Mine ventilation intelligent management system based on internet of things

Abstract

This invention relates to the field of mine ventilation management technology and discloses an IoT-based intelligent management system for mine ventilation, comprising: a sensing layer including sensors underground for collecting ventilation parameters; and an edge computing module deployed in pre-defined verification zones underground, each verification zone configured with one or more edge computing gateways. The edge computing gateways are communicatively connected to the sensors within their jurisdiction and are used to perform distributed data verification. The edge computing gateways have a built-in verification engine, which includes a time-series verifier used to perform univariate time-series analysis on the current readings of each sensor at fixed intervals to determine whether there are significant jumps. This invention avoids the computational resource accumulation and response latency problems caused by fixed-period full-scale verification and reduces the normal load on edge computing nodes.

Description

Technical Field

[0001] This invention relates to the field of mine ventilation management technology, and more specifically to an intelligent mine ventilation management system based on the Internet of Things. Background Technology

[0002] Mine ventilation systems are a core component in ensuring safe coal mine production, and their reliable operation relies on real-time and accurate monitoring of key parameters such as wind speed, wind pressure, and methane concentration. Existing IoT-based mine ventilation management systems generally employ a centralized data processing architecture, collecting ventilation parameters through numerous sensors deployed underground and transmitting them to a central platform for analysis and decision-making via an industrial ring network. However, the harsh underground environment, characterized by high humidity, high dust levels, and electromagnetic interference, can cause sensors to experience zero-point drift, measurement deviations, or even complete failure after prolonged operation. Traditional systems lack effective online self-verification mechanisms for sensor data, making it difficult to distinguish whether measurement anomalies stem from sensor malfunctions or actual changes in ventilation conditions. This can lead to the central platform making ventilation network calculations and control decisions based on erroneous data, resulting in anything from wasted energy to serious safety accidents.

[0003] Existing systems typically perform full data verification on all sensors at fixed intervals, resulting in high computational resource consumption, high response latency, and difficulty in accurately locating the fault source when multiple verification rules simultaneously trigger anomaly flags. Furthermore, the existing verification process lacks a hierarchical scheduling mechanism, failing to dynamically adjust the verification depth based on anomaly type. This leads to misclassifying real physical events such as damper movements and gas outbursts as sensor faults, or misclassifying multi-sensor coupled faults as single faults, reducing the system's reliability and usability. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent management system for mine ventilation based on the Internet of Things (IoT), thereby solving the above-mentioned technical problems:

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] An IoT-based intelligent management system for mine ventilation includes: a sensing layer, comprising sensors underground for collecting ventilation parameters; and an edge computing module deployed in pre-defined verification zones underground, each verification zone being configured with one or more edge computing gateways, the edge computing gateways being communicatively connected to sensors within their jurisdiction for performing distributed data verification.

[0007] The edge computing gateway has a built-in verification engine, which includes: a timing verifier, used to perform univariate timing analysis on the current readings of each sensor at fixed intervals to determine whether there are significant jumps; a spatial verifier, which is triggered when the timing verifier determines that there are significant jumps, and performs multi-sensor joint verification on the node or loop where the jump sensor is located based on the ventilation network topology; and a coupling verifier, which is triggered when the spatial verifier cannot uniquely identify the abnormal sensor, and performs in-depth verification based on the material and energy conservation relationships between cross-type sensors.

[0008] As a further technical solution, the system also includes: a data transmission module, comprising an industrial Ethernet ring network and a wireless communication network, used to establish data connections between the sensing layer and the central platform and execution module; a central platform, communicatively connected to the edge computing module, used to receive verified ventilation parameters and perform real-time ventilation network calculations, safety situation assessments, and intelligent control decisions; and an execution module, including a main ventilator, local ventilators, automatic dampers, and automatic windows, used to receive control commands from the central platform and the edge computing module and execute ventilation control. The ventilation parameters include wind speed, wind pressure, gas concentration, temperature, and dust concentration.

[0009] As a further technical solution, the timing verifier performs the following verification process: For each sensor, a sliding window is preset to store a predetermined number of its most recent historical reading sequences, and the mean and standard deviation of the readings within the sliding window are calculated; the reading at the current sampling time is compared with the mean, and the standardized deviation is calculated; if the standardized deviation is greater than a preset timing jump threshold, it is determined that the sensor has undergone a significant jump, the sensor is marked as a state to be verified, and the jump direction is recorded;

[0010] If the standardized deviation is less than or equal to the timing jump threshold, the sensor remains in normal condition and the spatial verifier is not triggered.

[0011] As a further technical solution, the verification process performed by the spatial verifier on the sensor marked as to be verified includes:

[0012] Step A: Determine the ventilation network node where the sensor to be verified is located, denoted as node j. Node j is connected to m branches, and each branch is equipped with a wind speed sensor. The measured air volume values ​​are Q1, Q2...Qm, where the inflow node is positive and the outflow node is negative.

