Weight vector-based coal mine disaster assessment real-time scoring method and system

Abstract

The invention relates to the technical field of coal mine safety monitoring, and discloses a coal mine disaster assessment real-time scoring method and system based on a weight vector, and the method comprises the steps: collecting multi-dimensional parameter data representing disaster risks in real time through a plurality of sensors disposed in an underground coal mine, and carrying out the preprocessing; based on a predefined multi-dimensional index system, combining multi-dimensional parameter data collected in real time, and generating a weight vector matched with the current underground environment through a dynamic adjustment mechanism; performing standardization processing on the multi-dimensional parameter data collected in real time to obtain a standardization score of each index, and performing weighted calculation on the standardization score and the weight vector to generate a comprehensive real-time risk assessment score; and outputting a real-time risk assessment score, and automatically triggering an early warning signal when the score exceeds a preset safety threshold. According to the method, the problems of static evaluation model, weight immobilization and the like in the prior art are solved, and dynamic, accurate and real-time evaluation of the coal mine disaster risk is realized.

Description

Technical Field

[0001] This invention relates to the field of coal mine safety monitoring technology, and more specifically, to a real-time scoring method and system for coal mine disaster assessment based on weight vectors. Background Technology

[0002] With increasing coal mining depth and intensity, the risks to coal mine safety are becoming increasingly complex. Traditional coal mine hazard assessment methods mainly rely on static weight allocation and periodic manual assessment, which are difficult to adapt to the dynamic changes in the underground environment. Current technologies typically use fixed weights to comprehensively assess parameters such as gas concentration, roof displacement, coal seam temperature, and water inflow, but this static assessment method cannot reflect the actual risk changes of different hazard types at different mining stages and under different geological conditions.

[0003] Current coal mine monitoring systems still have significant shortcomings in risk assessment. Their assessment models lack dynamic adaptability and cannot automatically adjust assessment weights according to the changing trends of real-time monitoring data, resulting in discrepancies between assessment results and actual risk conditions.

[0004] In view of this, a real-time scoring method and system for coal mine disaster assessment based on weight vectors is proposed. Summary of the Invention

[0005] In view of this, the present invention proposes a real-time scoring method and system for coal mine disaster assessment based on weight vectors, in order to overcome the problems of static assessment models, fixed weights, and simplistic early warning mechanisms in the existing technology, and to realize dynamic, accurate, and real-time assessment of coal mine disaster risks.

[0006] To achieve the above objectives, in a first aspect, the present invention provides a real-time scoring method for coal mine disaster assessment based on weight vectors, comprising: Multiple sensors deployed underground in coal mines are used to collect multidimensional parameter data characterizing disaster risk in real time and perform preprocessing. Based on a predefined multi-dimensional index system, combined with the real-time collected multi-dimensional parameter data, a weight vector matching the current downhole environment is generated through a dynamic adjustment mechanism. The real-time collected multidimensional parameter data is standardized to obtain the standardized score of each indicator, and then weighted with the weight vector to generate a comprehensive real-time risk assessment score. The system outputs the real-time risk assessment score and automatically triggers an early warning signal when the score exceeds a preset safety threshold.

[0007] Preferably, the real-time acquisition and preprocessing of multidimensional parameter data characterizing disaster risk specifically includes: The sensor's data acquisition interval is 10 to 30 seconds; The multidimensional parameter data includes gas concentration, roof displacement, coal seam temperature, and water inflow rate; The collected raw data is preprocessed, including outlier detection, standardization transformation, and time series data alignment.

[0008] Preferably, the weight vector based on the predefined multi-dimensional index system, combined with the real-time collected multi-dimensional parameter data, is dynamically adjusted to generate a weight vector that matches the current downhole environment. Specifically: Based on the aforementioned multi-dimensional indicator system, the initial weights of each evaluation indicator are determined using the analytic hierarchy process (AHP) to form an initial weight vector. Continuously monitor the changes in the real-time collected multidimensional parameter data; When the data fluctuation of a specific indicator exceeds the preset fluctuation threshold, the dynamic adjustment model is activated to correct the initial weight vector.

[0009] Preferably, the preset fluctuation threshold is 5% / minute.

[0010] Preferably, the standardized scores of each indicator are obtained, and then weighted and calculated with the weight vector to generate a comprehensive real-time risk assessment score, specifically as follows: The standardized scores of each indicator are weighted and calculated together with the weight vector to form a real-time comprehensive score ranging from 0 to 100 points, where a higher score indicates a higher level of disaster risk.

[0011] Preferably, the step of outputting the real-time risk assessment score and automatically triggering an early warning signal when the score exceeds a preset safety threshold is as follows: The real-time risk assessment score and the scores of each indicator are displayed in the form of visual charts; The warning signals include audible and visual alarms and the display of a red warning icon in a visual interface.

[0012] Preferably, the multi-dimensional indicator system includes assessment indicators related to gas disasters, roof disasters, and water hazards.

