alert: Blue alert for the area where the enterprise is located (rainfall, high temperature, typhoon, etc.). Rainfall: Major risk ≥50mm / h, significant risk 30-50mm / h, general risk 15-30mm / h, low risk <15mm / h. Road surface temperature: Major risk -10℃ and severe icing, significant risk -5-10℃ and considerable icing, general risk -2-5℃ and some icing, low risk -1-2℃ and no icing. Wind force level: Major risk ≥ level 8, significant risk level 6-8, general risk level 4-6, low risk level < level 4.

[0091] The risk assessment model is an earthquake risk early warning model, applicable to tailings ponds, underground mines, and open-pit mines. Data sources include earthquake monitoring data. ① Earthquakes below magnitude 3: Key mines within 20km are classified as Level 2 risk, other mines as Level 3 risk, and mines within 20-50km are classified as Level 4 risk. ② Earthquakes of magnitude 3.0-5.0: Key mining enterprises within 50km are classified as Level 1 risk, and other mines as Level 2 risk; Key mining enterprises within 50-100km are classified as Level 2 risk, and other mines as Level 3 risk; All mines within 100-150km are classified as Level 3 risk, and those within 150-200km are classified as Level 4 risk; ③ Earthquakes of magnitude 5.0-7.0: Enterprises within 100km are classified as Level 1 risk; those within 150km are classified as Level 2 risk; those within 200km are classified as Level 3 risk, and those within 200-300km are classified as Level 4 risk; ④ Earthquakes of magnitude 7.0 and above: Enterprises within 200km are classified as Level 1 risk; those within 300km are classified as Level 2 risk; those within 400km are classified as Level 3 risk, and those within 400-600km are classified as Level 4 risk.

[0092] The results of the dynamic risk classification assessment are statistically analyzed according to dimensions such as region, type of non-coal mine, basic information of non-coal mine, risk assessment level, and risk assessment time. Combined with a visualization platform, multi-dimensional data analysis is achieved, and information such as the current risk status and weak links of non-coal mine safety supervision is extracted and summarized for non-coal mine safety inspectors to view.

[0093] Based on the different characteristics of non-coal mines, this invention combines the business needs of safety production risk early warning and prevention with risk assessment indicators, configures a risk assessment model, and uses the model to output the dynamic risk classification assessment results of non-coal mines in real time and perform statistical analysis. The dynamic risk classification assessment results and statistical analysis results are displayed to help inspectors make a comprehensive judgment on the risk situation of non-coal mines and improve the efficiency of inspection.

[0094] In one embodiment, the method for constructing a risk assessment model based on combined weighting is as follows:

[0095] ;

[0096] in, For the first The first sample The evaluation value of each indicator, The first subjective evaluation The weight of each indicator, The first determined for the combined evaluation The weight of each indicator, For the number of samples, For the number of indicators, For the first The weighted comprehensive evaluation of each sample is based on the preset normalized benchmark value.

[0097] The working principle and beneficial effects of the above technical solution are as follows:

[0098] This invention introduces a risk assessment model based on the analytic hierarchy process (AHP), which optimizes both subjective and objective factors and takes into account both experience and data. It is more suitable for risk assessment of complex and multi-source non-coal mine regulatory data.

[0099] In one embodiment, the method for constructing a risk assessment model based on extension theory is as follows:

[0100] Each indicator system has a membership function constructed, and the indicator system is divided into: positive indicators and negative indicators;

[0101] Each indicator system is divided into four levels: red, orange, yellow, and blue, with each level corresponding to its own range.

[0102] Perform inverse normalization on the inverse values ​​within the interval;

[0103] For each normalized value, correlation analysis using correlation functions in extension theory is performed.

[0104] The higher the level to which the indicator data falls, the smaller the weight that indicator should be assigned.

[0105] The working principle and beneficial effects of the above technical solution are as follows:

[0106] This invention introduces a risk assessment model based on extension theory, which can dynamically handle contradictory risks and more flexibly classify risk levels for complex and multi-source non-coal mine regulatory data.