[0013] Step B: Calculate the algebraic sum of airflow at node j, denoted as the total node residual. :

[0014]

[0015] Step C: [The text appears to be incomplete and contains several grammatical errors. A more accurate translation would require the full context With respect to the preset spatial tolerance threshold Comparison:

[0016] like The spatial consistency was determined to be satisfactory.

[0017] like The space consistency is determined to be violated.

[0018] As a further technical solution, the verification process performed by the spatial verifier on the sensors marked as to be verified also includes:

[0019] Step D: When spatial consistency is violated, calculate the deviation of each branch's airflow from its historical baseline value:

[0020]

[0021] in This is the historical baseline value for the air volume of the k-th branch;

[0022] Step E: Compare the deviations of each branch. Total residual at nodes Relationship:

[0023] The deviation of each branch The absolute value of the node residuals The comparison is based on the following criteria:

[0024] If there exists a unique branch p that satisfies the following three conditions:

[0025] Condition one: ,and ,in It is a preset, tiny positive number;

[0026] Condition 2: The sensor on branch p is currently in a state of pending verification;

[0027] Condition 3: For all All have ;

[0028] If the sensor in branch p is determined to be the only source of failure, its status is updated to fault, and subsequent verification is terminated.

[0029] If the above conditions are not met, but multiple branches exist... Greater than or equal to If so, the status of the sensors to be verified on these branches is updated to suspicious, and the coupling checker is triggered.

[0030] As a further technical solution, the verification process performed by the coupling checker on the sensor group marked as suspicious includes:

[0031] For the mining face area, the wind speed sensor readings and gas concentration sensor readings are collected at the same time to calculate the working face air volume, which is equal to the wind speed multiplied by the roadway cross-sectional area.

[0032] Based on the principle of material conservation, the working face air volume is multiplied by the gas concentration to obtain the calculated gas emission rate. Based on historical data regression or the gas emission rate monitoring value of the adjacent area, the expected gas emission rate of the current working face is obtained.

[0033] The relative deviation is calculated by dividing the absolute value of the difference between the calculated gas emission and the expected gas emission by the sum of the expected gas emission and a zero-prevention constant.

[0034] The relative deviation is compared with a preset physical coupling threshold:

[0035] If the relative deviation is less than or equal to the physical coupling threshold, the physical coupling relationship is determined to be established, the state of the suspicious sensor is restored to normal, and this event is recorded as a false alarm of the timing rule.

[0036] If the relative deviation is greater than the physical coupling threshold, the self-diagnostic status information of the sensor is read again: if the self-diagnostic status shows an abnormality, it is determined to be a sensor fault, and the status is updated to fault; if the self-diagnostic status shows a normal status, it is determined to be a real physical event, the status is updated to physical event, and the event is reported to the central platform.

[0037] As a further technical solution, the verification engine also includes a dynamic threshold adjuster, which is used to adaptively update the temporal jump threshold, spatial tolerance threshold and physical coupling threshold based on statistical information of historical verification results.

[0038] The dynamic threshold adjuster performs the following adjustment process:

[0039] The false alarm rate of each sensor within a preset time window is obtained. If the false alarm rate is greater than the preset upper limit of the false alarm rate, the timing jump threshold of the sensor is increased. If the false alarm rate is less than the preset lower limit of the false alarm rate and the missed alarm rate increases, the timing jump threshold of the sensor is decreased.

[0040] The spatial tolerance threshold is adjusted based on the continuous normal operation time of the entire mine: when the continuous fault-free operation time exceeds the first time threshold, the spatial tolerance threshold is reduced; when a real fault or physical event is detected, the spatial tolerance threshold is immediately restored to the initial tolerance value.

[0041] The physical coupling threshold is adaptively set based on the historical standard deviation of gas emission from the working face.

[0042] The beneficial effects of this invention are:

[0043] (1) The present invention sets up a timing checker, a spatial checker and a coupling checker in the verification engine built into the edge computing gateway, and performs the verification in the progressive order of timing verification, spatial verification and physical coupling verification to form an event-driven dynamic scheduling mechanism. Spatial verification is triggered only when there is a significant jump in the timing verification judgment, and coupling verification is triggered only when the spatial verification cannot uniquely determine the abnormal sensor. Therefore, the present invention avoids the problem of computing resource accumulation and response delay caused by fixed period full verification, and reduces the normal load of edge computing nodes.

[0044] (2) The present invention can uniquely identify faulty sensors and terminate subsequent verification by using precise judgment conditions based on node air volume algebra and deviation from each branch in the spatial verification device, or trigger coupled verification for in-depth verification when multiple sensors are abnormal. At the same time, by comparing the spatial consistency of upstream and downstream nodes, it can effectively distinguish between sensor faults and real physical events, and avoid misjudging normal physical changes such as damper operation and gas outburst as sensor faults.

[0045] (3) The present invention enables the verification engine to adapt to different mine conditions and sensor aging characteristics by using a dynamic threshold adjuster to adaptively update the timing jump threshold based on the historical false alarm rate, adjust the spatial tolerance threshold based on the continuous normal operation duration, and set the physical coupling threshold based on the standard deviation of gas emission fluctuation. Attached Figure Description

[0046] The invention will now be further described with reference to the accompanying drawings.

