A safety warning method and system for a coal mine sensing device

Abstract

The application discloses a kind of safety early warning method and system of coal mine sensing equipment, it is related to coal mine technical field, and its technical solution main points include following steps: by intelligent sensor, the real-time environmental parameter of coal mine operating environment and the equipment running data of coal mine target equipment are acquired, the safety early warning demand of coal mine equipment is acquired;Real-time monitoring model is constructed according to real-time environmental parameter and equipment running data, and fault correlation data block is obtained by analysis according to fault feature library and real-time monitoring model;According to safety early warning demand, obtain early warning attention content and early warning time threshold, obtain target early warning data according to early warning attention content and fault correlation data block analysis;Early warning attention content, target early warning data and early warning time threshold are processed and analyzed to obtain comprehensive early warning coefficient;According to comprehensive early warning coefficient, the target early warning data corresponding to safety early warning demand is graded output, and the effect is to effectively reduce the probability of accident.

Description

Technical Field

[0001] This invention relates to the field of coal mining technology, and more specifically, to a safety early warning method and system for coal mine sensing equipment. Background Technology

[0002] Traditional coal mine safety monitoring methods mostly focus on monitoring environmental parameters or equipment operating status individually, failing to link the risks associated with these two aspects. For example, high-load equipment operation exacerbates fluctuations in environmental parameters, but single monitoring cannot reflect the interconnected risks, easily leading to one-sided risk identification. Furthermore, traditional early warning systems often rely on manual experience or simple threshold triggers, making it difficult to match historical fault characteristics to identify potential risks in advance, and also unable to filter core early warning information according to different safety needs, thus failing to meet the safety management requirements of complex operational scenarios. Most solutions can only issue uniform warning signals, unable to match corresponding response measures based on the severity and urgency of the risk. Over-responding to low-risk situations reduces operational efficiency, while under-responding to high-risk situations may trigger accidents, making it difficult to achieve a balance between safety and production. Summary of the Invention

[0003] In view of the shortcomings of the existing technology, the purpose of this invention is to provide a safety early warning method and system for coal mine sensing equipment.

[0004] To achieve the above objectives, the present invention provides the following technical solution: A safety early warning method based on coal mine sensing equipment, the method comprising the following steps: By acquiring real-time environmental parameters of the coal mine operating environment and equipment operation data of target equipment in the coal mine through intelligent sensors, safety early warning requirements for coal mine equipment can be obtained. The environmental risk assessment window is obtained by analyzing real-time environmental parameters, the equipment failure risk level is obtained by analyzing the multi-dimensional operating characteristics of equipment operation data, and the corresponding early warning association level within the environmental risk assessment window is obtained by analyzing the environmental dynamic parameters within the environmental risk assessment window. A fault feature library is constructed based on historical fault data of target equipment in coal mines. A real-time monitoring model is constructed based on real-time environmental parameters and equipment operation data. Fault-related data blocks are obtained by analyzing the fault feature library and the real-time monitoring model. Based on the safety early warning requirements, the content to be concerned and the warning time threshold are obtained. The target early warning data is obtained by analyzing the data blocks related to the faults and the content to be concerned in the early warning. The comprehensive early warning coefficient is obtained by processing and analyzing the content of the early warning, the target early warning data, and the early warning time threshold. Based on the comprehensive early warning coefficient, target early warning data corresponding to the safety early warning needs are output in a graded manner.

[0005] Preferably, the environmental risk assessment window is obtained by analyzing real-time environmental parameters, specifically including the following steps: The time-series environmental data of the coal mine operation environment is obtained based on real-time environmental parameters; wherein, the time-series environmental data includes basic environmental characteristics, dynamic interference characteristics, and time period characteristics; The time-series environmental data of the coal mine operation environment is input into the environmental risk prediction model. The environmental risk prediction model outputs the risk prediction value of the coal mine operation environment based on the time-series environmental data. Each risk prediction value corresponds to a time window. The risk prediction value includes the predicted gas concentration, the predicted dust content, and the predicted temperature fluctuation range. The environmental risk prediction model obtains the assessment priority coefficient corresponding to the time window based on the risk prediction value, and then obtains the environmental risk assessment window based on the assessment priority coefficient.

[0006] Preferably, the equipment failure risk level is obtained by analyzing the multi-dimensional operational characteristics of the equipment operation data, specifically including the following steps: Multi-dimensional operational characteristics include equipment load intensity, runtime, fault history, and component wear level; Set up an equipment evaluation model, which includes multi-dimensional operating characteristic weights, including load intensity weight, runtime weight, fault record weight, and wear level weight; The multi-dimensional operational characteristics of equipment operation data are input into the equipment evaluation model. The equipment evaluation model performs corresponding feature weighting and accumulation on the multi-dimensional operational characteristics of the equipment operation data according to the multi-dimensional operational characteristics, and obtains and outputs the fault risk coefficient corresponding to the equipment operation data. Set a preset number of fault risk level ranges, and each fault risk level range corresponds to a fault risk level. The fault risk coefficient corresponding to the equipment operation data is compared with the fault risk level range, and the fault risk level corresponding to the fault risk level range where the fault risk coefficient is located is marked as the equipment fault risk level.

[0007] Preferably, the environmental dynamic parameters within the environmental risk assessment window are analyzed to obtain the corresponding early warning correlation level within the environmental risk assessment window, specifically including the following steps: The environmental dynamic parameters include the rate of change of environmental parameters and the frequency of parameter fluctuations within the environmental risk assessment window; Obtain the number of fault risk level intervals, and divide the environmental parameter safety threshold range based on the number of level intervals to obtain the environmental dynamic interval; By comparing the frequency of parameter fluctuations within the environmental risk assessment window with the environmental dynamic range, the environmental dynamic range in which the frequency of parameter fluctuations is located can be obtained. The fault risk level corresponding to the environmental dynamic range is marked as the corresponding early warning association level within the environmental risk assessment window.