[0107] In one embodiment, the statistical analysis of the dynamic risk grading assessment results and their corresponding display include:

[0108] Calculate the risk assessment level for each risk indicator and obtain a collaborative analysis strategy for risk assessment indicators;

[0109] The risk assessment classification of each risk indicator and the statistical analysis results determined by the collaborative analysis strategy based on the risk assessment indicators are visualized.

[0110] The collaborative analysis strategy for obtaining risk assessment indicators includes:

[0111] The risk assessment indicators are grouped according to the same dynamic risk level, and the grouped risk indicators are sorted in descending order of their corresponding dynamic risk levels to obtain the first grouped indicator sequence.

[0112] The risk assessment indicators are grouped according to their correlation with other indicators, and the grouped risk indicators are sorted in descending order of the sum of their corresponding dynamic risk levels to obtain the second grouped indicator sequence.

[0113] Determine the positioning boxes that move from the first sequence position from the first aggregation index sequence and the second aggregation index sequence respectively. The selection content of the positioning boxes includes: the first aggregation index and the second aggregation index.

[0114] If the positioning box meets the positioning box stopping condition, extract the selection content features of the corresponding selected content;

[0115] Based on the characteristics of the selected content, determine the collaborative analysis strategy for risk assessment indicators;

[0116] The stopping condition for the positioning box is: the risk dynamic level corresponding to the first aggregation indicator is greater than the preset first threshold and / or the risk dynamic level corresponding to the second aggregation indicator is greater than the preset second threshold, and there is a target risk assessment indicator in the positioning box, and the target risk assessment indicator exists in both the first aggregation indicator and the second aggregation indicator selected by the positioning box.

[0117] The selected content features include: the difference between the dynamic risk level of the target risk assessment indicator and the first threshold, the correlation between the target risk assessment indicator and other risk assessment indicators in the corresponding second aggregation indicators, and the comparison relationship of the dynamic risk levels.

[0118] The working principle and beneficial effects of the above technical solution are as follows:

[0119] After obtaining the risk assessment level for each risk indicator, and considering the correlation between indicators, monitoring personnel often need to coordinate the analysis of multiple indicators. This invention uses a machine learning model based on a risk assessment indicator collaborative analysis strategy to obtain statistical analysis results. Finally, it presents the risk assessment level and statistical analysis results for each risk indicator together, which is highly intelligent. The suitability of the risk assessment indicator collaborative analysis strategy greatly affects the accuracy of machine analysis.

[0120] Existing technologies often involve extensive comparative analysis of risk indicators based on their correlations, which results in a massive amount of data being processed and low analysis efficiency.

[0121] This invention addresses the aforementioned problems by first analyzing scenarios requiring collaborative analysis of indicators. The specific implementation process is as follows: First, risk assessment indicators with the same risk assessment level are grouped into the same indicator set, resulting in multiple aggregated risk indicators. Then, these aggregated risk indicators are sorted in descending order of their corresponding dynamic risk levels to obtain a first aggregated indicator sequence. Next, the sum of the dynamic risk levels corresponding to all risk indicators in the aggregated risk indicators grouped according to their correlation relationships is calculated. Finally, the aggregated risk indicators are sorted in descending order of their dynamic risk level sums to obtain a second aggregated indicator sequence. The correlation relationships are obtained from a pre-configured risk indicator correlation table, such as the correlation between mining depth and surface subsidence / vegetation destruction rate.

[0122] The first set of aggregated indicators is ranked and summarized according to their respective dynamic risk levels. The earlier an indicator appears in the first set of aggregated indicators, the higher the level of risk it represents. The second set of aggregated indicators is ranked according to the dynamic risk levels of related risk assessment indicators. The earlier a related risk assessment indicator appears in the second set of aggregated indicators, the higher the level of risk obtained from the collaborative analysis of those related risk assessment indicators.

[0123] Determine the positioning boxes that move from the first sequence position from the first aggregation index sequence and the second aggregation index sequence respectively. The selection of the positioning boxes includes the first aggregation index and the second aggregation index.

