Coal mine underground safety monitoring data comprehensive processing method

Abstract

The invention discloses a coal mine underground safety monitoring data comprehensive processing method, and relates to the technical field of coal mine underground data processing, and the method comprises the following steps: setting an initial vibration frequency for a coal mine underground conveying system, carrying out the vibration processing of conveyed coal through a natural vibration amplitude, and building a stable coal transportation vibration working condition. According to the method, the gas monitoring data is deeply analyzed, the trend feature vector is extracted, and the pre-trained support vector machine model is utilized to predict the gas change trend and the gathering risk in the coal conveying process in real time. When abnormal gas gathering is detected, the vibration frequency is automatically regulated and controlled, coal disturbance is reduced, gas precipitation is inhibited, and concentration rise is prevented. Regulation and control are intelligently executed on the basis of risk prediction and combined with real-time feedback optimization, a closed-loop control strategy is formed, accurate gas management and control are achieved, the explosion risk is reduced, and intelligent guarantee is provided for coal mine safety production.

Description

Technical Field

[0001] The present invention relates to the technical field of underground coal mine data processing, and particularly to a comprehensive processing method for underground coal mine safety monitoring data. Background Art

[0002] The comprehensive processing of underground coal mine safety monitoring data refers to the centralized collection, unified storage, real-time analysis and systematic processing of multi-source heterogeneous data (such as gas concentration, oxygen content, temperature and humidity, carbon monoxide content, wind speed and air volume, mine pressure, and equipment operation status, etc.) collected by various safety monitoring sensor devices in the underground coal mine. By using data fusion technology, intelligent algorithms and information analysis means, abnormal conditions or potential safety hazards can be detected in a timely manner, accurate early warning and alarm information can be given, and visual monitoring reports and safety assessment indicators can be generated to assist coal mine production management personnel in making scientific decisions and ensuring the safety of miners and the stable operation of coal mine production.

[0003] The role of coal transportation safety monitoring is to monitor, analyze and give early warning of potential safety hazards that may occur during the coal transportation process in real time, ensure the stable operation of the transportation system, and guarantee the safety of coal mine production. Its core functions include gas monitoring (preventing large amounts of gas from being released and accumulating, leading to explosion), coal temperature monitoring (preventing coal from spontaneous combustion due to overheating caused by oxidation or friction), dust concentration detection (avoiding the risk of coal dust explosion), monitoring of the state of transportation equipment (detecting faults such as belt wear, motor overheating, and roller slippage), coal flow accumulation and ventilation regulation (optimizing the coal flow transportation method to prevent safety problems caused by blockage and air stagnation), etc. Through an intelligent monitoring system and an automatic early warning mechanism, coal transportation safety monitoring can detect abnormalities in a timely manner and take corresponding control measures, reduce the probability of accidents, improve the safety and production efficiency of the coal mine transportation system, thereby reducing economic losses and the risk of casualties.

[0004] The existing technologies have the following deficiencies: In the existing technologies, the mined coal is usually transported through a transportation system (such as a conveyor belt, a belt transfer machine, etc.). To prevent coal from piling up, spilling and improve the stacking efficiency, the transportation system usually applies a certain vibration. However, since the coal itself contains a certain amount of gas (mainly methane), during the transportation process, affected by the vibration, the gas may be released in large amounts and quickly accumulate in local areas. However, it is often difficult for the existing technologies to accurately identify and respond to this hidden danger in real time. When the gas concentration in the air reaches 5% - 15% and encounters a fire source (such as conveyor belt friction, electric spark, etc.), it is extremely easy to cause a violent explosion. Once the gas concentration in the transportation channel reaches the explosion limit and is ignited, it may lead to a chain explosion, releasing high-temperature and high-pressure shock waves, which will not only destroy the coal mine transportation system, but also pose a serious threat to underground workers, and even trigger a large-scale mine disaster.

[0005] The above information disclosed in the background section is only used to enhance the understanding of the background of the present disclosure. Therefore, it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0006] The object of the present invention is to provide a comprehensive processing method for coal mine underground safety monitoring data. By deeply analyzing gas monitoring data, characteristic vectors with trend and predictability are generated, and a pre-trained support vector machine model is used to predict the change trend of gas and potential accumulation risks during coal transportation in real time. When the prediction model detects abnormal gas concentration accumulation, a dynamic vibration frequency regulation mechanism is automatically activated to reduce the actual vibration frequency of the conveying system, thereby reducing mechanical disturbance of the coal body, inhibiting the gas evolution rate, and preventing the gas concentration from further increasing, so as to solve the problems in the above background technology.

[0007] To achieve the above object, the present invention provides the following technical solution: A comprehensive processing method for coal mine underground safety monitoring data, including the following steps: Set an initial vibration frequency for the conveying system in the coal mine underground, vibrate the conveyed coal with the inherent vibration amplitude, and establish a stable coal transportation vibration working condition; Real-time monitor and record the gas data evolved due to vibration during coal transportation through high-precision gas sensors arranged at key positions of the conveying equipment; Transmit and store the gas concentration data collected by the gas sensors to the cloud server in real time, store it in chronological order, and automatically generate a historical data analysis set of gas concentration. Extract gas accumulation characteristic indicators from the analysis set, construct a characteristic vector of gas change, and input the characteristic vector into the pre-trained support vector machine model to predict the change trend of gas and potential accumulation risks during coal transportation in real time; When the prediction model identifies the existence of gas accumulation risk, activate the frequency regulation mechanism, and automatically reduce the actual vibration frequency of the conveying system according to the degree of accumulation risk predicted by the model, reduce the intensity of coal body disturbance, reduce the gas evolution rate, and relieve the situation of a large amount of gas aggregation; Conduct real-time feedback evaluation on the effect after regulation to form a closed-loop control.

[0008] Preferably, the gas sensors are composed of a sensing network, and are arranged in a multi-point layout and multi-angle distribution to cover the key areas of the conveying system.