[0008] Preferably, a fault feature database is constructed based on historical fault data of target equipment in the coal mine, specifically including the following steps: The historical fault data of the target equipment in the coal mine is divided into historical fault data blocks; Fault feature nodes are generated based on historical fault data blocks, and feature identification values ​​corresponding to the fault feature nodes are obtained by extracting features from the historical fault data blocks. A fault feature library for coal mine target equipment is constructed based on fault feature nodes and their corresponding feature identifier values.

[0009] Preferably, a real-time monitoring model is constructed based on real-time environmental parameters and equipment operation data, specifically including the following steps: Real-time environmental parameters and equipment operation data are divided into real-time monitoring data blocks; Real-time monitoring nodes are generated based on real-time monitoring data blocks, and the monitoring identifier values ​​corresponding to the real-time monitoring nodes are obtained by feature extraction from the real-time monitoring data blocks. A real-time monitoring model is constructed based on the real-time monitoring nodes and the monitoring identifier values ​​corresponding to the real-time monitoring nodes.

[0010] Preferably, the fault-related data block is obtained by analyzing the fault feature library and the real-time monitoring model, specifically including the following steps: The fault feature nodes in the fault feature library are compared with the corresponding identifier values ​​of the real-time monitoring nodes in the real-time monitoring model. If there is a real-time monitoring node in the real-time monitoring model whose matching degree with the identifier value corresponding to the fault feature node exceeds a preset threshold, then the real-time monitoring node is marked as an associated node, and the real-time monitoring data block corresponding to the associated node is recorded as a fault associated data block.

[0011] Preferably, the target early warning data is obtained by analyzing the early warning focus content and fault-related data blocks, specifically including the following steps: The standard parameters of the warning content are compared with the standard parameters of the storage of the fault-related data block to obtain the parameter matching ratio between the storage standard parameters and the warning standard parameters. If the percentage of matching between the storage standard parameters and the early warning standard parameters is greater than a preset matching percentage threshold, then the fault-related data block corresponding to the storage standard parameters is determined to be the target early warning data.

[0012] Preferably, the comprehensive early warning coefficient is obtained by processing and analyzing the content of the early warning, the target early warning data, and the early warning time threshold, specifically including the following steps: Obtain the parameter feature matching degree between the warning focus content and the target warning data; The first warning coefficient corresponding to the target warning data is obtained by summing up the matching degree of each parameter feature of the warning focus and the target warning data. Obtain the monitoring time node corresponding to the target warning data, and calculate the second warning coefficient corresponding to the target warning data by the difference between the monitoring time node and the warning time threshold of the warning content. The comprehensive early warning coefficient is obtained based on the first and second early warning coefficients.

[0013] A safety early warning system based on coal mine sensing equipment includes: Acquisition module: Acquires real-time environmental parameters of the coal mine operating environment and equipment operation data of target equipment in the coal mine, and acquires safety early warning requirements for coal mine equipment; The first analysis module analyzes real-time environmental parameters to obtain an environmental risk assessment window, analyzes the multi-dimensional operating characteristics of equipment operation data to obtain the equipment failure risk level, and analyzes the environmental dynamic parameters in the environmental risk assessment window to obtain the corresponding early warning association level within the environmental risk assessment window. Construction module: Construct a fault feature library based on historical fault data of target equipment in coal mines, construct a real-time monitoring model based on real-time environmental parameters and equipment operation data, and analyze the fault feature library and real-time monitoring model to obtain fault-related data blocks; The second analysis module: Based on the safety early warning requirements, it obtains the content of concern for early warning and the early warning time threshold, and analyzes the data blocks related to the faults to obtain the target early warning data. Processing module: Processes and analyzes the warning content, target warning data, and warning time threshold to obtain a comprehensive warning coefficient; Output module: Based on the comprehensive early warning coefficient, outputs the target early warning data corresponding to the safety early warning requirements.

[0014] Compared with the prior art, the present invention has the following beneficial effects: This invention collects real-time parameters of the coal mine operating environment and equipment operation data. It identifies high-risk operating periods through an environmental risk assessment window, clarifies the degree of equipment hazard based on equipment failure risk levels, and correlates dynamic environmental parameters with equipment failure levels. This allows for a linked assessment of environmental and equipment risks, avoiding omissions due to monitoring only the environment or equipment, and ensuring more comprehensive safety early warning coverage. By constructing a historical fault feature database and a real-time monitoring model, it matches fault-related data blocks with similar characteristics to historical faults from real-time data. Combined with safety early warning requirements, it filters target early warning data, reducing interference from invalid warnings and enabling relevant personnel to quickly identify core risk points. A comprehensive early warning coefficient quantifies and integrates the matching intensity and urgency of risks. The tiered response mechanism ensures that low-risk situations do not disrupt normal operations while allowing high-risk hazards to be addressed promptly, improving the efficiency and flexibility of safety management. This method does not only respond to risks that have already occurred, but also predicts risks in the future by using an environmental risk prediction model. It combines a historical fault feature database to identify real-time states similar to past faults in advance, helping coal mine operations to prepare for risk prevention and control in advance, effectively reducing the probability of accidents, and providing more reliable technical support for the safe operation of operations. Attached Figure Description

[0015] Figure 1 This is a schematic diagram illustrating the steps of a safety early warning method based on coal mine sensing equipment, as provided in an embodiment of the present invention. Figure 2 This invention provides a schematic diagram of a safety early warning system based on coal mine sensing equipment. Detailed Implementation

[0016] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0017] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0018] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single embodiment or an embodiment selectively excluded from other embodiments.