[0124] The positioning box stopping condition is set as follows: the dynamic risk level corresponding to the first aggregated indicator is greater than a preset first threshold and / or the dynamic risk level corresponding to the second aggregated indicator is greater than a preset second threshold. The preset first and second thresholds are both manually set, for example, the first threshold is 3 and the second threshold is 10. The purpose of setting the positioning box stopping condition is to constrain the risk indicators in the selected content to achieve at least one level of early warning, including early warning from a single indicator analysis and collaborative early warning from multiple indicator analyses. Furthermore, the positioning box contains a risk assessment indicator (target risk assessment indicator) that exists in both the first and second aggregated indicators within the selected box. This positioning box stopping condition is used for subsequent feature extraction of the selected content, constraining at least one level of early warning and the existence of two related indicators (target risk assessment indicators), effectively filtering out situations requiring collaborative indicator analysis and improving the efficiency of feature extraction for subsequent selected content.

[0125] Next, the features of the selected content are further extracted. These features include: the difference between the dynamic risk level of the target risk assessment indicator and the first threshold, the correlation between the target risk assessment indicator and other risk assessment indicators in the corresponding second set of indicators, and the comparison of their dynamic risk levels. When determining the collaborative analysis strategy for risk assessment indicators, the selected content features can be matched and obtained from a pre-configured collaborative analysis strategy library. The library contains multiple pre-configured, one-to-one matching selected content features and corresponding collaborative analysis strategies. For example, if the selected content features are: the target risk assessment indicator is mining depth, the difference between its dynamic risk level and the first threshold is -1, and the correlation between mining depth and surface subsidence and vegetation destruction rate (corresponding to other risk assessment indicators in the second set of indicators) is: the deeper the mining depth, the greater the corresponding surface subsidence and vegetation destruction rate; the dynamic risk level of mining depth is lower than the dynamic risk level of greater surface subsidence and vegetation destruction rate. Therefore, the corresponding collaborative analysis strategy for risk assessment indicators is: collaboratively analyze surface subsidence, vegetation destruction rate, and other related risk assessment indicators besides mining depth. The selected content corresponding to the scenarios requiring collaborative analysis of indicators is characterized, and further matched to determine the collaborative analysis strategy for risk assessment indicators, thereby further improving the suitability of the collaborative analysis strategy formulation.

[0126] In one embodiment, the statistical analysis and display of the dynamic risk grading assessment results also includes:

[0127] Records are determined based on historical monitoring intentions, and multiple standard intention trend descriptions are identified.

[0128] Based on the locations where the inspectors conduct their inspections, determine the sequence of risk assessment items to be reviewed;

[0129] Based on the target project type switching position in the risk assessment item sequence, the risk assessment item sequence can be split and viewed to obtain the split sub-sequences;

[0130] Traverse the subsequences based on their relative sequence position to the sequence of risk assessment items being viewed, and extract the intent trend description of the subsequence being traversed.

[0131] Match the intent trend description with the standard intent trend description to obtain the matching rate fluctuation curve during the traversal process;

[0132] If the curve characteristics of the matching rate fluctuation curve conform to the characteristics of the standard curve, then obtain the target intent corresponding to the standard intent trend description of the corresponding fluctuation curve.

[0133] A quick way to view the target intent sent to the monitoring personnel.

[0134] The working principle and beneficial effects of the above technical solution are as follows:

[0135] When inspectors view the interface, inspection points (obtained using gaze extraction technology) are generated on the large visual screen. Based on these inspection points, they can access and view risk assessment items. These risk assessment items are ordered according to the order in which they appear on the corresponding inspection points, resulting in a sequence of risk assessment items. The risk assessment items include: the risk assessment level of an individual risk assessment indicator and the results of collaborative analysis of risk assessment indicators. Generally, inspectors will first view the risk assessment level of an individual risk assessment indicator, and then view the collaborative analysis results. The inspectors' viewing intentions are implicitly conveyed through the alternation of these item types.

[0136] Therefore, a target project type switching position is introduced. This position is the location in the sequence where the risk assessment item changes from the collaborative analysis result of risk assessment indicators to the risk assessment level of a single risk assessment indicator. The split subsequence obtained based on the target project type switching position has the first part of the sequence as the risk assessment level of a single risk assessment indicator and the second part as the collaborative analysis result of risk assessment indicators. Each split subsequence implies an intention trend.