[0009] Preferably, gas accumulation characteristic indicators are extracted from the analysis set. The extracted characteristic indicators include the cumulative duration of the gas concentration continuously being higher than the safety threshold and the cumulative increment of the gas concentration relative to the baseline. Under the monitoring window, after deeply analyzing the extracted characteristics, a cross-threshold duration indicator and a short-term cumulative increment indicator are respectively generated. A characteristic vector of gas change is constructed through the cross-threshold duration indicator and the short-term cumulative increment indicator to quantify the cumulative effect and overlimit persistence of gas within the monitoring window, evaluate the severity of the gas concentration exceeding the safety threshold and its enrichment trend in a short period of time, and identify the evolution process of the gas explosion risk.

[0010] Preferably, the cross-threshold duration indicator and the short-term cumulative increment indicator are input into a pre-trained support vector machine model, and an instantaneous enrichment reference value is output through the support vector machine model. Based on the instantaneous enrichment reference value, the change trend and potential accumulation risk of gas during the coal conveying process are predicted in real time.

[0011] Preferably, the instantaneous enrichment reference value generated when the pre-trained support vector machine model predicts the change trend of gas during the coal conveying process is compared and analyzed with the preset instantaneous enrichment reference threshold to identify the gas accumulation risk during the coal conveying process. The specific identification process is as follows: If the instantaneous enrichment reference value is greater than the instantaneous enrichment reference threshold, the gas evolution situation during the coal conveying process under this monitoring window is classified as gas accumulation; if the instantaneous enrichment reference value is less than or equal to the instantaneous enrichment reference threshold, the gas evolution situation during the coal conveying process under this monitoring window is classified as normal evolution.

[0012] Preferably, when the prediction model identifies the existence of gas accumulation risk, a frequency regulation mechanism is activated, and the actual vibration frequency of the conveying system is automatically reduced according to the degree of accumulation risk predicted by the model. The specific steps are as follows: When the prediction model identifies the existence of gas accumulation risk, the vibration frequency adjustment amount is dynamically calculated according to the ratio of the instantaneous enrichment reference value to the instantaneous enrichment reference threshold. The calculation expression is: , Where: is the vibration frequency adjustment amount, and a negative value indicates the reduction amplitude of the vibration frequency. is the initial vibration frequency, that is, the standard vibration frequency of the coal conveying system when there is no gas risk. is the instantaneous enrichment reference value, which is used to measure the current gas enrichment degree. is the instantaneous enrichment reference threshold, which is the critical value of the gas enrichment risk. is the adjustment coefficient, which is used to control the sensitivity of the vibration frequency adjustment. is the non-linear weight factor, which is used to strengthen the adjustment amplitude in high-risk situations.

[0013] Preferably, a feedback evaluation mechanism is introduced. Within the adjusted monitoring window, the instantaneous enrichment reference value is recalculated and compared with that before adjustment to form a dynamic closed-loop optimization control. The expression for updating the vibration frequency through the dynamic closed-loop optimization control is: , where: is the actual vibration frequency after adjustment, is the minimum allowable vibration frequency, is the vibration frequency of the previous monitoring window, is the recalculated instantaneous enrichment reference value after adjustment, used to evaluate the adjustment effect, is the feedback adjustment factor, used to control the response rate of the feedback adjustment.

[0014] Preferably, under the monitoring window, the specific steps for generating the cross-threshold duration index by deeply analyzing the cumulative duration of the gas concentration continuously higher than the safety threshold are as follows: Within the monitoring window, the gas concentration is monitored in real time, and a safety threshold is set. When the gas concentration exceeds the safety threshold, the cumulative value of the duration of the over-threshold state is recorded. The statistical expression for the cumulative value of the duration of the over-threshold state is: , where: is the cumulative value of the duration of the over-threshold state, indicating the total time that the gas is in a high-risk state within the monitoring window, is the Heaviside step function, used to determine whether the gas concentration exceeds the threshold, defined as: , where, is the gas concentration at time t, is the set safety threshold of the gas, represents the time range of the monitoring window, used to count the gas over-threshold situation within the monitoring window. Among them, and represent the start point and end point of the monitoring window respectively; Based on the cumulative value of the duration of the over-threshold state, considering the exponential amplification effect of the over-threshold duration, a cross-threshold duration index is constructed, and the risk characterization ability is enhanced through the exponential transformation in the time dimension. The construction expression of the cross-threshold duration index is: , where: is the cross-threshold duration index, used to quantify the cumulative effect of gas over-threshold, is the adjustment coefficient, used to control the amplitude of exponential growth, is the exponential growth factor, used to adjust the influence of the over-threshold time on the final cross-threshold duration index.

[0015] Preferably, under the monitoring window, the specific steps for generating the short-term cumulative increment index after deeply analyzing the cumulative increment of the gas concentration relative to the baseline are as follows: Calculate the short-term cumulative amount above the baseline, considering not only the degree above the baseline but also the cumulative impact of the state above the baseline. The calculation expression is: , where: is the short-term cumulative amount above the baseline, used to measure the total accumulation degree of the part of the gas above the baseline within the monitoring window, is the gas concentration at the i th sampling point within the monitoring window, is the set baseline gas concentration, is the weight factor, which assigns different influence weights to different time points within the monitoring window to highlight high-risk areas, is the non-linear adjustment parameter; Introduce a mapping function to convert the short-term cumulative amount above the baseline into a short-term cumulative increment index, directly quantifying the severity of the gas above the baseline. The specific conversion expression is: , where: is the short-term cumulative increment index, used to quantify the persistence of the gas above the baseline within the monitoring window and provide a trend assessment, M is the reference limit value, used for normalization, is the logarithmic mapping coefficient, used to control the sensitivity at low cumulative amounts above the baseline, is the non-linear normalization coefficient, controlling the non-linear amplification degree of the short-term cumulative increment index, is the cumulative amount above the baseline trend weighting coefficient, controlling the impact of the cumulative amount above the baseline on the final short-term cumulative increment index.