[0013] Secondly, the present invention provides a real-time scoring system for coal mine disaster assessment based on weight vectors, comprising: The data acquisition module is used to collect multi-dimensional parameter data characterizing disaster risk in real time through multiple sensors deployed underground in coal mines; The weight generation module is used to generate a weight vector that matches the current downhole environment based on a predefined multi-dimensional indicator system and the real-time collected multi-dimensional parameter data through a dynamic adjustment mechanism. The scoring calculation module is used to standardize the real-time collected multi-dimensional parameter data to obtain the standardized score of each indicator, and to perform weighted calculation with the weight vector to generate a comprehensive real-time risk assessment score. The output warning module is used to output the real-time risk assessment score and automatically trigger a warning signal when the score exceeds a preset safety threshold.

[0014] Preferably, the data acquisition module specifically comprises: The sensor's data acquisition interval is 10 to 30 seconds; The multidimensional parameter data includes gas concentration, roof displacement, coal seam temperature, and water inflow rate; The collected raw data is preprocessed, including outlier detection, standardization transformation, and time series data alignment.

[0015] Preferably, the weight generation module specifically comprises: Based on the aforementioned multi-dimensional indicator system, the initial weights of each evaluation indicator are determined using the analytic hierarchy process (AHP) to form an initial weight vector. Continuously monitor the changes in the real-time collected multidimensional parameter data; When the data fluctuation of a specific indicator exceeds the preset fluctuation threshold, the dynamic adjustment model is activated to correct the initial weight vector.

[0016] This application discloses a real-time scoring method and system for coal mine disaster assessment based on weighted vectors. This method collects multi-dimensional parameter data characterizing disaster risk in real time using multiple sensors deployed underground in the coal mine. The raw data is preprocessed to ensure the quality and consistency of the assessment data, effectively solving the problem that multi-source heterogeneous data is difficult to directly use for comprehensive assessment, thus laying a data foundation for subsequent accurate assessment. Based on a predefined multi-dimensional indicator system and combined with the real-time collected multi-dimensional parameter data, a weighted vector matching the current underground environment is generated through a dynamic adjustment mechanism. This enables the assessment model to dynamically adjust the weights of each indicator according to actual risk changes, significantly improving the accuracy and timeliness of the assessment results. After standardizing the real-time collected multi-dimensional parameter data, a weighted calculation is performed with the dynamically generated weighted vector to generate a comprehensive real-time risk assessment score. This score can comprehensively and objectively reflect the current overall safety status of the coal mine, providing a scientific basis for safety management decisions. By outputting a real-time risk assessment score and automatically triggering an early warning signal when the score exceeds a preset safety threshold, a seamless connection from risk assessment to early warning response is achieved. This changes the limitations of traditional single-parameter alarms, enabling the early detection of potential complex risks, effectively reducing false alarm and missed alarm rates, and improving the reliability and effectiveness of coal mine safety early warning. Attached Figure Description

[0017] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 A flowchart illustrating a real-time scoring method for coal mine disaster assessment based on weight vectors, provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of a real-time scoring system for coal mine disaster assessment based on weight vectors, provided in an embodiment of the present invention. Detailed Implementation

[0018] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, embodiments and features in the embodiments of the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0019] like Figure 1 As shown in some embodiments of this application, this embodiment provides a real-time scoring method for coal mine disaster assessment based on weight vectors. Specifically, the method includes the following steps: Step S101: Collect multi-dimensional parameter data characterizing disaster risk in real time through multiple sensors deployed underground in the coal mine, and perform preprocessing.

[0020] As mentioned above, this step aims to obtain raw data reflecting the safety status of underground coal mines and to standardize it, providing reliable data input for subsequent weight vector generation and comprehensive score calculation.

[0021] Specifically, the system deploys a monitoring network composed of various types of sensors in key underground coal mine operating areas, geologically complex areas, and areas prone to historical disasters. These sensors operate continuously at preset acquisition frequencies, collecting multi-dimensional parameter data closely related to coal mine disaster risks in real time. The collected raw data first undergoes data cleaning to remove outliers caused by sensor malfunctions, signal interference, etc.; then, it undergoes standardization transformation to convert parameter data with different dimensions and value ranges into comparable standardized scores; finally, it performs time-series alignment processing to unify data from different acquisition frequencies onto the same time base, ensuring the consistency of each parameter data in the time dimension, and providing a time-synchronized data foundation for subsequent dynamic weight adjustments and comprehensive scoring.

[0022] For example, taking a typical coal mining face as an example, the system deploys the following sensors: methane concentration sensors are placed in the intake airway, return airway and goaf of the working face to collect methane concentration data every 15 seconds; displacement sensors and pressure sensors are placed on the roof and sides of the working face to collect roof subsidence and surrounding rock pressure data every 20 seconds; temperature sensors are placed on the coal wall and roadway floor to collect coal seam temperature data every 30 seconds; and flow meters and water pressure sensors are placed in drainage ditches and borehole water outlets to collect water inflow and water pressure data every 30 seconds.

[0023] When the system is running, these sensors continuously collect data. For example, the raw data collected at a certain moment might be: gas concentration 0.8%, roof displacement 15mm, coal seam temperature 35℃, and water inflow 2.5m³ / h. The system first determines whether these data are within a reasonable range (e.g., whether the gas concentration exceeds the instrument's range). If a significant anomaly is found (e.g., the gas concentration suddenly jumps to 10%), it is marked as an outlier and interpolated. Then, each parameter is converted into a standardized score of 0-10 (e.g., 0.8% gas concentration corresponds to 6 points, 15mm roof displacement corresponds to 5 points, etc.). Finally, the data collected at different frequencies are aligned to the most recent whole-minute time point to form a set of time-synchronized standardized data for subsequent analysis.