[0019] Reference Figures 1-2 As shown.

[0020] The embodiments further illustrate the safety early warning method and system for coal mine sensing equipment proposed in this invention.

[0021] A safety early warning method based on coal mine sensing equipment, the method comprising the following steps: By acquiring real-time environmental parameters of the coal mine operating environment and equipment operation data of target equipment in the coal mine through intelligent sensors, safety early warning requirements for coal mine equipment can be obtained. The environmental risk assessment window is obtained by analyzing real-time environmental parameters, the equipment failure risk level is obtained by analyzing the multi-dimensional operating characteristics of equipment operation data, and the corresponding early warning association level within the environmental risk assessment window is obtained by analyzing the environmental dynamic parameters within the environmental risk assessment window. A fault feature library is constructed based on historical fault data of target equipment in coal mines. A real-time monitoring model is constructed based on real-time environmental parameters and equipment operation data. Fault-related data blocks are obtained by analyzing the fault feature library and the real-time monitoring model. Based on the safety early warning requirements, the content to be concerned and the warning time threshold are obtained. The target early warning data is obtained by analyzing the data blocks related to the faults and the content to be concerned in the early warning. The comprehensive early warning coefficient is obtained by processing and analyzing the content of the early warning, the target early warning data, and the early warning time threshold. Based on the comprehensive early warning coefficient, target early warning data corresponding to the safety early warning needs are output in a graded manner.

[0022] First, it is necessary to clarify the grading standards corresponding to the comprehensive early warning coefficient. Typically, the comprehensive early warning coefficient is divided into multiple level ranges based on actual safety needs, with each range corresponding to a specific early warning level. For example, four level ranges can be set: a comprehensive early warning coefficient between 0 and 1 corresponds to Level 1 early warning, between 1 and 2 corresponds to Level 2 early warning, between 2 and 3 corresponds to Level 3 early warning, and 3 and above corresponds to Level 4 early warning, where Level 1 is a low-risk warning and Level 4 is an extremely high-risk warning.

[0023] The overall warning coefficient is compared with these level ranges to determine the corresponding warning level. For example, if the overall warning coefficient of a target's warning data is 1.42, which falls within the range of 1 to 2, then the corresponding warning level is Level II.

[0024] Based on the determined warning level, corresponding target warning data is output. For example, a Level 1 warning only requires a pop-up message on the equipment management terminal, briefly displaying the core parameters of the target warning data; a Level 2 warning prompts the terminal and pushes a message to the on-site operation supervisor, supplementing the explanation of the basic impact range of the risk; a Level 3 warning will activate on-site audible and visual alarms, and simultaneously push detailed target warning data (including parameter change trends and associated equipment status) to the management backend and the operation supervisor; a Level 4 warning immediately triggers an emergency shutdown command for the equipment, simultaneously pushing the highest priority target warning data to the mine-wide safety management system, along with emergency evacuation instructions.

[0025] Taking a comprehensive early warning coefficient of 2.8 as an example, the corresponding early warning level is Level III. At this time, the on-site audible and visual alarm device is activated, and the detailed content of the target early warning data, including the real-time changes in gas concentration and the fluctuation curve of equipment load, is simultaneously sent to the monitoring screen of the coal mine safety management backend and the mobile terminal of the on-site work team leader, to ensure that relevant personnel can obtain complete risk information in a timely manner and take action.

[0026] This tiered output approach avoids low-risk warnings from excessively interfering with normal operations, while ensuring that high-risk warnings receive sufficient attention and rapid response, thus achieving more precise and efficient safety warnings.

[0027] The environmental risk assessment window is obtained by analyzing real-time environmental parameters, specifically including the following steps: The time-series environmental data of the coal mine operation environment is obtained based on real-time environmental parameters; the time-series environmental data includes basic environmental characteristics, dynamic disturbance characteristics, and time period characteristics. The time-series environmental data of the coal mine operation environment is input into the environmental risk prediction model. The environmental risk prediction model outputs the risk prediction values ​​of the coal mine operation environment based on the time-series environmental data. Each risk prediction value corresponds to a time window. Among them, the risk prediction values ​​include the predicted gas concentration, the predicted dust content, and the predicted temperature fluctuation range. The environmental risk prediction model obtains the assessment priority coefficient corresponding to the time window based on the risk prediction value, and then obtains the environmental risk assessment window based on the assessment priority coefficient.

[0028] First, time-series environmental data is extracted from real-time environmental parameters of the coal mine operating environment. Basic environmental characteristics are the baseline state parameters of the operating environment, such as stable baseline values ​​for methane concentration and normal air temperature during normal operations. These values ​​serve as the basic reference for judging whether the environment is abnormal. Dynamic disturbance characteristics are the changes in environmental parameters, such as the rate of increase in methane concentration within a short period and the degree of sudden fluctuation in dust content. These changes are direct manifestations of environmental risk. Time-period characteristics are the operating period information corresponding to the parameters, such as the early shift and the middle shift. The intensity of work varies at different times, and the scope of environmental risk impact will also differ.

[0029] Time-series environmental data is input into the environmental risk prediction model. Based on this data, the model calculates predicted risk values ​​for the coal mine operating environment, with each predicted risk value corresponding to a time window. These predicted risk values ​​specifically include predicted gas concentration, predicted dust content, and predicted temperature fluctuations. For example, the environmental risk prediction model might predict that within a 10-minute time window, the gas concentration will rise to 0.9% and the dust content will reach 15 mg / m³. 3 The temperature fluctuates between 25℃ and 30℃.