[0016] In the above technical solution, the technical effects and advantages provided by the present invention are: The present invention deeply analyzes gas monitoring data, generates characteristic vectors with trend and predictability, and uses a pre-trained support vector machine model to predict the change trend and potential accumulation risk of gas during coal transportation in real time. When the prediction model detects abnormal accumulation of gas concentration, it automatically activates a dynamic vibration frequency regulation mechanism to reduce the actual vibration frequency of the transportation system, thereby reducing mechanical disturbance of the coal body, inhibiting the gas evolution rate, and preventing further increase in gas concentration. The adjustment of the vibration frequency is intelligently executed based on the predicted risk level, enabling the transportation condition to be dynamically matched with the gas evolution condition. Finally, through real-time feedback to evaluate the regulation effect, a closed-loop control strategy is formed to achieve precise control of gas during coal transportation, greatly reducing the risk of gas explosion and providing a more efficient, intelligent, and refined safety guarantee for coal mine safety production. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0018] Figure 1 It is a method flow chart of a comprehensive processing method for safety monitoring data in a coal mine underground of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] Now, the exemplary embodiments will be described more comprehensively with reference to the drawings. However, the exemplary embodiments can be implemented in various forms and should not be construed as limited to the examples set forth herein; rather, these examples are provided so that the present disclosure will be more complete and comprehensive, and will fully convey the concept of the exemplary embodiments to those skilled in the art.

[0020] The present invention provides a comprehensive processing method for safety monitoring data in a coal mine underground as shown in Figure 1 the following, including the following steps: Set an initial vibration frequency for the transportation system in the coal mine underground (including transportation equipment such as conveyor belts and belt transfer machines), vibrate the transported coal with an inherent vibration amplitude, establish a stable coal transportation vibration condition, and enable the coal to be transported evenly and orderly; The selection of the initial vibration frequency is mainly based on the comprehensive consideration of multiple factors such as conveying efficiency, safety, and equipment reliability during the coal transportation process. The specific requirements include the following aspects: First, it is necessary to ensure that the vibration frequency is moderate, which can effectively prevent coal from accumulating, sticking, or blocking on the conveyor belt surface, and ensure that the coal can move smoothly and spread evenly. Second, the selected frequency should ensure that no violent vibration is generated to avoid excessive fragmentation of coal particles during transportation, which may produce a large amount of dust and increase the explosion risk. Third, the setting of the vibration frequency should keep the conveying equipment itself within a reasonable operating condition range to avoid resonance or mechanical fatigue damage to the conveyor belt, support structure, and driving device. In addition, the initial frequency should consider the properties of the transported coal (such as physical parameters like coal particle size, humidity, viscosity, etc.) to achieve the best conveying efficiency and safety performance. Therefore, considering the above multiple factors, the initial vibration frequency is usually determined through experimental verification based on factors such as the actual coal mine transportation equipment specifications, coal types, and on-site environment.

[0021] The function of this step is to establish a stable vibration condition for coal transportation, enabling the coal to be conveyed evenly and orderly, and at the same time providing a stable basic state for the subsequent gas emission situation.

[0022] The high-precision gas sensors set at key positions of the conveying equipment (such as the conveyor belt surface, transfer points, upper part of the coal bunker, and conveying space, etc.) are used to monitor and record the gas data (mainly methane concentration values) emitted due to vibration during coal transportation in real time. To improve the monitoring accuracy and reliability, these gas sensors can form a sensing network and adopt a multi-point layout and multi-angle distribution to cover the key areas of the conveying system. The function of this step is to achieve accurate and real-time acquisition and monitoring of the gas concentration during the conveying process, thus providing rich data support for the subsequent analysis, modeling, and early warning of gas risks.

[0023] Forming a sensing network with gas sensors and adopting a multi-point layout and multi-angle distribution to cover the key areas of the conveying system means reasonably configuring multiple gas sensors according to specific spatial positions, different heights, and different angles to form a tight sensing network in the conveying system and the surrounding space. Specifically, this layout method can include the distribution of sensors along the length, width, and vertical height directions of the conveyor belt. For example, gas sensors are deployed at key positions such as the head and tail of the conveying system, transfer points, feeding points, top and bottom of the coal bunker, side walls of the conveying channel, and corner areas where gas accumulation or leakage is likely to occur. Thus, it can comprehensively capture the gas emission situation in different regions and positions during coal transportation, identify the dynamic change trend of local gas accumulation in real time, eliminate monitoring blind spots, and maximize the accuracy and real-time performance of gas risk monitoring, providing reliable data support for subsequent gas safety early warning.

[0024] The gas concentration data collected by the gas sensor is transmitted and stored in the cloud server in real time in the form of time series, and a gas concentration historical data analysis set is automatically generated. The gas concentration characteristic indicators are extracted from the analysis set, and the characteristic vector of gas change is constructed. The characteristic vector is input into the pre-trained support vector machine model to predict the change trend of gas and potential accumulation risks in the coal transportation process in real time. The main function of storing and automatically generating gas concentration historical data analysis sets in time series is to provide continuous, trending and predictive data support, thereby improving the accuracy and intelligence of gas safety monitoring. Specifically, time series storage can record the changes in gas concentration at different time points, not only tracking the instantaneous concentration of gas in real time, but also analyzing its long-term trend and periodic fluctuations, thereby identifying the law of gas precipitation during the transportation process.

[0025] Gas concentration characteristic indicators are extracted from the analysis set, where the extracted characteristic indicators include the cumulative duration that the gas concentration is continuously higher than the safety threshold and the cumulative increment of the gas concentration relative to the baseline. Under the monitoring window, after in-depth analysis of the extracted features, the cross-threshold duration indicator and the short-time cumulative increment indicator are generated respectively. The characteristic vector of gas change is constructed through the cross-threshold duration indicator and the short-time cumulative increment indicator, and the cumulative effect and over-limit persistence of gas within the monitoring window are quantified to evaluate the severity of the gas concentration exceeding the safety threshold and its enrichment trend in a short time, and to identify the evolution process of the gas explosion risk.

[0026] If the gas concentration remains above the safety threshold for a long period of time during coal transportation, it means that the gas concentration underground or in the transportation channel is in a dangerous range for a long time and is difficult to be diluted or discharged in time, which greatly increases the risk of gas accumulation in the coal transportation environment. Continuous high concentrations not only increase the possibility of gas explosion accidents, but also easily cause poisoning or hypoxia and suffocation of employees. At the same time, long-term dangerous concentrations also indicate that ventilation measures or transportation strategies have failed to effectively diffuse or discharge gas. If not intervened in time, the degree of gas accumulation will further deepen, which can easily trigger the critical point of explosion and lead to serious safety accidents.