[0024] It should be noted that, in specific implementation scenarios, an adaptive sensor network deployment scheme can be adopted based on the above solutions. This means the system can dynamically adjust the deployment location and density of sensors according to changes in coal mining progress and geological conditions. For example, when the working face approaches a fault zone, the number of sensors in that area can be automatically increased to improve monitoring accuracy. A multimodal data fusion scheme can also be adopted, integrating non-contact monitoring devices such as video surveillance, infrared thermal imaging, and acoustic emission in addition to conventional sensors to collect visual, thermal, and acoustic multimodal data, enriching the information dimensions of risk assessment. An edge computing preprocessing scheme can also be adopted, deploying edge computing devices at underground monitoring nodes to achieve localized data preprocessing. This reduces the computational burden on the ground data processing center and ensures data processing continuity during network interruptions. All of the above optional solutions fall within the scope of protection of this application.

[0025] Step S102: Based on a predefined multi-dimensional index system and combined with the real-time collected multi-dimensional parameter data, a weight vector matching the current downhole environment is generated through a dynamic adjustment mechanism.

[0026] As mentioned above, this step aims to overcome the shortcomings of fixed weights in traditional assessment methods, enabling the assessment model to automatically adjust the relative importance of each indicator according to changes in the actual downhole risk situation.

[0027] A multi-dimensional indicator system is pre-established, covering major coal mine hazard types, including but not limited to gas hazards, roof falls, and water hazards. Each hazard type includes several specific assessment indicators; for example, gas hazard indicators include gas concentration, gas concentration change rate, and gas emission rate. The system determines initial weights for each indicator based on the analytic hierarchy process (AHP) or expert experience, forming an initial weight vector.

[0028] Building upon this foundation, the system introduces a dynamic adjustment mechanism to continuously monitor changes in the real-time collected multi-dimensional parameter data. When the fluctuation range of a certain indicator exceeds a preset fluctuation threshold, the system determines that the current risk of that indicator has significantly increased, and its influence in the comprehensive assessment should be increased. At this point, the system activates the dynamic adjustment model to correct the initial weight vector, appropriately increasing the weight proportion of that indicator while correspondingly decreasing the weights of other indicators, ensuring that the sum of the weight vectors is 1. In this way, the generated weight vector can reflect the main risk characteristics of the current downhole environment in real time, providing a dynamically adapted weight basis for subsequent comprehensive scoring.

[0029] For example, suppose the system's preset multi-dimensional indicator system includes three main disaster types: gas disaster, roof disaster, and water disaster, with initial weights of 0.5, 0.3, and 0.2, respectively.

[0030] Under normal mining conditions, the system uses this initial weight vector for evaluation. At a certain moment, the system detected that the gas concentration in the return airway of the working face increased from 0.6% to 0.9% within 1 minute, a change of 5%, exceeding the preset 5% fluctuation threshold. The system determined that the gas risk had increased significantly and activated the dynamic adjustment mechanism.

[0031] The system automatically increases the weight of gas hazard indicators, for example, from 0.5 to 0.7, while correspondingly decreasing the weights of roof hazard and water hazard to 0.2 and 0.1, respectively, forming a new weight vector [0.7, 0.2, 0.1]. This adjustment means that in the current environment, gas risk has become the most important assessment factor, and the assessment model will pay more attention to changes in gas-related parameters.

[0032] For example, if the system detects that the flow rate in a certain area increases by more than 5% in a short period of time, it will increase the weight of the water hazard index and decrease the weight of other indicators, so that the assessment focus shifts to water hazard prevention and control.

[0033] It should be noted that, in specific implementation scenarios, a multi-level threshold dynamic adjustment scheme can be adopted based on the above solutions. This involves setting multiple fluctuation threshold levels (e.g., 3%, 5%, 8%), with different levels of data fluctuation triggering different degrees of weight adjustments, achieving a more refined risk response. A time-window weighted adjustment scheme can be adopted, where the dynamic adjustment model adjusts weights based on data change trends within a sliding time window (e.g., the most recent 5 minutes, 10 minutes), avoiding erroneous adjustments caused by instantaneous abnormal fluctuations. A disaster coupling weight adjustment scheme can be adopted, where when multiple disaster indicators show abnormal fluctuations simultaneously, the system can identify disaster coupling phenomena and adjust the weights of related indicators collaboratively. For example, when roof displacement increases simultaneously with gas emission, the weights of both roof and gas indicators are increased. A regionally differentiated weight strategy scheme can be adopted, where the system can preset different initial weight vectors and adjustment strategies based on the geological conditions and mining stages of different underground areas. For example, deeper mining areas initially assign higher weights to gas indicators, while shallower areas focus more on roof stability. All of the above optional schemes fall within the scope of protection of this application.