[0137] The process iterates through the subsequences based on their relative positions to the risk assessment item sequence, extracting the intent trend description of the currently being iterated subsequence. This intent trend description is a descriptive vector corresponding to the intent trend described in the currently being iterated subsequence. Since monitors may not be able to accurately view the desired risk assessment indicator's collaborative analysis results all at once, multiple intent trend descriptions will be generated. The nth intent trend description is generated jointly based on the (n-1)th intent description and the nth subsequence. Furthermore, historical monitoring intent determination records include the monitors' historical viewing actions, and the average intent trend description for each stage is determined based on these viewing actions as the standard intent trend description. The matching rate fluctuation curve is a curve showing the change in the matching results of the intention trend description and the standard intention trend description as the action stage progresses. If the curve characteristics of the matching rate fluctuation curve conform to the characteristics of the standard curve (e.g., the difference between the highest and lowest points of the curve does not exceed 5% in matching rate, and the average fluctuation rate of the curve is 80%), then the historical intention described by the corresponding standard intention trend description of the fluctuation curve is taken as the target intention, and a quick viewing path for the target intention is delivered to the monitoring personnel. The quick viewing path is the best indicator viewing and switching strategy corresponding to the target intention on the visualization screen, such as: first switch to which single indicator analysis, and then switch to which multiple indicator analysis results. This invention pushes the target intention between the monitoring personnel's indicator viewing, seamlessly assisting the monitoring personnel in viewing the evaluation results, making it more intelligent.

[0138] This invention provides a dynamic risk classification and assessment system for non-coal underground mine supervision data, such as... Figure 2 As shown, it includes:

[0139] Indicator Construction Module 1 is used to construct risk assessment indicators for monitoring and early warning in non-coal mines;

[0140] Model configuration module 2 is used to configure risk assessment models based on the different characteristics of non-coal mines, combined with the business needs and risk assessment indicators of safety production risk early warning and prevention.

[0141] The assessment result output module 3 is used to output the dynamic risk classification assessment results of non-coal mines based on the risk assessment model.

[0142] Analysis and Display Module 4 is used to statistically analyze the results of dynamic risk grading assessment and display them accordingly.

[0143] In one embodiment, the risk assessment model configured by the model configuration module includes:

[0144] Risk assessment models based on the analytic hierarchy process (AHP), the entropy weight method, the combined weighting method, and extension theory.

[0145] In one embodiment, the model configuration module constructs a risk assessment model based on combined weighting as follows:

[0146] ;

[0147] in, For the first The first sample The evaluation value of each indicator, The first subjective evaluation The weight of each indicator, The first determined for the combined evaluation The weight of each indicator, For the number of samples, For the number of indicators, For the first The weighted comprehensive evaluation of each sample is based on the preset normalized benchmark value.

[0148] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for dynamic risk classification and assessment of non-coal underground mine regulatory data, characterized in that, include: Construct risk assessment indicators for monitoring and early warning of non-coal mines; Based on the different characteristics of non-coal mines, and combined with the business needs of safety production risk early warning and prevention and risk assessment indicators, a risk assessment model is configured. Based on the risk assessment model, output the dynamic risk classification assessment results for non-coal mines; Statistical analysis of the dynamic risk grading assessment results and corresponding display; The statistical analysis of the dynamic risk grading assessment results and their corresponding display include: Calculate the risk assessment level for each risk indicator and obtain a collaborative analysis strategy for risk assessment indicators; The risk assessment classification of each risk indicator and the statistical analysis results determined by the collaborative analysis strategy based on the risk assessment indicators are visualized. Among them, the collaborative analysis strategy for obtaining risk assessment indicators includes: The risk assessment indicators are grouped according to the same dynamic risk level, and the grouped risk indicators are sorted in descending order of their corresponding dynamic risk levels to obtain the first grouped indicator sequence. The risk assessment indicators are grouped according to their correlation with other indicators, and the grouped risk indicators are sorted in descending order of the sum of their corresponding dynamic risk levels to obtain the second grouped indicator sequence. Determine the positioning boxes that move from the first sequence position from the first aggregation index sequence and the second aggregation index sequence respectively. The selection content of the positioning boxes includes: the first aggregation index and the second aggregation index. If the positioning box meets the positioning box stopping condition, extract the selection content features of the corresponding selected content; Based on the characteristics of the selected content, determine the collaborative analysis strategy for risk assessment indicators; The stopping condition for the positioning box is: the risk dynamic level corresponding to the first aggregation indicator is greater than the preset first threshold and / or the risk dynamic level corresponding to the second aggregation indicator is greater than the preset second threshold, and there is a target risk assessment indicator in the positioning box, and the target risk assessment indicator exists in both the first aggregation indicator and the second aggregation indicator selected by the positioning box. The selected content features include: the difference between the dynamic risk level of the target risk assessment indicator and the first threshold, the correlation between the target risk assessment indicator and other risk assessment indicators in the corresponding second aggregation indicators, and the comparison relationship of the dynamic risk levels. The statistical analysis and display of the dynamic risk grading assessment results also includes: Records are determined based on historical monitoring intentions, and multiple standard intention trend descriptions are identified. Based on the locations where the inspectors conduct their inspections, determine the sequence of risk assessment items to be reviewed; Based on the target project type switching position in the risk assessment item sequence, the risk assessment item sequence can be split and viewed to obtain the split sub-sequences; Traverse the subsequences based on their relative sequence position to the sequence of risk assessment items being viewed, and extract the intent trend description of the subsequence being traversed. Match the intent trend description with the standard intent trend description to obtain the matching rate fluctuation curve during the traversal process; If the curve characteristics of the matching rate fluctuation curve conform to the characteristics of the standard curve, then obtain the target intent corresponding to the standard intent trend description of the corresponding fluctuation curve. A quick way to view the target intent sent to the monitoring personnel.