[0027] When transporting coal underground in a coal mine through a conveying system, if the cumulative duration during which the gas concentration is continuously higher than the safety threshold is relatively long during the conveying process, this indeed means that the potential risk of gas accumulation during coal transportation is greater. Specifically, when the gas concentration exceeds the safety threshold (such as the commonly set safety standard value of 1% or the lower explosive limit of 5%) and remains at a high level for a long time, it indicates that the gas cannot be diffused and discharged in a timely and effective manner, but accumulates and gathers continuously in a local space. This phenomenon directly leads to a further increase in the gas concentration and gradually approaches or even enters the dangerous range of gas explosion (5% - 15%). As the residence time of high-concentration gas becomes longer, the probability of encountering a fire source (such as conveyor belt friction, electrical equipment sparks, etc.) will also increase significantly, thereby greatly increasing the likelihood of gas explosion accidents. Therefore, the cumulative duration of gas concentration exceeding the limit can be used as a key indicator to timely identify potential risks during coal transportation and prevent serious safety accidents from occurring.

[0028] The specific steps for generating the cross-threshold duration index after deeply analyzing the cumulative duration during which the gas concentration is continuously higher than the safety threshold under the monitoring window are as follows: Within the monitoring window, the gas concentration is monitored in real time, and a safety threshold is set. When the gas concentration exceeds this safety threshold, the cumulative value of the duration of the over-threshold state is recorded, providing a basis for the calculation of the cross-threshold duration index. The statistical expression for the cumulative value of the duration of the over-threshold state is: , Where: is the cumulative value of the duration of the over-threshold state, representing the total time that the gas is in a high-risk state within the monitoring window. is the Heaviside step function, used to determine whether the gas concentration exceeds the threshold, defined as: , Where, is the gas concentration at time t. is the set safety threshold of the gas, usually set to 1% or 5%. represents the time range of the monitoring window, used to count the gas over-threshold situation within the monitoring window. Among them, and represent the start point and end point of the monitoring window respectively. This step calculates through integration to continuously track the cumulative duration of the gas over-threshold state within the monitoring window, providing basic data for the subsequent cross-threshold duration index processing. If the over-threshold duration is longer, it means that the gas aggregation situation is more serious, and corresponding control measures need to be taken.

[0029] Based on the first-step calculation, consider the exponential amplification effect of the over-threshold duration. That is, when the gas over-threshold time is short, the impact on safety is small; but when the over-threshold duration is long, the risk will increase exponentially. Therefore, a cross-threshold duration index is constructed to enhance the risk characterization ability through exponential transformation in the time dimension. The construction expression of the cross-threshold duration index is: , where: is the cross-threshold duration index, used to quantify the cumulative effect of gas over-threshold. The larger the value, the longer the gas stays in the over-limit state, and the higher the risk. is the adjustment coefficient, used to control the amplitude of exponential growth, usually taking values between 1 and 10. is the exponential growth factor, used to adjust the impact of the over-threshold time on the final cross-threshold duration index, often set to 0.01 - 0.05 to ensure that the impact of short-term over-threshold is small and the impact of long-term over-threshold is significant. e is the base of the natural logarithm, used to exponentially amplify the time effect; Through exponential transformation, this cross-threshold duration index can more sensitively capture the duration effect of gas in the over-threshold state, so that short-term over-threshold will not trigger excessive alarms, while long-term over-threshold will lead to an exponential surge, thus providing a more targeted risk assessment standard. When the cumulative value of the over-threshold state duration is very small, is close to 1, and the exponential change is gentle, indicating that the short-term over-threshold state will not cause the system to overreact; when the cumulative value of the over-threshold state duration increases to a certain threshold, the exponential value rises sharply, quickly reflecting the severity of gas accumulation and providing a strong signal for the early warning system.

[0030] From the cross-threshold duration index, it can be seen that under the monitoring window, the larger the performance value of the cross-threshold duration index generated after in-depth analysis of the cumulative duration of the gas concentration continuously higher than the safety threshold, the longer the time that the gas concentration continuously exceeds the safety threshold during the coal transportation process, that is, the more obvious the cumulative effect of gas in the local space, which means the greater the potential gas accumulation risk, and the probability of safety hazards such as explosion, fire or asphyxiation also increases accordingly. On the contrary, if the performance value of the cross-threshold duration index is small, it means that the duration of the gas concentration over-threshold state is short, the gas can diffuse or be discharged quickly, and the gas accumulation trend during the coal transportation process is weak, and the safety risk is low.

[0031] When coal in the underground coal mine is transported through the conveying system, if the cumulative increment of the gas concentration relative to the baseline is high, it usually means that the total amount of gas released from the coal within a certain period of time has increased significantly, and it cannot be quickly diffused or discharged and gradually accumulates in local areas. At this time, the gas release rate significantly exceeds the average level under normal conveying conditions or the historical safety baseline level, indicating that the gas is in a state of rapid enrichment, and the potential aggregation risk has increased significantly. If this situation is not dealt with in a timely and effective manner, it is extremely easy to cause the local gas concentration to exceed the safety threshold (the explosion range of 5%-15%), and trigger a violent gas explosion after encountering ignition sources such as conveyor belt friction and electrical equipment sparks. Therefore, the higher the cumulative increment, the more serious the hidden danger of rapid gas accumulation during the conveying process.