[0034] Step S103: Standardize the real-time collected multi-dimensional parameter data to obtain the standardized score of each indicator, and calculate the weighted score with the weight vector to generate a comprehensive real-time risk assessment score.

[0035] As mentioned above, this step aims to combine the preprocessed multidimensional parameter data with the dynamically generated weight vector to generate a comprehensive score that can fully and intuitively reflect the overall safety status of the current coal mine.

[0036] In the specific execution process, the system first converts the preprocessed multidimensional parameter data from step S101 into standardized scores with unified dimensions according to preset standardization rules. The standardization process considers the physical characteristics and risk thresholds of each indicator, mapping the original parameter values ​​to a scoring range of 0-10 points, where 0 points represent no risk and 10 points represent extremely high risk. For example, a high score corresponds to a gas concentration close to the safety limit, and a low score corresponds to a concentration far from the limit.

[0037] Subsequently, the system invokes the dynamic weight vector generated in step S102, which matches the current environment. The standardized score of each indicator is weighted and calculated by multiplying its corresponding weight value; specifically, it is the sum of the products of each indicator's score and its weight. This calculation process ensures that higher-risk indicators (reflected by dynamic weights) have a larger weight in the final score, thus accurately reflecting the main sources of current risk.

[0038] Ultimately, the system converts the weighted calculation results into a comprehensive real-time risk assessment score ranging from 0 to 100. This score is updated every 30 seconds to 1 minute to ensure the timeliness of the assessment results and provide a continuous and dynamic risk view for coal mine safety management.

[0039] For example, suppose the system currently collects the following standardized scores: gas hazard index is 7.5 points; roof hazard index is 6.0 points; and water hazard index is 4.5 points. Meanwhile, according to the dynamic adjustment mechanism in step S102, the system generates the following weight vectors: gas hazard weight is 0.7; roof hazard weight is 0.2; and water hazard weight is 0.1.

[0040] The system performs a weighted calculation: gas contribution is 7.5 × 0.7 = 5.25; roof contribution is 6.0 × 0.2 = 1.2; water hazard contribution is 4.5 × 0.1 = 0.45. The weighted sum of these contributions is: 5.25 + 1.2 + 0.45 = 6.9. Finally, this result is linearly converted into a comprehensive score of 0-100: 6.9 × 10 = 69 points.

[0041] Therefore, the system outputs a current real-time risk assessment score of 69, which falls under the medium risk level (60-80 points), prompting managers to pay close attention to gas risks and take corresponding prevention and control measures.

[0042] It should be noted that, in specific implementation scenarios, a nonlinear scoring mapping scheme can be adopted based on the above solutions. This means the standardization process can use a nonlinear mapping function, such as an exponential or logarithmic function, to give higher score increases to parameter values ​​approaching the danger threshold, thus enhancing sensitivity to critical risks. A multi-level comprehensive scoring scheme can also be adopted, where the system can generate multi-level scoring results, including an overall comprehensive score, sub-scores for each hazard type (e.g., gas risk score, roof risk score), and regional scores (e.g., working face score, roadway score), forming a multi-dimensional risk view. Finally, a scoring confidence assessment scheme can be adopted, where the system can calculate the confidence level of the comprehensive score based on factors such as data quality, sensor stability, and weight adjustment range, providing managers with a reference for the reliability of the scoring. All of the above optional schemes fall within the scope of protection of this application.

[0043] Step S104: Output the real-time risk assessment score, and automatically trigger an early warning signal when the score exceeds a preset safety threshold.

[0044] As mentioned above, this step aims to transform the abstract comprehensive score into an intuitive information presentation and establish an automated early warning and linkage mechanism to ensure that risk information can be communicated to relevant personnel in a timely and effective manner, thus gaining valuable time for emergency decision-making and the implementation of prevention and control measures.

[0045] The system outputs and displays the real-time risk assessment score generated in step S103 through various methods. The current comprehensive score and its corresponding risk level are displayed in real time on the main control screen of the coal mine monitoring center, the mobile terminals of managers at all levels, and display devices in the on-site work area. The scoring information is presented in multiple formats, including numbers, color coding, and progress bars, ensuring the information is intuitive and easy to read.

[0046] Meanwhile, the system presets one or more safety thresholds to classify different risk levels and trigger corresponding early warning responses. When the comprehensive score exceeds the preset safety threshold for the first time, the system immediately and automatically triggers an early warning signal. The early warning signal includes not only traditional audible and visual alarms but also information push notifications through various communication channels to ensure that early warning information reaches all relevant personnel. The system also supports a tiered early warning mechanism, with different score ranges corresponding to different levels of early warning response measures, achieving precise and differentiated management of early warnings.

[0047] For example, suppose the system sets the following safety thresholds and warning rules: 60 points triggers a yellow warning (medium risk); 80 points triggers an orange warning (high risk); and 90 points triggers a red warning (extremely high risk).

[0048] When the system calculates a real-time risk assessment score of 65 points, exceeding the yellow warning threshold of 60 points, the system automatically performs the following actions: On the monitoring center's large screen, the comprehensive scoring area changes from green to yellow, displaying a "medium risk" label; an intermittent beeping sound lasts for 5 seconds; a yellow warning notification is pushed to the mobile terminals of the shift leader and safety officer, prompting "Gas risk increased, please strengthen monitoring"; and the video surveillance footage of the relevant area is automatically retrieved and displayed on the side screen of the monitoring center.