[0030] The environmental risk prediction model calculates an assessment priority coefficient for each time window based on the predicted risk values. This priority coefficient is derived by combining the weights of each predicted risk value. For example, if the weight of predicted methane concentration is set to 0.5, predicted dust content to 0.3, and predicted temperature fluctuation range to 0.2, the assessment priority coefficient = 0.5 × predicted methane concentration + 0.3 × predicted dust content + 0.2 × predicted temperature fluctuation range. Assuming a predicted methane concentration of 0.9, predicted dust content of 15, and predicted temperature fluctuation range of 5 for a certain time window, substituting these values ​​into the formula yields the assessment priority coefficient = 0.5 × 0.9 + 0.3 × 15 + 0.2 × 5 = 0.45 + 4.5 + 1 = 5.95.

[0031] The environmental risk prediction model selects time windows based on the magnitude of the assessment priority coefficient, identifying those with higher priority coefficients as environmental risk assessment windows. For example, when the assessment priority coefficient threshold is set to 5, the time window corresponding to 5.95 calculated above is included in the environmental risk assessment window for subsequent focused monitoring of environmental risks during that period.

[0032] The equipment failure risk level is determined by analyzing the multi-dimensional operational characteristics of equipment operation data, specifically including the following steps: Multi-dimensional operational characteristics include equipment load intensity, runtime, fault history, and component wear level; Set up an equipment evaluation model, which includes multi-dimensional operating characteristic weights, including load intensity weight, runtime weight, fault record weight, and wear level weight; The multi-dimensional operational characteristics of equipment operation data are input into the equipment evaluation model. The equipment evaluation model performs corresponding feature weighting and accumulation on the multi-dimensional operational characteristics of the equipment operation data according to the multi-dimensional operational characteristics, and obtains and outputs the fault risk coefficient corresponding to the equipment operation data. Set a preset number of fault risk level ranges, and each fault risk level range corresponds to a fault risk level. The fault risk coefficient corresponding to the equipment operation data is compared with the fault risk level range, and the fault risk level corresponding to the fault risk level range where the fault risk coefficient is located is marked as the equipment fault risk level.

[0033] First, it's crucial to define the equipment's multi-dimensional operational characteristics, specifically including load intensity, runtime, fault history, and component wear. Load intensity refers to the workload the equipment bears during operation; for example, if a coal mine transport equipment is currently operating at 90% of its rated load, this is its current load intensity. Runtime is the duration of continuous operation; for example, if the transport equipment has been running continuously for 8 hours. Fault history records past malfunctions; for example, if the equipment has experienced two minor malfunctions in the past month. Component wear refers to the wear or aging of the equipment's core components; for example, if the transmission components of the equipment have reached 30% wear.

[0034] The equipment evaluation model includes multi-dimensional operating characteristic weights, which means that each operating characteristic is assigned a corresponding weight coefficient. Common weight settings can be load intensity weight of 0.3, running time weight of 0.2, fault record weight of 0.3, and wear level weight of 0.2.

[0035] The multi-dimensional operational characteristics corresponding to the equipment operation data are input into the equipment evaluation model. The equipment evaluation model performs a weighted sum of these characteristics to obtain the fault risk coefficient. Fault Risk Coefficient = Load Intensity × Load Intensity Weight + Running Time × Running Time Weight + Fault History Record × Fault Record Weight + Component Wear Level × Wear Level Weight. Assuming the quantified value of load intensity is 90, the quantified value of running time is 8, the quantified value of fault history record is 2, and the quantified value of component wear level is 30, substituting these values ​​into the formula yields the fault risk coefficient = 90 × 0.3 + 8 × 0.2 + 2 × 0.3 + 30 × 0.2 = 27 + 1.6 + 0.6 + 6 = 35.2.

[0036] A preset number of fault risk level ranges need to be set, with each range corresponding to a specific fault risk level. For example, four ranges can be set: 0 to 20 corresponds to a low risk level, 21 to 40 to a medium risk level, 41 to 60 to a high risk level, and 61 and above to an extremely high risk level.

[0037] The equipment failure risk coefficient is compared with the failure risk level range to determine the range in which the equipment failure risk coefficient falls, and the failure risk level corresponding to that range is then marked as the equipment's failure risk level. For example, if the failure risk coefficient is 35.2 and falls within the range of 21 to 40, then the failure risk level corresponding to this equipment is medium risk.

[0038] The environmental dynamic parameters within the environmental risk assessment window are analyzed to obtain the corresponding early warning correlation level within the window. This process includes the following steps: Environmental dynamic parameters include the rate of change of environmental parameters and the frequency of parameter fluctuations within the environmental risk assessment window; Obtain the number of fault risk level intervals, and divide the environmental parameter safety threshold range based on the number of level intervals to obtain the environmental dynamic interval; By comparing the frequency of parameter fluctuations within the environmental risk assessment window with the environmental dynamic range, the environmental dynamic range in which the frequency of parameter fluctuations is located can be obtained. The fault risk level corresponding to the environmental dynamic range is marked as the corresponding early warning association level within the environmental risk assessment window.

[0039] Environmental dynamic parameters include the rate of change of environmental parameters and the frequency of parameter fluctuations within the environmental risk assessment window. The rate of change of environmental parameters refers to the degree of change of environmental parameters per unit time. For example, if the gas concentration rises from 0.5% to 0.7% within 1 minute, the rate of change is 0.2% / minute. The frequency of parameter fluctuations refers to the number of times an environmental parameter fluctuates within the environmental risk assessment window. For example, if the dust content shows 6 significant fluctuations within a 10-minute assessment window, this value is the frequency of parameter fluctuations within that window.