[0032] The specific steps for generating the short-term cumulative increment index after in-depth analysis of the cumulative increment of the gas concentration relative to the baseline under the monitoring window are as follows: Within the monitoring window, the gas concentration may be higher than the set baseline value (i.e., the gas concentration in the historical safe and stable state) during some time periods. To accurately measure the degree of exceeding the baseline and its cumulative impact of the gas within this window, first calculate the short-term cumulative amount exceeding the baseline, that is, within the entire monitoring window, how the part of the gas concentration exceeding the baseline value accumulates over time to form a "risk pool". The short-term cumulative amount exceeding the baseline not only considers the degree of exceeding the baseline but also the cumulative impact of the state of exceeding the baseline, so as to identify persistent risks. The calculation expression is: , Where: is the short-term cumulative amount exceeding the baseline, which is used to measure the total accumulation degree of the part of the gas exceeding the baseline within the monitoring window, is the gas concentration at the i th sampling point within the monitoring window, is the set baseline gas concentration, usually set based on the gas concentration in the historical stable operating state, is the weight factor, which assigns different influence weights to different time points within the monitoring window to highlight high-risk areas (such as greater weights at gas sudden increase points), is the non-linear adjustment parameter, which is used to amplify the data points with a higher degree of exceeding the baseline, making the contribution of the period with a serious exceeding of the baseline more significant; The weight factor is adopted, so that the contributions of the points exceeding the baseline within different monitoring windows are different. For example, if the gas concentration suddenly rises and remains at a high level for a long time during a certain period, the weight of this area will be higher, enhancing the sensitivity of the algorithm to dangerous moments. This cumulative amount not only considers the degree of exceeding the baseline but also combines the cumulative impact, so as to more comprehensively characterize the overall trend of short-term gas exceeding the baseline.

[0033] The short-term super-baseline cumulative quantity only reflects the cumulative degree of gas concentration exceeding the baseline within the monitoring window, but fails to intuitively quantify the risk level or form a trend assessment. Therefore, a mapping function is introduced to convert the short-term super-baseline cumulative quantity into a short-term cumulative increment index, so as to directly quantify the severity of gas exceeding the baseline and be used for trend prediction and risk warning. The specific conversion expression is: , where: is the short-term cumulative increment index, which is used to quantify the persistence of gas exceeding the baseline within the monitoring window and provide a trend assessment. M is the reference limit value, which is used for normalization to prevent the short-term cumulative increment index from being too large. It can usually be set as the historical maximum risk cumulative quantity or an empirically set value. is the logarithmic mapping coefficient, which is used to control the sensitivity at low super-baseline cumulative quantities so that the low-risk interval can still be quantified. is the non-linear normalization coefficient, which controls the non-linear amplification degree of the short-term cumulative increment index, making the exponential value more distinguishable when severely exceeding the baseline. is the super-baseline trend weighting coefficient, which controls the influence of the super-baseline cumulative quantity on the final short-term cumulative increment index, causing the index to rise sharply when the gas concentration significantly exceeds the baseline. The logarithmic transformation is adopted so that even when the super-baseline cumulative quantity is small, the short-term cumulative increment index can still output reasonable values, avoiding the fuzzification of the short-term cumulative increment index in the low-risk interval. By the exponential transformation of the non-linear normalization term ( ), the exponential value in the high-risk interval is amplified, making the high gas accumulation risk more prominent and improving the resolution of the warning. The short-term cumulative increment index can sensitively reflect the overall trend of short-term gas exceeding the baseline and provide more fine-grained safety warning data support.

[0034] From the short-term cumulative increment index, it can be seen that under the monitoring window, the larger the performance value of the short-term cumulative increment index generated after in-depth analysis of the cumulative increment of gas concentration relative to the baseline, the higher the cumulative increment of gas concentration relative to the baseline within the current monitoring window, that is, the more significant the situation that gas continuously exceeds the historical stable level in a short time. At this time, the gas evolution rate is fast, and it cannot be diffused or discharged in time, resulting in a relatively high degree of local aggregation and obvious gas accumulation risk. The higher this risk, the larger the value of the short-term cumulative increment index, indicating a significant increase in the probability of gas explosion or safety accident. On the contrary, if the performance value of the short-term cumulative increment index is small, it means that the difference between the gas concentration and the baseline is not obvious or the duration is short, and no significant cumulative effect of gas exceeding the baseline is formed. Therefore, the gas accumulation risk is relatively low and the conveying environment is relatively safe. Therefore, the size of the performance value of the short-term cumulative increment index can effectively quantify and reflect the potential gas accumulation risk during coal conveying.

[0035] Input the cross-threshold duration index and the short-term cumulative increment index into the pre-trained support vector machine model, and output the instantaneous enrichment reference value through the support vector machine model. Based on the instantaneous enrichment reference value, the change trend and potential accumulation risk of gas during the coal conveying process are predicted in real time.

[0036] The pre-trained support vector machine model refers to comprehensively training and optimizing the support vector machine (SVM) model based on historical monitoring data and gas risk cases before the coal mine gas monitoring and safety analysis system is officially put into real-time operation. Specifically, first, it is necessary to use the historical data of the coal mine conveying process, especially relevant characteristic indexes such as the "cross-threshold duration index" and the "short-term cumulative increment index" as the model input, and combine the real annotations of actual gas abnormal events (such as rapid gas accumulation or even explosion accidents) as the model output target to establish a training data set. After fully cleaning and preprocessing the training data, use the support vector machine algorithm to construct the optimal decision boundary in the high-dimensional space, so that the model can accurately distinguish between the "normal conveying state" and the "gas abnormal enrichment state". During the training process, by repeatedly trying different model kernel functions (such as linear kernel, radial basis function kernel, Gaussian kernel), penalty coefficients (C value), kernel function parameters (such as Gaussian kernel width) and other key parameters, and combining cross-validation and grid search techniques, the best model structure and hyperparameter configuration are determined. After the above repeated training, verification and optimization, a support vector machine model with good generalization performance is finally obtained, that is, the "pre-trained support vector machine model" in this paper.

[0037] Furthermore, the significance of "pre-training completed" lies in that the support vector machine model already possesses powerful pattern recognition and trend prediction capabilities before the coal mine conveyor safety monitoring system is officially put into use. When real-time data of the "crossing threshold duration index" and the "short-term cumulative increment index" are continuously input into this model, the model can quickly compare and identify the current gas characteristics with the patterns learned in advance, and output the instantaneous enrichment reference value representing the degree of rapid gas enrichment and risk in real time. Since the support vector machine model has previously learned the internal relationship between the rapid increase in gas concentration and the coal conveyor environment and conveyor characteristics from a large amount of historical data and actual cases, it can respond quickly and accurately to new monitoring data during real-time operation, and highly reliable risk prediction results can be obtained without re-training. The advantage of this pre-training is also reflected in its operation efficiency. The calculation process of the support vector machine model itself is relatively simple and the amount of calculation is small. Especially after the model parameters and structure are determined in advance, the real-time prediction process can be completed within milliseconds, so it is very suitable for the coal mine safety conveyor scenario that requires rapid response and real-time monitoring and early warning.