[0049] If the score continues to rise to 85 points, exceeding the orange warning threshold of 80 points, the system will further upgrade the warning: the comprehensive scoring area will change from yellow to orange, displaying a "high risk" label; the beeping sound will change to a rapid, continuous tone; an orange warning will be pushed to the on-duty mine manager, ventilation area manager, and other management personnel, prompting "Activate the emergency plan and prepare to evacuate"; the emergency broadcast in the relevant area will be automatically turned on, playing a pre-recorded warning voice.

[0050] If the score reaches 92 points, a red alert is triggered, and the system will execute the highest level of response: the comprehensive scoring area will turn red and flash, displaying "extremely high risk, evacuate immediately"; the highest priority alarm will sound; the evacuation order will be automatically broadcast to the entire mine; and emergency ventilation, power outage and other automatic control measures will be activated simultaneously.

[0051] It should be noted that, in specific implementation scenarios, a multi-channel early warning linkage scheme can be adopted based on the above solutions. This means that early warning signals can be simultaneously released through multiple channels such as audible and visual alarms, LED displays, mobile APP push notifications, SMS, telephone, emergency broadcasts, and automatic control systems, ensuring full coverage and high reliability of information transmission. A regionalized precision early warning scheme can be adopted, whereby the system can issue warnings only to high-risk areas and adjacent areas based on the regional location of the scoring source, avoiding resource waste and personnel panic caused by indiscriminate alarms throughout the mine. An automatic early warning response recording scheme can be adopted, whereby the system automatically records the trigger time, scoring value, triggering reason, response measures, and handling results of each early warning, forming a complete early warning event archive for post-event analysis and accountability. All of the above optional schemes fall within the scope of protection of this application.

[0052] In some embodiments of this application, the real-time acquisition and preprocessing of multidimensional parameter data characterizing disaster risk specifically includes: The sensor's data acquisition interval is 10 to 30 seconds; The multidimensional parameter data includes gas concentration, roof displacement, coal seam temperature, and water inflow rate; The collected raw data is preprocessed, including outlier detection, standardization transformation, and time series data alignment.

[0053] As mentioned above, various types of sensors are deployed in different areas of the coal mine to collect key parameter data reflecting disaster risk in real time. The sensors continuously collect data at set time intervals of 10 to 30 seconds, ensuring real-time data transmission while avoiding excessive network and system load due to high collection frequency. Specific parameters collected include: gas concentration, used to monitor the accumulation of combustible gas underground and prevent explosions caused by excessive gas levels; roof displacement, used to monitor the stability of the roof in roadways and working faces and assess the risk of roof collapse; coal seam temperature, used to monitor the spontaneous combustion trend of coal and identify potential spontaneous combustion hazards; and water inflow rate, used to understand the underground water situation and assess the degree of water hazard threat.

[0054] The collected raw data is transmitted to the data processing center via wireless or wired transmission. Due to potential sensor errors, signal interference, or transient malfunctions, the raw data may contain outliers that significantly deviate from the normal range, necessitating outlier detection. The system employs statistical methods to identify data points exceeding reasonable fluctuation ranges and marks or replaces them to ensure data reliability.

[0055] Then, a standardization transformation is performed to convert parameter data with different physical dimensions and numerical ranges into comparable standardized scores. For example, raw values ​​such as gas concentration and roof displacement are mapped to a scoring range of 0-10 points, giving each indicator a unified measurement standard and facilitating subsequent comprehensive evaluation.

[0056] Because different sensors may have different acquisition frequencies, the data may be out of sync in time, necessitating time-series data alignment. The system interpolates or samples the data of each parameter using a unified time base (such as every 15 seconds or every 30 seconds) to ensure that the data of each indicator are complete and corresponding at the same point in time, providing a time-consistent data foundation for subsequent dynamic weight adjustments and comprehensive scoring.

[0057] In some embodiments of this application, to ensure that the comprehensive score accurately reflects the most prominent security threats, a weight vector matching the current downhole environment is generated through a dynamic adjustment mechanism based on a predefined multi-dimensional indicator system and the real-time collected multi-dimensional parameter data. Specifically: Based on the aforementioned multi-dimensional indicator system, the initial weights of each evaluation indicator are determined using the analytic hierarchy process (AHP) to form an initial weight vector. Continuously monitor the changes in the real-time collected multidimensional parameter data; When the data fluctuation of a specific indicator exceeds the preset fluctuation threshold, the dynamic adjustment model is activated to correct the initial weight vector.

[0058] As described above, the system pre-establishes a multi-dimensional indicator system, which divides the main types of coal mine hazards into several assessment dimensions, such as gas hazards, roof falls, and water hazards. Each dimension contains one or more specific assessment indicators. Based on this indicator system, the analytic hierarchy process (AHP) is used to compare the importance of each indicator pairwise. By constructing a judgment matrix and performing a consistency check, the relative importance of each indicator in the comprehensive assessment is finally determined, forming an initial weight vector. For example, the gas hazard indicator may be assigned a higher initial weight, while the roof fall and water hazard indicators are assigned corresponding weights based on historical data and geological conditions of the mining area. The sum of the weights of all indicators is 1.