2. The method for dynamic risk classification and assessment of non-coal underground mine regulatory data as described in claim 1, characterized in that, The configured risk assessment model includes: Risk assessment models based on the analytic hierarchy process (AHP), the entropy weight method, the combined weighting method, and extension theory.

3. The method for dynamic risk classification and assessment of non-coal underground mine regulatory data as described in claim 2, characterized in that, The construction method of the risk assessment model based on combined weighting is as follows: ； in, For the first The first sample The evaluation value of each indicator, The first subjective evaluation The weight of each indicator, The first determined for the combined evaluation The weight of each indicator, For the number of samples, For the number of indicators, For the first The weighted comprehensive evaluation of each sample is based on the preset normalized benchmark value.

4. The method for dynamic risk classification and assessment of non-coal underground mine regulatory data as described in claim 2, characterized in that, The method for constructing a risk assessment model based on extension theory is as follows: Each indicator system has a membership function constructed, and the indicator system is divided into: positive indicators and negative indicators; Each indicator system is divided into four levels: red, orange, yellow, and blue, with each level corresponding to its own range. Perform inverse normalization on the inverse values ​​within the interval; For each normalized value, correlation analysis using correlation functions in extension theory is performed. The higher the level to which the indicator data falls, the smaller the weight that indicator should be assigned.

5. A dynamic risk classification and assessment system for non-coal underground mine regulatory data, employing the dynamic risk classification and assessment method for non-coal underground mine regulatory data as described in any one of claims 1-4, characterized in that, include: The indicator construction module is used to construct risk assessment indicators for monitoring and early warning of non-coal mines. The model configuration module is used to configure risk assessment models based on the different characteristics of non-coal mines, combined with the business needs and risk assessment indicators for safety production risk early warning and prevention. The assessment results output module is used to output the dynamic risk classification assessment results of non-coal mines based on the risk assessment model. The analysis and display module is used to statistically analyze the results of dynamic risk grading assessment and to display them accordingly.

6. The non-coal underground mine regulatory data risk dynamic classification and assessment system as described in claim 5, characterized in that, The risk assessment models configured in the model configuration module include: Risk assessment models based on the analytic hierarchy process (AHP), the entropy weight method, the combined weighting method, and extension theory.

7. The non-coal underground mine regulatory data risk dynamic classification and assessment system as described in claim 6, characterized in that, The model configuration module constructs a risk assessment model based on combined weighting as follows: ； in, For the first The first sample The evaluation value of each indicator, The first subjective evaluation The weight of each indicator, The first determined for the combined evaluation The weight of each indicator, For the number of samples, For the number of indicators, For the first The weighted comprehensive evaluation of each sample is based on the preset normalized benchmark value.

Images