[0038] In summary, the "pre-trained support vector machine model" means that it has mastered the typical feature patterns of gas accumulation from a sufficiently rich historical dataset, and can achieve rapid and accurate analysis and prediction of real-time data, effectively ensuring the timeliness and accuracy of gas risk management during the coal mine conveyor process.

[0039] The support vector machine model is not specifically limited here. Any deep learning model that can comprehensively analyze the crossing threshold duration index and the short-term cumulative increment index to generate the instantaneous enrichment reference value is acceptable. To implement the technical solution of the present invention, the present invention provides a specific implementation; the expression for generating the instantaneous enrichment reference value is: , where , are the preset proportionality factors of the crossing threshold duration index and the short-term cumulative increment index respectively, and , are both greater than 0. The preset proportionality factors ( and ) here refer to the parameters used to adjust and balance the contribution degrees of different indicators to the instantaneous enrichment reference value. Specifically, the crossing threshold duration index and the short-term cumulative increment index may have different importance and influence when calculating the instantaneous enrichment reference value. Therefore, it is necessary to use the preset proportionality factors and to assign their weights to the final calculation result. These factors are usually obtained based on empirical data, historical statistical analysis, or machine learning optimization to ensure that the calculated instantaneous enrichment reference value can more accurately reflect the enrichment risk of gas. For example, if the impact of the gas overlimit duration on the risk is more significant, then may be larger; while if the rapid change of short-term gas accumulation is more critical, then may be larger. Therefore, the setting of these proportionality factors is crucial for balancing the impacts of different risk factors to ensure that the instantaneous enrichment reference value can accurately characterize the change trend and potential risks of gas during the coal transportation process.

[0040] It can be seen from the instantaneous enrichment reference value that under the monitoring window, the larger the performance value of the cross-threshold duration index generated after in-depth analysis of the cumulative duration of the gas concentration continuously higher than the safety threshold, and the larger the performance value of the short-term cumulative increment index generated after in-depth analysis of the cumulative increment of the gas concentration relative to the baseline, that is, the larger the performance value of the instantaneous enrichment reference value generated when predicting the change trend of gas during the coal transportation process through the pre-trained support vector machine model, the greater the risk of potential gas accumulation during the coal transportation process. On the contrary, it indicates that the risk of potential gas accumulation during the coal transportation process is smaller.

[0041] Compare and analyze the instantaneous enrichment reference value generated when predicting the change trend of gas during the coal transportation process by the pre-trained support vector machine model with the pre-set instantaneous enrichment reference threshold to identify the gas accumulation risk during the coal transportation process. The specific identification process is as follows: If the instantaneous enrichment reference value is greater than the instantaneous enrichment reference threshold, then classify the gas evolution situation during the coal transportation under this monitoring window as gas accumulation; if the instantaneous enrichment reference value is less than or equal to the instantaneous enrichment reference threshold, then classify the gas evolution situation during the coal transportation under this monitoring window as normal evolution.

[0042] When the prediction model identifies the risk of gas accumulation, start the frequency regulation mechanism, automatically reduce the actual vibration frequency of the conveying system according to the degree of accumulation risk predicted by the model to reduce the intensity of coal body disturbance, reduce the gas evolution rate, relieve the large accumulation situation of gas, and conduct real-time feedback evaluation on the effect after regulation to form a closed-loop control; When the prediction model identifies the risk of gas accumulation, start the frequency regulation mechanism, and automatically reduce the actual vibration frequency of the conveying system according to the degree of accumulation risk predicted by the model. The specific steps are as follows: When the prediction model identifies the risk of gas accumulation, the vibration frequency adjustment amount is dynamically calculated according to the ratio of the instantaneous enrichment reference value to the instantaneous enrichment reference threshold. Through this calculation, it can be ensured that the adjusted vibration frequency can effectively inhibit the gas evolution rate while avoiding excessive reduction of the conveying efficiency. The vibration frequency adjustment amount depends on the degree of enrichment risk exceeding the limit, that is, the deviation degree between the instantaneous enrichment reference value and the instantaneous enrichment reference threshold. The calculation formula is: , where: is the vibration frequency adjustment amount, and a negative value indicates the amplitude of the vibration frequency reduction. is the initial vibration frequency, that is, the standard vibration frequency of the coal conveying system when there is no gas risk. is the instantaneous enrichment reference value, which is used to measure the current gas enrichment degree. is the instantaneous enrichment reference threshold, which is the critical value of the gas enrichment risk. is the adjustment coefficient, which is used to control the sensitivity of the vibration frequency adjustment, and is usually determined by experimental data or the coal mine conveying working conditions. is the non-linear weight factor, which is used to strengthen the adjustment amplitude in high-risk situations to make it more responsive. If , then the vibration frequency adjustment amount is a negative value, indicating that the vibration frequency needs to be reduced. The greater the degree of exceeding the limit (that is, is much higher than ), the greater the reduction amplitude of the vibration frequency to strongly inhibit the gas evolution rate. By using the exponential amplification effect of the non-linear weight factor , it can be ensured that in high-risk situations, the vibration adjustment is more sensitive, while in slightly exceeding the limit situations, the adjustment is more gentle.