[0059] During operation, the system continuously receives and analyzes preprocessed multidimensional parameter data, monitoring changes in each indicator in real time. The monitoring includes not only the current value but also the trend and amplitude of these changes. When a parameter value of a particular indicator changes significantly within a short period, exceeding a preset fluctuation threshold, the system determines that the risk represented by that indicator is escalating and requires greater attention in the comprehensive assessment.

[0060] At this point, the system initiates a dynamic adjustment model to correct the initial weight vector. Specifically, the system increases the weight of the indicator whose fluctuation exceeds the limit, while proportionally reducing the weights of other indicators to ensure that the sum of the adjusted weight vectors remains 1. For example, when the gas concentration rises by more than 5% within one minute, the system automatically increases the weight of the gas hazard indicator and correspondingly reduces the weights of the roof or water hazard indicators, generating a new weight vector that matches the current downhole risk characteristics. This dynamic adjustment mechanism enables the assessment model to adapt to changes in the downhole environment, ensuring that the comprehensive score accurately reflects the most prominent safety threats.

[0061] In some embodiments of this application, to ensure that the risk assessment results are consistent with the actual downhole safety conditions, the preset fluctuation threshold is 5% / minute.

[0062] As mentioned above, the system sets a fluctuation threshold of 5% per minute to determine whether the change in a certain disaster indicator parameter reaches a level of significance that requires adjustment of the assessment weight. This threshold indicates that the relative rate of change of a monitored parameter exceeds 5% of its baseline value within a unit of time. For example, if the gas concentration rises from 0.6% to 0.63% or more within one minute, i.e., the change reaches or exceeds 5%, the system determines that the data fluctuation of that indicator exceeds the preset threshold. Similarly, if the roof displacement increases by more than 5% of its current value within one minute, or the water inflow increases by more than 5% within one minute, it is also considered to have reached the fluctuation threshold.

[0063] The threshold setting comprehensively considers the development speed of various disasters in coal mines, sensor measurement accuracy, and the response time of safety warnings. A threshold of 5% / minute effectively identifies abnormally rapid changes in parameters, avoiding unnecessary weight adjustments triggered by normal fluctuations or measurement noise, while also promptly capturing the accelerating development trend of potential risks. When the system detects that the rate of change of any indicator exceeds this threshold consecutively or multiple times, it is considered that the disaster risk represented by that indicator is significantly increasing, requiring its influence to be increased in the comprehensive assessment. This triggers a dynamic adjustment model to correct the weight vector, ensuring that the risk assessment results are consistent with the actual underground safety situation.

[0064] In some embodiments of this application, in order to accurately and comprehensively characterize the real-time safety status of coal mines and provide a quantitative basis for safety management, the standardized scores of each indicator are obtained, and then weighted and calculated with the weight vector to generate a comprehensive real-time risk assessment score, specifically as follows: The standardized scores of each indicator are weighted and calculated together with the weight vector to form a real-time comprehensive score ranging from 0 to 100 points, where a higher score indicates a higher level of disaster risk.

[0065] As described above, after data preprocessing and weight vector generation, the system performs a weighted calculation by summing the standardized scores of each disaster indicator with their corresponding weight values ​​to generate a comprehensive score that fully reflects the current safety status of the coal mine. The standardized scores of each indicator are obtained by converting the preprocessed parameter data into a unified score within the range of 0-10 according to preset rules, ensuring the comparability of parameters with different dimensions. The weight vector is dynamically generated in step S102, containing the relative importance of each indicator in the current environment.

[0066] The system multiplies the standardized score of each indicator by its corresponding weight, then sums all the products to obtain a weighted sum. This weighted sum reflects the overall risk contribution of each indicator under the current weight allocation. Subsequently, the system converts this weighted sum into a real-time comprehensive score of 0 to 100 using a linear or non-linear mapping method. For example, multiplying the weighted sum by 10 expands its range from 0-10 to 0-100.

[0067] The final real-time comprehensive score is updated every 30 seconds to 1 minute, continuously reflecting the dynamic changes in underground risks. The score results are output in numerical form, and risk levels are categorized according to the score range: 0-30 points indicate low risk, 30-60 points indicate medium risk, 60-80 points indicate high risk, and 80-100 points indicate extremely high risk. A higher score indicates a higher overall disaster risk facing the coal mine. This comprehensive score integrates multi-source parameter information and dynamic weighting, enabling a more accurate and comprehensive characterization of the real-time safety status of the coal mine and providing a quantitative basis for safety management.

[0068] In some embodiments of this application, in order to achieve timely and intuitive communication of risk information and provide effective support for emergency decision-making and the implementation of safety control measures, the real-time risk assessment score is output, and an early warning signal is automatically triggered when the score exceeds a preset safety threshold. Specifically: The real-time risk assessment score and the scores of each indicator are displayed in the form of visual charts; The warning signals include audible and visual alarms and the display of a red warning icon in a visual interface.