[0040] Obtain the number of risk level intervals for equipment failure. Assuming the equipment failure risk level was previously divided into four intervals—low, medium, high, and extremely high—then the number of intervals is four. Next, divide the environmental parameter safety threshold ranges into corresponding dynamic environmental intervals. For example, the safety threshold range for dust content is 0 to 20 mg / m³. 3 The concentration was divided into four intervals, resulting in 0-5 mg / m³. 3 5-10 mg / m 3 10-15mg / m 3 15-20mg / m 3The environmental dynamic range, and each environmental dynamic range corresponds to the low, medium, high and extremely high levels of equipment failure risk in sequence.

[0041] The frequency of parameter fluctuations within an environmental risk assessment window is compared with the environmental dynamic range to determine the environmental dynamic range in which that frequency falls. For example, if the quantified value corresponding to the frequency of dust content fluctuations within a certain environmental risk assessment window is 12, and 12 falls within the range of 10-15 mg / m³, then... 3 Within the corresponding range, the environmental dynamic range in which the fluctuation frequency of this parameter falls is 10-15 mg / m³. 3 The corresponding interval.

[0042] The fault risk level corresponding to the environmental dynamic range is marked as the corresponding early warning association level within the environmental risk assessment window. For example, 10-15 mg / m³. 3 If the corresponding equipment failure risk level is high risk, then the warning correlation level corresponding to the environmental risk assessment window is also high risk, thereby realizing the correlation between environmental risk and equipment failure risk.

[0043] A fault feature database is constructed based on historical fault data of target equipment in coal mines, specifically including the following steps: The historical fault data of the target equipment in the coal mine is divided into historical fault data blocks; Fault feature nodes are generated based on historical fault data blocks, and feature identification values ​​corresponding to the fault feature nodes are obtained by extracting features from the historical fault data blocks. A fault feature library for coal mine target equipment is constructed based on fault feature nodes and their corresponding feature identifier values.

[0044] First, the historical fault data of the target equipment in the coal mine needs to be divided into historical fault data blocks. For example, the historical fault data of a coal mine's ventilation equipment may contain multiple fault records. The complete information corresponding to each fault, including the equipment's load intensity, operating time, and environmental parameters at the time of the fault, will be divided into a separate historical fault data block. Assuming that the equipment has 3 historical faults, then 3 independent historical fault data blocks will be formed.

[0045] Fault feature nodes are generated for each historical fault data block, and feature extraction is performed on the historical fault data blocks to obtain the feature identifier value corresponding to the fault feature node. A fault feature node is an abstract classification of the core features of a single fault. For example, if a fault corresponding to a certain historical fault data block is component damage caused by excessive load intensity and runtime exceeding a threshold, the core features of this fault will form a fault feature node. Feature extraction quantifies these core features into feature identifier values. For example, the quantized values ​​of load intensity and runtime are combined and calculated according to a preset rule. Assuming the quantized load intensity is 95 and the quantized runtime is 10, the feature identifier value = load intensity × 0.6 + runtime × 0.4. Substituting the values, we get the feature identifier value = 95 × 0.6 + 10 × 0.4 = 57 + 4 = 61.

[0046] A fault feature library for coal mine target equipment is constructed based on fault feature nodes and their corresponding feature identifier values. This involves integrating the fault feature nodes corresponding to all historical fault data blocks, along with the feature identifier values ​​of each node, into a unified database. For example, three fault feature nodes and their feature identifier values ​​corresponding to three historical faults are stored in the database. During subsequent real-time monitoring, the feature identifier values ​​can be compared to match the corresponding historical fault features, thus assisting in determining whether the current equipment status poses a fault risk.

[0047] A real-time monitoring model is constructed based on real-time environmental parameters and equipment operation data, specifically including the following steps: Real-time environmental parameters and equipment operation data are divided into real-time monitoring data blocks; Real-time monitoring nodes are generated based on real-time monitoring data blocks, and the monitoring identifier values ​​corresponding to the real-time monitoring nodes are obtained by feature extraction from the real-time monitoring data blocks. A real-time monitoring model is constructed based on the real-time monitoring nodes and the monitoring identifier values ​​corresponding to the real-time monitoring nodes.

[0048] First, real-time environmental parameters and equipment operation data are divided into real-time monitoring data blocks. For example, the real-time data of a coal mining machine, including the gas concentration and dust content within a 5-minute period, as well as the equipment operation data such as the load intensity and operating duration of the mining machine during that period, are integrated into a single real-time monitoring data block. If divided at 5-minute intervals, 12 real-time monitoring data blocks will be formed within one hour.

[0049] Real-time monitoring nodes are generated based on each real-time monitoring data block, and feature extraction is performed on the real-time monitoring data blocks to obtain the monitoring identifier value corresponding to the real-time monitoring node. A real-time monitoring node is an abstract classification of the core characteristics of the equipment and environmental status during the current time period. For example, if a real-time monitoring data block corresponds to a state where the gas concentration is slightly high and the equipment load is at a medium level, the core characteristics of this state will form a real-time monitoring node. Feature extraction quantifies these core characteristics into monitoring identifier values: Monitoring identifier value = quantified environmental parameter value × 0.5 + quantified equipment operating parameter value × 0.5. Assuming the quantified value of the gas concentration during this time period is 60 and the quantified value of the equipment load intensity is 70, substituting these values ​​into the formula yields the monitoring identifier value = 60 × 0.5 + 70 × 0.5 = 30 + 35 = 65.