[0043] After adjusting the vibration frequency, it is necessary to continuously monitor the adjustment effect to ensure that the gas evolution rate is inhibited and at the same time ensure that the coal conveying efficiency is not overly weakened. Therefore, a feedback evaluation mechanism is introduced. Within the monitored window after adjustment, the instantaneous enrichment reference value is recalculated and compared with that before adjustment to form a dynamic closed-loop optimization control. If the instantaneous enrichment reference value after adjustment still exceeds the limit, the vibration frequency is further reduced; if the gas evolution trend tends to be stable, the vibration frequency is appropriately restored to maintain the normal conveying efficiency. The expression for updating the vibration frequency through the dynamic closed-loop optimization control is: , where: is the actual vibration frequency after adjustment. is the minimum allowable vibration frequency to prevent the vibration frequency from dropping too low and affecting the conveying efficiency. is the vibration frequency of the previous monitored window to ensure the smoothness of the adjustment. is the recalculated instantaneous enrichment reference value after adjustment, which is used to evaluate the adjustment effect. is the feedback adjustment factor, which is used to control the response rate of feedback adjustment and ensure the stability of the adjustment process. is the feedback index, whose function is to dynamically adjust the recovery rate of the vibration frequency, making the adjustment process of the vibration frequency more smooth and self-adaptive, so as to ensure the effective suppression of the gas enrichment risk, and at the same time avoid excessive frequency adjustment from affecting the coal conveying efficiency. If the instantaneous enrichment reference value after adjustment is still over the limit, then the feedback index is small, resulting in a further decrease in the actual vibration frequency after adjustment to further strengthen the control of gas desorption. If the instantaneous enrichment reference value after adjustment tends to be normal, the exponential decay effect weakens, causing the vibration frequency to gradually recover to maintain the conveying efficiency. The exponential feedback mechanism can ensure a smooth transition during the adjustment process and avoid unstable coal flow or increased equipment wear caused by frequent vibration adjustment.

[0044] When the prediction model identifies the risk of gas accumulation, the frequency control mechanism is activated. By dynamically adjusting the actual vibration frequency of the conveying system in real time, it can automatically decrease according to the predicted degree of gas accumulation risk, effectively reducing the mechanical disturbance intensity of the coal body during the conveying process. After the coal body disturbance intensity decreases, it can significantly inhibit the desorption rate of internal gas (methane) in the coal, slow down the rapid release of gas from the coal body to the air, and thus avoid the gas concentration reaching or exceeding the explosion critical range (5%-15%) rapidly in the key areas of the conveying equipment. This mechanism enables the system to respond in a timely manner to the potential gas accumulation risk during coal conveying, proactively take control measures to inhibit the continuous increase of gas concentration, and reduce the explosion risk. In addition, after the conveying system reduces the vibration frequency, it will also conduct real-time feedback and dynamic evaluation on the gas concentration change through the sensing network, recalculate the instantaneous enrichment reference value in combination with the short-term cumulative increment index and the cross-threshold duration index, and perform adaptive fine adjustment on the frequency according to the feedback index, forming a closed-loop adaptive control system to achieve intelligent precise guarantee and efficient operation of conveying safety.

[0045] Through the above comprehensive processing method for underground coal mine safety monitoring data, it is possible to effectively achieve accurate monitoring and intelligent control of gas risks during the coal transportation process. Specifically, by reasonably setting the initial vibration frequency, while ensuring the coal transportation efficiency, it reduces the accumulation and spillage of coal. Then, through the real-time monitoring and recording of high-precision gas sensors at key positions of the transportation equipment, it can promptly capture the dynamic of rapid gas (methane) evolution caused by vibration during transportation, eliminating the blind spots and omissions in local areas in traditional monitoring methods. At the same time, the real-time monitored gas data is uploaded and stored in the cloud database in chronological order, generating trend-based and predictive feature vectors through in-depth analysis, and using a pre-trained support vector machine (SVM) model for intelligent prediction, further improving the accuracy and real-time performance of gas potential risk identification. When the prediction model detects an obvious risk of gas accumulation, the system automatically activates the dynamic vibration frequency regulation mechanism to reduce the actual vibration frequency of the transportation system, thereby effectively reducing the intensity of mechanical disturbance of the coal body, inhibiting the gas evolution speed, and preventing the further increase of gas concentration. The dynamic adjustment of the vibration frequency is automatically executed based on the predicted risk level, enabling the transportation working conditions to be dynamically adapted to the gas evolution situation. Finally, by feedback evaluating the regulation effect, a real-time closed-loop control strategy is formed to achieve precise gas safety control during the coal transportation process, greatly reducing the risk of gas explosion and providing a higher-level, more refined, and more intelligent guarantee means for coal mine production safety.

[0046] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0047] Only some exemplary embodiments of the present invention have been described above by way of illustration. Undoubtedly, for those of ordinary skill in the art, without departing from the spirit and scope of the present invention, the described embodiments can be modified in various different ways. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the protection scope of the claims of the present invention.

[0048] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed in this application can easily think of changes or substitutions, which should all be covered within the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims

1. A comprehensive processing method for safety monitoring data in coal mines, characterized in that The steps include the following: Set an initial vibration frequency for the conveying system in the coal mine underground, vibrate the conveyed coal with the inherent vibration amplitude, and establish a stable vibration working condition for coal transportation; Real-time monitor and record the gas data released due to vibration during coal transportation through high-precision gas sensors set at key positions of the conveying equipment; Transmit and store the gas concentration data collected by the gas sensors to the cloud server in real time, store it in chronological order, automatically generate a historical data analysis set of gas concentration, extract gas accumulation characteristic indicators from the analysis set, construct a characteristic vector of gas change, and input the characteristic vector into a pre-trained support vector machine model to predict the change trend and potential accumulation risk of gas during coal transportation in real time; When the prediction model identifies the existence of gas accumulation risk, start the frequency regulation mechanism, automatically reduce the actual vibration frequency of the conveying system according to the degree of accumulation risk predicted by the model, reduce the disturbance intensity of the coal body, reduce the gas release rate, and relieve the large-scale accumulation condition of gas; Conduct real-time feedback evaluation on the effect after regulation to form a closed-loop control.

2. The integrated processing method of coal mine underground safety monitoring data according to claim 1, characterized in that Form a sensing network with gas sensors, adopt multi-point layout and multi-angle distribution to cover the key areas of the conveying system.

3. A comprehensive processing method for safety monitoring data in coal mines according to claim 1, characterized in that Extract gas accumulation characteristic indicators from the analysis set. Among them, the extracted characteristic indicators include the cumulative duration of the gas concentration continuously higher than the safety threshold and the cumulative increment of the gas concentration relative to the baseline. Under the monitoring window, after deeply analyzing the extracted characteristics, generate a cross-threshold duration indicator and a short-term cumulative increment indicator respectively. Construct a characteristic vector of gas change through the cross-threshold duration indicator and the short-term cumulative increment indicator to quantify the cumulative effect and over-limit persistence of gas within the monitoring window, evaluate the severity of the gas concentration exceeding the safety threshold and its enrichment trend in a short period of time, and identify the evolution process of gas explosion risk.