[0069] As described above, the system outputs and displays the generated real-time risk assessment score and the standardized scores of each individual indicator through a visual interface. The current overall score and the scores of each indicator are presented intuitively in chart form on the main control screen of the coal mine monitoring center, the dispatch terminal, and the mobile devices of relevant management personnel. Visualization formats include numerical displays, progress bars, dashboards, trend curves, and bar charts, enabling users to quickly grasp the overall risk level and the contribution of each disaster factor.

[0070] When the real-time comprehensive score exceeds the preset safety threshold, the system automatically triggers an early warning signal. The early warning signal includes two main methods: audible and visual alarms and interface prompts. Audible and visual alarm devices installed in the monitoring center and key underground areas will immediately activate, emitting a specific frequency beep or voice prompt, accompanied by flashing warning lights, alerting on-site personnel to the risk. Simultaneously, in the visual interface, the comprehensive score area automatically turns red and displays prominent red warning labels such as "High Risk" or "Danger," with relevant indicators also highlighted to facilitate quick identification of the risk source by operators.

[0071] The triggering of early warning signals is automatic, requiring no manual intervention, ensuring an immediate response when risks reach a critical state. The system supports setting multiple safety thresholds, corresponding to different levels of early warning response. For example, a yellow warning is activated when the score exceeds 60 points, and a red warning is activated when it exceeds 80 points, achieving tiered early warning management. Through this approach, the system achieves timely and intuitive communication of risk information, providing effective support for emergency decision-making and the implementation of safety control measures.

[0072] In some embodiments of this application, in order to comprehensively and systematically identify and quantify the risks of different disaster types, providing a foundation for dynamic weight adjustment and comprehensive scoring, the multi-dimensional indicator system includes assessment indicators related to gas disasters, roof disasters, and water hazards.

[0073] As mentioned above, the multi-dimensional indicator system is a structured framework for comprehensively assessing coal mine disaster risks. This system divides major disaster types into several independent but comprehensive assessment dimensions. Among them, the gas disaster dimension includes assessment indicators related to gas accumulation, outburst, and explosion risks, such as gas concentration, gas concentration change rate, gas emission rate, and ventilation velocity, used to assess the degree of danger of combustible gases underground. The roof disaster dimension includes assessment indicators reflecting the stability of the surrounding rock in roadways and working faces, such as roof displacement, roof delamination value, anchor bolt stress, and surrounding rock pressure, used to determine the risk of roof collapse or fall. The water hazard dimension includes assessment indicators related to groundwater activity and water inrush risks, such as water inflow rate, water pressure, water temperature, borehole water output, and aquifer water level changes, used to monitor and warn of water inrush accidents.

[0074] The assessment indicators for each dimension are based on real-time data collected by sensors and incorporated into the overall assessment process. This multi-dimensional system not only covers common major disaster types in coal mines but also supports the integration of assessment results from various dimensions through a unified weight vector to form a comprehensive risk score. The system architecture is scalable, allowing for the addition or adjustment of relevant indicators based on the specific geological conditions and mining processes of the mine, ensuring the applicability and accuracy of the assessment model. Through this indicator system, the system can comprehensively and systematically identify and quantify the risks of different disaster types, providing fundamental support for dynamic weight adjustment and comprehensive scoring.

[0075] Compared with existing technologies, this application discloses a real-time scoring method for coal mine disaster assessment based on weighted vectors. This method collects multi-dimensional parameter data characterizing disaster risk in real time using multiple sensors deployed underground in the coal mine. The raw data is preprocessed to ensure the quality and consistency of the assessment data, effectively solving the problem that multi-source heterogeneous data is difficult to directly use for comprehensive assessment, thus laying a data foundation for subsequent accurate assessment. Based on a predefined multi-dimensional indicator system and combined with the real-time collected multi-dimensional parameter data, a weighted vector matching the current underground environment is generated through a dynamic adjustment mechanism. This enables the assessment model to dynamically adjust the weights of each indicator according to actual risk changes, significantly improving the accuracy and timeliness of the assessment results. After standardizing the real-time collected multi-dimensional parameter data, a weighted calculation is performed with the dynamically generated weighted vector to generate a comprehensive real-time risk assessment score. This score can comprehensively and objectively reflect the current overall safety status of the coal mine, providing a scientific basis for safety management decisions. By outputting a real-time risk assessment score and automatically triggering an early warning signal when the score exceeds a preset safety threshold, a seamless connection from risk assessment to early warning response is achieved. This changes the limitations of traditional single-parameter alarms, enabling the early detection of potential complex risks, effectively reducing false alarm and missed alarm rates, and improving the reliability and effectiveness of coal mine safety early warning.

[0076] Based on the same inventive concept as the methods described above, this application also proposes a real-time scoring system for coal mine disaster assessment based on weight vectors, such as... Figure 2 The diagram shown is a schematic representation of a real-time scoring system for coal mine disaster assessment based on weight vectors. The system includes: The data acquisition module is used to collect multi-dimensional parameter data characterizing disaster risk in real time through multiple sensors deployed underground in coal mines; The weight generation module is used to generate a weight vector that matches the current downhole environment based on a predefined multi-dimensional indicator system and the real-time collected multi-dimensional parameter data through a dynamic adjustment mechanism. The scoring calculation module is used to standardize the real-time collected multi-dimensional parameter data to obtain the standardized score of each indicator, and to perform weighted calculation with the weight vector to generate a comprehensive real-time risk assessment score. The output warning module is used to output the real-time risk assessment score and automatically trigger a warning signal when the score exceeds a preset safety threshold.