[0050] A real-time monitoring model is constructed based on real-time monitoring nodes and their corresponding monitoring identifier values. All real-time monitoring nodes corresponding to all real-time monitoring data blocks, along with the monitoring identifier values ​​of each node, are integrated into a dynamically updated model. For example, the 12 real-time monitoring nodes and their monitoring identifier values ​​within one hour are continuously stored in the real-time monitoring model. Subsequently, by comparing these monitoring identifier values ​​with feature identifier values ​​in the fault feature database, it is possible to quickly identify whether the current state matches historical fault characteristics and promptly detect potential risks.

[0051] The fault-related data blocks are obtained through analysis based on the fault feature library and real-time monitoring model, specifically including the following steps: The fault feature nodes in the fault feature library are compared with the corresponding identifier values ​​of the real-time monitoring nodes in the real-time monitoring model. If there is a real-time monitoring node in the real-time monitoring model whose matching degree with the identifier value corresponding to the fault feature node exceeds a preset threshold, then the real-time monitoring node is marked as an associated node, and the real-time monitoring data block corresponding to the associated node is recorded as a fault associated data block.

[0052] First, the fault feature nodes in the fault feature library need to be compared with the corresponding identifier values ​​of the real-time monitoring nodes in the real-time monitoring model. The identifier value of the fault feature node is the quantification result of the core features of historical faults, while the identifier value of the real-time monitoring node is the quantification result of the current equipment and environmental status. The comparison is to determine the feature similarity between the two.

[0053] During the comparison process, it is necessary to calculate the matching degree between the real-time monitoring node identifier value and the fault feature node identifier value. Matching degree = 1 - |Real-time monitoring identifier value - Fault feature identifier value| / Fault feature identifier value. For example, if the identifier value of a fault feature node is 61 and the identifier value of a real-time monitoring node is 65, substituting these values ​​into the formula yields a matching degree of 1 - |65 - 61| / 61 = 1 - 4 / 61 ≈ 0.934, meaning the matching degree is approximately 93.4%.

[0054] A preset threshold needs to be set to determine whether the match is valid; a common threshold is 80%. If a real-time monitoring node in the real-time monitoring model has a match rate with the identifier value of a fault feature node that exceeds this preset threshold, then the real-time monitoring node is marked as an associated node.

[0055] The real-time monitoring data block corresponding to the associated node is recorded as the fault-associated data block. For example, if the associated node corresponds to a data block of equipment operation data and environmental parameters within 5 minutes, this data block is identified as the fault-associated data block.

[0056] The target warning data is obtained by analyzing the warning focus and fault correlation data blocks, specifically including the following steps: The standard parameters of the warning content are compared with the standard parameters of the storage of the fault-related data block to obtain the parameter matching ratio between the storage standard parameters and the warning standard parameters. If the percentage of matching between the storage standard parameters and the early warning standard parameters is greater than a preset matching percentage threshold, then the fault-related data block corresponding to the storage standard parameters is determined to be the target early warning data.

[0057] First, clarify the standard parameters for the warning focus and the storage standard parameters for the fault-related data blocks. The warning focus is the core concern of the safety warning. For example, if the current warning focuses on excessive gas concentration and excessive equipment load, the corresponding standard parameters would include the gas concentration threshold and the equipment load intensity threshold. The storage standard parameters for the fault-related data blocks are the actual environmental and equipment parameters recorded in the data block, such as the gas concentration value and the equipment load intensity value stored in a certain fault-related data block.

[0058] The standard parameters for early warning are compared with the storage standard parameters of the fault-related data block, and the parameter matching ratio is calculated. Parameter matching ratio = number of matched parameter items ÷ total number of early warning standard parameters × 100%. For example, if the early warning concern content corresponds to two standard parameters: gas concentration ≥ 0.8% and equipment load intensity ≥ 90%; and the storage standard parameters of a fault-related data block are a gas concentration of 0.9% (meets the early warning standard) and an equipment load intensity of 85% (does not meet the early warning standard), then the number of matched parameter items is 1. Substituting these into the formula, the parameter matching ratio is 1 ÷ 2 × 100% = 50%.

[0059] A preset matching percentage threshold needs to be set, for example, setting the matching percentage threshold to 60%. Determine if the parameter matching percentage is greater than the matching percentage threshold: if it is, then the fault-related data block is identified as target warning data; if it is not, then the data block does not meet the warning focus and is not included in the target warning data.

[0060] If the gas concentration in the storage standard parameters of a certain fault-related data block is 0.9% (compliant) and the equipment load intensity is 92% (compliant), then the number of matched parameters is 2, and the parameter matching ratio = 2 ÷ 2 × 100% = 100%. Since this is greater than the matching ratio threshold of 60%, the fault-related data block is determined to be the target warning data.

[0061] The comprehensive early warning coefficient is obtained by processing and analyzing the content of the early warning focus, the target early warning data, and the early warning time threshold. The specific steps include: Obtain the parameter feature matching degree between the warning focus content and the target warning data; The first warning coefficient corresponding to the target warning data is obtained by summing up the matching degree of each parameter feature of the warning focus and the target warning data. Obtain the monitoring time node corresponding to the target warning data, and calculate the second warning coefficient corresponding to the target warning data by the difference between the monitoring time node and the warning time threshold of the warning content. The comprehensive early warning coefficient is obtained based on the first and second early warning coefficients.

[0062] First, it is necessary to obtain the parameter feature matching degree between the warning focus content and the target warning data. The parameter feature matching degree refers to the degree of fit between each parameter in the target warning data and the corresponding parameter in the warning focus content. For example, if the warning focus content includes two parameters, gas concentration and equipment load intensity, the corresponding parameter feature matching degree quantifies the fit between these two parameters. For example, the matching degree for gas concentration is 0.9, and the matching degree for equipment load intensity is 0.8.