4. A comprehensive processing method for safety monitoring data in coal mines according to claim 3, characterized in that, Input the cross-threshold duration indicator and the short-term cumulative increment indicator into a pre-trained support vector machine model, and output an instantaneous enrichment reference value through the support vector machine model. Based on the instantaneous enrichment reference value, predict the change trend and potential accumulation risk of gas during coal transportation in real time.

5. A comprehensive processing method for coal mine underground safety monitoring data according to claim 4, characterized in that Compare and analyze the instantaneous enrichment reference value generated when the pre-trained support vector machine model predicts the change trend of gas during coal transportation with the pre-set instantaneous enrichment reference threshold to identify the gas accumulation risk during coal transportation. The specific identification process is as follows: If the instantaneous enrichment reference value is greater than the instantaneous enrichment reference threshold, classify the gas release situation during coal transportation under this monitoring window as gas accumulation; if the instantaneous enrichment reference value is less than or equal to the instantaneous enrichment reference threshold, classify the gas release situation during coal transportation under this monitoring window as normal release.

6. A comprehensive processing method for coal mine underground safety monitoring data according to claim 5, characterized in that When the prediction model identifies the existence of gas accumulation risk, start the frequency regulation mechanism, and automatically reduce the actual vibration frequency of the conveying system according to the degree of accumulation risk predicted by the model. The specific steps are as follows: When the prediction model identifies the existence of gas accumulation risk, dynamically calculate the vibration frequency adjustment amount according to the ratio of the instantaneous enrichment reference value to the instantaneous enrichment reference threshold. The calculation formula is: ， Wherein: is the vibration frequency adjustment amount, and a negative value indicates the amplitude of the vibration frequency reduction, is the initial vibration frequency, that is, the standard vibration frequency of the coal conveying system when there is no gas risk, is the instantaneous enrichment reference value, which is used to measure the current gas enrichment degree, is the instantaneous enrichment reference threshold, which is the critical value of the gas enrichment risk, is the adjustment coefficient, which is used to control the sensitivity of the vibration frequency adjustment, is the non-linear weight factor, which is used to strengthen the adjustment amplitude in high-risk situations.

7. A comprehensive processing method for underground coal mine safety monitoring data according to claim 6, characterized in that Introduce a feedback evaluation mechanism. Within the adjusted monitoring window, recalculate the instantaneous enrichment reference value and compare it with that before adjustment to form a dynamic closed-loop optimization control. The expression for updating the vibration frequency through the dynamic closed-loop optimization control is as follows: ， Wherein: is the actual vibration frequency after adjustment, is the lowest allowable vibration frequency, is the vibration frequency of the previous monitoring window, is the instantaneous enrichment reference value recalculated after adjustment, used to evaluate the adjustment effect, is the feedback adjustment factor, used to control the response rate of the feedback adjustment.

8. A comprehensive processing method for safety monitoring data in coal mines according to claim 3, characterized in that Under the monitoring window, the specific steps for generating the cross-threshold duration index after deeply analyzing the cumulative duration during which the gas concentration continuously exceeds the safety threshold are as follows: Within the monitoring window, the gas concentration is monitored in real time, and a safety threshold is set. When the gas concentration exceeds this safety threshold, record the cumulative value of the duration of the over-threshold state. The statistical expression for the cumulative value of the duration of the over-threshold state is: ， Wherein: is the cumulative value of the duration of the over-threshold state, representing the total time that the gas is in a high-risk state within the monitoring window, is the Heaviside step function, used to determine whether the gas concentration exceeds the threshold, and is defined as: ， Among them, is the gas concentration at time t, is the safety threshold set for gas, represents the time range of the monitoring window, which is used to count the situation of gas exceeding the threshold within the monitoring window. Among them, and represent the starting point and the ending point of the monitoring window respectively; Based on the cumulative value of the duration of the over-threshold state, considering the exponential amplification effect of the over-threshold duration, construct the cross-threshold duration index, and enhance the risk characterization ability through the exponential transformation in the time dimension. The construction expression for the cross-threshold duration index is: ， Wherein: is the threshold-crossing duration index, used to quantify the cumulative effect of gas exceeding the threshold; is the adjustment coefficient, used to control the amplitude of exponential growth; is the exponential growth factor, used to adjust the influence of the time exceeding the threshold on the final threshold-crossing duration index.

9. A comprehensive processing method for safety monitoring data in coal mines according to claim 3, characterized in that Under the monitoring window, the specific steps for generating the short-term cumulative increment index after deeply analyzing the cumulative increment of the gas concentration relative to the baseline are as follows: Calculate the short-term cumulative amount above the baseline, considering not only the degree above the baseline but also the cumulative impact of the state above the baseline. The calculation expression is: ， Wherein: is the short-term cumulative amount exceeding the baseline, which is used to measure the total accumulation degree of the part of gas exceeding the baseline within the monitoring window; is the gas concentration at the i th sampling point within the monitoring window; is the set baseline gas concentration; is the weight factor, which assigns different influence weights to different time points within the monitoring window to highlight high-risk areas; is the non-linear adjustment parameter. Introduce a mapping function to convert the short-term cumulative amount above the baseline into a short-term cumulative increment index to directly quantify the severity of the gas exceeding the baseline. The specific conversion expression is: ， Wherein: is a short-term cumulative increment index, used to quantify the persistence of gas exceeding the baseline within the monitoring window and provide a trend assessment, M is a reference limit value, used for normalization processing, is a logarithmic mapping coefficient, used to control the sensitivity when the cumulative amount of gas exceeding the baseline is low, is a non-linear normalization coefficient, controlling the non-linear amplification degree of the short-term cumulative increment index, is a trend weighting coefficient for gas exceeding the baseline, controlling the influence of the cumulative amount of gas exceeding the baseline on the final short-term cumulative increment index.

Images