[0077] Preferably, the data acquisition module specifically comprises: The sensor's data acquisition interval is 10 to 30 seconds; The multidimensional parameter data includes gas concentration, roof displacement, coal seam temperature, and water inflow rate; The collected raw data is preprocessed, including outlier detection, standardization transformation, and time series data alignment.

[0078] Preferably, the weight generation module specifically comprises: Based on the aforementioned multi-dimensional indicator system, the initial weights of each evaluation indicator are determined using the analytic hierarchy process (AHP) to form an initial weight vector. Continuously monitor the changes in the real-time collected multidimensional parameter data; When the data fluctuation of a specific indicator exceeds the preset fluctuation threshold, the dynamic adjustment model is activated to correct the initial weight vector.

[0079] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program goods. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program goods embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0080] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program goods according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0081] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0082] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A real-time scoring method for coal mine disaster assessment based on weight vectors, characterized in that, include: Multiple sensors deployed underground in coal mines are used to collect multidimensional parameter data characterizing disaster risk in real time and perform preprocessing. Based on a predefined multi-dimensional index system, combined with the real-time collected multi-dimensional parameter data, a weight vector matching the current downhole environment is generated through a dynamic adjustment mechanism. The real-time collected multidimensional parameter data is standardized to obtain the standardized score of each indicator, and then weighted with the weight vector to generate a comprehensive real-time risk assessment score. The system outputs the real-time risk assessment score and automatically triggers an early warning signal when the score exceeds a preset safety threshold.

2. The method as described in claim 1, characterized in that, The real-time acquisition and preprocessing of multidimensional parameter data characterizing disaster risk includes: The sensor's data acquisition interval is 10 to 30 seconds; The multidimensional parameter data includes gas concentration, roof displacement, coal seam temperature, and water inflow rate; The collected raw data is preprocessed, including outlier detection, standardization transformation, and time series data alignment.

3. The method as described in claim 1, characterized in that, The predefined multi-dimensional indicator system, combined with the real-time collected multi-dimensional parameter data, generates a weight vector that matches the current downhole environment through a dynamic adjustment mechanism, specifically: Based on the aforementioned multi-dimensional indicator system, the initial weights of each evaluation indicator are determined using the analytic hierarchy process (AHP) to form an initial weight vector. Continuously monitor the changes in the real-time collected multidimensional parameter data; When the data fluctuation of a specific indicator exceeds the preset fluctuation threshold, the dynamic adjustment model is activated to correct the initial weight vector.

4. The method as described in claim 3, characterized in that, The preset fluctuation threshold is 5% / minute.

5. The method as described in claim 1, characterized in that, The standardized scores of each indicator are obtained, and then weighted by the weight vector to generate a comprehensive real-time risk assessment score, specifically: The standardized scores of each indicator are weighted and calculated together with the weight vector to form a real-time comprehensive score ranging from 0 to 100 points, where a higher score indicates a higher level of disaster risk.

6. The method as described in claim 1, characterized in that, The real-time risk assessment score is output, and an early warning signal is automatically triggered when the score exceeds a preset safety threshold. Specifically: The real-time risk assessment score and the scores of each indicator are displayed in the form of visual charts; The warning signals include audible and visual alarms and the display of a red warning icon in a visual interface.

7. The method as described in claim 3, characterized in that, The multi-dimensional indicator system includes assessment indicators related to gas disasters, roof disasters, and water hazards.

8. A real-time scoring system for coal mine disaster assessment based on weight vectors, characterized in that, include: The data acquisition module is used to collect multi-dimensional parameter data characterizing disaster risk in real time through multiple sensors deployed underground in coal mines; The weight generation module is used to generate a weight vector that matches the current downhole environment based on a predefined multi-dimensional indicator system and the real-time collected multi-dimensional parameter data through a dynamic adjustment mechanism. The scoring calculation module is used to standardize the real-time collected multi-dimensional parameter data to obtain the standardized score of each indicator, and to perform weighted calculation with the weight vector to generate a comprehensive real-time risk assessment score. The output warning module is used to output the real-time risk assessment score and automatically trigger a warning signal when the score exceeds a preset safety threshold.

9. The system as described in claim 8, characterized in that, The data acquisition module is specifically: The sensor's data acquisition interval is 10 to 30 seconds; The multidimensional parameter data includes gas concentration, roof displacement, coal seam temperature, and water inflow rate; The collected raw data is preprocessed, including outlier detection, standardization transformation, and time series data alignment.

10. The system as described in claim 8, characterized in that, The weight generation module specifically comprises: Based on the aforementioned multi-dimensional indicator system, the initial weights of each evaluation indicator are determined using the analytic hierarchy process (AHP) to form an initial weight vector. Continuously monitor the changes in the real-time collected multidimensional parameter data; When the data fluctuation of a specific indicator exceeds the preset fluctuation threshold, the dynamic adjustment model is activated to correct the initial weight vector.

Images