[0063] The first warning coefficient corresponding to the target warning data is obtained by summing the matching degrees of each parameter feature. The first warning coefficient = the sum of the matching degrees of each parameter feature. Substituting the data, we get the first warning coefficient = 0.9 + 0.8 = 1.7. The first warning coefficient reflects the risk matching strength between the target warning data and the content of concern in the warning; the higher the value, the stronger the matching degree.

[0064] It is necessary to obtain the monitoring time nodes corresponding to the target early warning data, and at the same time, clarify the early warning time thresholds for the content of concern. The monitoring time node is the time when the target early warning data is collected; for example, the monitoring time node for a target's early warning data is 9:00 AM. The early warning time threshold is the required advance warning time; for example, if an early warning is required one hour in advance, the corresponding early warning time threshold is 10:00 AM, meaning the early warning needs to be issued before 10:00 AM. Then, the difference between the monitoring time node and the early warning time threshold is calculated to obtain the second early warning coefficient. Second early warning coefficient = Early warning time threshold - Monitoring time node (in hours). Substituting the time into the example, we get the second early warning coefficient = 10 - 9 = 1. The second early warning coefficient reflects the urgency of the early warning; the smaller the value, the closer to the warning deadline, and the higher the urgency.

[0065] The comprehensive early warning coefficient is obtained by combining the first and second early warning coefficients. The comprehensive early warning coefficient = first early warning coefficient × 0.6 + (2 - second early warning coefficient) × 0.4, where 2 is a reasonable upper limit for the second early warning coefficient. Subtracting the second early warning coefficient from 2 allows for a higher coefficient percentage for situations with greater urgency. Substituting the previous values, we get the comprehensive early warning coefficient = 1.7 × 0.6 + (2 - 1) × 0.4 = 1.02 + 0.4 = 1.42. The comprehensive early warning coefficient serves as the core basis for subsequent tiered early warning output. A higher value indicates a higher degree of combined risk and urgency, requiring a higher level of early warning response measures.

[0066] A safety early warning system based on coal mine sensing equipment includes: Acquisition module: Acquires real-time environmental parameters of the coal mine operating environment and equipment operation data of target equipment in the coal mine, and acquires safety early warning requirements for coal mine equipment; The first analysis module analyzes real-time environmental parameters to obtain an environmental risk assessment window, analyzes the multi-dimensional operating characteristics of equipment operation data to obtain the equipment failure risk level, and analyzes the environmental dynamic parameters in the environmental risk assessment window to obtain the corresponding early warning association level within the environmental risk assessment window. Construction module: Construct a fault feature library based on historical fault data of target equipment in coal mines, construct a real-time monitoring model based on real-time environmental parameters and equipment operation data, and analyze the fault feature library and real-time monitoring model to obtain fault-related data blocks; The second analysis module: Based on the safety early warning requirements, it obtains the content of concern for early warning and the early warning time threshold, and analyzes the data blocks related to the faults to obtain the target early warning data. Processing module: Processes and analyzes the warning content, target warning data, and warning time threshold to obtain a comprehensive warning coefficient; Output module: Based on the comprehensive early warning coefficient, outputs the target early warning data corresponding to the safety early warning requirements.

[0067] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0068] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0069] 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 them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A safety early warning method based on coal mine sensing equipment, characterized in that, The method includes the following steps: To obtain real-time environmental parameters of the coal mine operating environment and equipment operation data of target equipment in the coal mine, and to obtain safety early warning requirements for coal mine equipment; The environmental risk assessment window is obtained by analyzing real-time environmental parameters, the equipment failure risk level is obtained by analyzing the multi-dimensional operating characteristics of equipment operation data, and the corresponding early warning association level within the environmental risk assessment window is obtained by analyzing the environmental dynamic parameters within the environmental risk assessment window. A fault feature library is constructed based on historical fault data of target equipment in coal mines. A real-time monitoring model is constructed based on real-time environmental parameters and equipment operation data. Fault-related data blocks are obtained by analyzing the fault feature library and the real-time monitoring model. Based on the safety early warning requirements, the content to be concerned and the warning time threshold are obtained. The target early warning data is obtained by analyzing the data blocks related to the faults and the content to be concerned in the early warning. The comprehensive early warning coefficient is obtained by processing and analyzing the content of the early warning, the target early warning data, and the early warning time threshold. Based on the comprehensive early warning coefficient, target early warning data corresponding to the safety early warning needs are output in a graded manner.

2. The safety early warning method based on coal mine sensing equipment according to claim 1, characterized in that, The environmental risk assessment window is obtained by analyzing real-time environmental parameters, specifically including the following steps: The time-series environmental data of the coal mine operation environment is obtained based on real-time environmental parameters; wherein, the time-series environmental data includes basic environmental characteristics, dynamic interference characteristics, and time period characteristics; The time-series environmental data of the coal mine operation environment is input into the environmental risk prediction model. The environmental risk prediction model outputs the risk prediction value of the coal mine operation environment based on the time-series environmental data. Each risk prediction value corresponds to a time window. The risk prediction value includes the predicted gas concentration, the predicted dust content, and the predicted temperature fluctuation range. The environmental risk prediction model obtains the assessment priority coefficient corresponding to the time window based on the risk prediction value, and then obtains the environmental risk assessment window based on the assessment priority coefficient.

3. A safety early warning method based on coal mine sensing equipment according to claim 2, characterized in that, The equipment failure risk level is determined by analyzing the multi-dimensional operational characteristics of equipment operation data, specifically including the following steps: Multi-dimensional operational characteristics include equipment load intensity, runtime, fault history, and component wear level; Set up an equipment evaluation model, which includes multi-dimensional operating characteristic weights, including load intensity weight, runtime weight, fault record weight, and wear level weight; The multi-dimensional operational characteristics of equipment operation data are input into the equipment evaluation model. The equipment evaluation model performs corresponding feature weighting and accumulation on the multi-dimensional operational characteristics of the equipment operation data according to the multi-dimensional operational characteristics, and obtains and outputs the fault risk coefficient corresponding to the equipment operation data. Set a preset number of fault risk level ranges, and each fault risk level range corresponds to a fault risk level. The fault risk coefficient corresponding to the equipment operation data is compared with the fault risk level range, and the fault risk level corresponding to the fault risk level range where the fault risk coefficient is located is marked as the equipment fault risk level.

4. A safety early warning method based on coal mine sensing equipment according to claim 3, characterized in that, The environmental dynamic parameters within the environmental risk assessment window are analyzed to obtain the corresponding early warning correlation level within the window. This process includes the following steps: The environmental dynamic parameters include the rate of change of environmental parameters and the frequency of parameter fluctuations within the environmental risk assessment window; Obtain the number of fault risk level intervals, and divide the environmental parameter safety threshold range based on the number of level intervals to obtain the environmental dynamic interval; By comparing the frequency of parameter fluctuations within the environmental risk assessment window with the environmental dynamic range, the environmental dynamic range in which the frequency of parameter fluctuations is located can be obtained. The fault risk level corresponding to the environmental dynamic range is marked as the corresponding early warning association level within the environmental risk assessment window.

5. A safety early warning method based on coal mine sensing equipment according to claim 4, characterized in that, A fault feature database is constructed based on historical fault data of target equipment in coal mines, specifically including the following steps: The historical fault data of the target equipment in the coal mine is divided into historical fault data blocks; Fault feature nodes are generated based on historical fault data blocks, and feature identification values ​​corresponding to the fault feature nodes are obtained by extracting features from the historical fault data blocks. A fault feature library for coal mine target equipment is constructed based on fault feature nodes and their corresponding feature identifier values.

6. A safety early warning method based on coal mine sensing equipment according to claim 5, characterized in that, A real-time monitoring model is constructed based on real-time environmental parameters and equipment operation data, specifically including the following steps: Real-time environmental parameters and equipment operation data are divided into real-time monitoring data blocks; Real-time monitoring nodes are generated based on real-time monitoring data blocks, and the monitoring identifier values ​​corresponding to the real-time monitoring nodes are obtained by feature extraction from the real-time monitoring data blocks. A real-time monitoring model is constructed based on the real-time monitoring nodes and the monitoring identifier values ​​corresponding to the real-time monitoring nodes.

7. A safety early warning method based on coal mine sensing equipment according to claim 6, characterized in that, The fault-related data blocks are obtained through analysis based on the fault feature library and real-time monitoring model, specifically including the following steps: The fault feature nodes in the fault feature library are compared with the corresponding identifier values ​​of the real-time monitoring nodes in the real-time monitoring model. If there is a real-time monitoring node in the real-time monitoring model whose matching degree with the identifier value corresponding to the fault feature node exceeds a preset threshold, then the real-time monitoring node is marked as an associated node, and the real-time monitoring data block corresponding to the associated node is recorded as a fault associated data block.

8. A safety early warning method based on coal mine sensing equipment according to claim 7, characterized in that, The target warning data is obtained by analyzing the warning focus and fault correlation data blocks, specifically including the following steps: The standard parameters of the warning content are compared with the standard parameters of the storage of the fault-related data block to obtain the parameter matching ratio between the storage standard parameters and the warning standard parameters. If the percentage of matching between the storage standard parameters and the early warning standard parameters is greater than a preset matching percentage threshold, then the fault-related data block corresponding to the storage standard parameters is determined to be the target early warning data.

9. A safety early warning method based on coal mine sensing equipment according to claim 8, characterized in that, The comprehensive early warning coefficient is obtained by processing and analyzing the content of the early warning focus, the target early warning data, and the early warning time threshold. The specific steps include: Obtain the parameter feature matching degree between the warning focus content and the target warning data; The first warning coefficient corresponding to the target warning data is obtained by summing up the matching degree of each parameter feature of the warning focus and the target warning data. Obtain the monitoring time node corresponding to the target warning data, and calculate the second warning coefficient corresponding to the target warning data by the difference between the monitoring time node and the warning time threshold of the warning content. The comprehensive early warning coefficient is obtained based on the first and second early warning coefficients.

10. A safety early warning system based on coal mine sensing equipment, applied to the safety early warning method based on coal mine sensing equipment as described in any one of claims 1 to 9, characterized in that, include: Acquisition module: Acquires real-time environmental parameters of the coal mine operating environment and equipment operation data of target equipment in the coal mine, and acquires safety early warning requirements for coal mine equipment; The first analysis module analyzes real-time environmental parameters to obtain an environmental risk assessment window, analyzes the multi-dimensional operating characteristics of equipment operation data to obtain the equipment failure risk level, and analyzes the environmental dynamic parameters in the environmental risk assessment window to obtain the corresponding early warning association level within the environmental risk assessment window. Construction module: Construct a fault feature library based on historical fault data of target equipment in coal mines, construct a real-time monitoring model based on real-time environmental parameters and equipment operation data, and analyze the fault feature library and real-time monitoring model to obtain fault-related data blocks; The second analysis module: Based on the safety early warning requirements, it obtains the content of concern for early warning and the early warning time threshold, and analyzes the data blocks related to the faults to obtain the target early warning data. Processing module: Processes and analyzes the warning content, target warning data, and warning time threshold to obtain a comprehensive warning coefficient; Output module: Based on the comprehensive early warning coefficient, outputs the target early warning data corresponding to the safety early warning requirements.

Images