[0047] Figure 1 This is a framework diagram of the IoT-based intelligent management system for mine ventilation in this invention. Detailed Implementation

[0048] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0049] Please see Figure 1As shown, an IoT-based intelligent management system for mine ventilation includes: a sensing layer, comprising sensors underground for collecting ventilation parameters. Specifically, the sensing layer can consist of multiple sensors deployed in mine roadways, working faces, and return air shafts, used to collect underground ventilation parameters in real time, including but not limited to wind speed, wind pressure, methane concentration, temperature, and dust concentration. An edge computing module is deployed in pre-defined verification zones underground. Each verification zone is configured with one or more edge computing gateways based on sensor distribution density and network topology. Each edge computing gateway establishes a data connection with the sensors within its jurisdiction via fieldbus or wireless communication, used to receive raw sensor data and perform distributed data verification locally, thereby avoiding network congestion and response delays caused by uploading all raw data to a central platform.

[0050] The edge computing gateway has a built-in verification engine. This engine, as a core data processing unit, runs as software or firmware within the edge computing gateway's processor and is used for real-time quality assessment and anomaly diagnosis of sensor data within its jurisdiction. The verification engine includes: a time-series verifier, configured to perform univariate time-series analysis on the current reading of each sensor at a fixed period (e.g., once per second). Specifically, it determines whether there is a significant jump by comparing the current reading with the mean and standard deviation within the sensor's historical sliding window. When the time-series verifier determines that a sensor reading has experienced a significant jump, it marks the sensor as pending verification and immediately triggers the spatial verifier. The spatial verifier, once triggered, performs multi-sensor joint verification based on the ventilation network topology (including the connection relationships between roadway branches and nodes, airflow conservation constraints, and air pressure balance constraints). It calculates the algebraic sum of node airflow and compares it with a preset spatial tolerance threshold to determine whether spatial consistency is satisfied. If the spatial verifier can uniquely identify a sensor as the fault source, it directly outputs a fault flag and terminates the verification process. The coupling checker is triggered only when the spatial checker cannot uniquely identify the abnormal sensor (e.g., multiple sensors exhibiting deviations simultaneously or the relationship between deviation and node residuals is unclear). This coupling checker performs in-depth verification based on material conservation relationships (e.g., the product of gas emission and airflow / gas concentration) and energy conservation relationships (e.g., the square relationship between wind pressure and wind speed) between different sensor types. By comparing the relative deviation between calculated and expected values, and combining this with the sensor's self-diagnostic status information, it ultimately determines whether the root cause of the anomaly is a sensor malfunction or a real physical event. In this embodiment, through the progressive collaborative work of the timing checker, spatial checker, and coupling checker, the verification engine achieves hierarchical and on-demand verification of sensor data, avoiding the waste of computational resources caused by fixed-cycle full-volume verification, while simultaneously improving the accuracy of fault location and the precision of physical event identification.

[0051] The system also includes a data transmission module, comprising an industrial Ethernet ring network and a wireless communication network. The industrial Ethernet ring network serves as the backbone network deployed in the main underground roadways and chambers, employing a redundant ring network topology to ensure communication reliability. The wireless communication network includes, but is not limited to, wireless communication standards such as WiFi, 4G, and 5G, used to cover areas where fixed sensors are difficult to wire or to transmit data for mobile inspection equipment. The data transmission module establishes a bidirectional data connection between the sensing layer, the central platform, and the execution module: on one hand, it uploads the raw data collected by the sensing layer and the data verified by the edge computing module to the central platform; on the other hand, it sends the control commands generated by the central platform to the execution module. The central platform consists of a server cluster and supporting data storage and computing resources. This central platform establishes a communication connection with the edge computing module through the data transmission module, and is used to receive ventilation parameters verified by the edge computing module and accompanied by status markers (such as normal, pending verification, suspicious, fault, physical event). The central platform integrates a real-time ventilation network calculation engine, a safety situation assessment model, and an intelligent control decision-making algorithm. The real-time ventilation network calculation engine, based on received reliable ventilation parameters, uses ventilation network calculation algorithms (such as the Scott-Hinsley method or Newton's iteration method) to calculate the airflow and resistance distribution of each branch of the entire mine. The safety situation assessment model compares the calculation results with preset safety thresholds (such as upper and lower limits of wind speed and gas concentration alarm values) to determine whether there are abnormal states such as insufficient airflow, excessive wind speed, or gas accumulation. When an anomaly is detected, the intelligent control decision-making algorithm automatically generates control commands based on the mine's production plan and energy consumption optimization targets, including adjusting the main ventilation fan speed, adjusting the air vent opening, and switching the air door status. The execution module, including the main ventilation fan, local ventilation fans, automatic air doors, and automatic air windows, receives control commands from the central platform and edge computing module and executes ventilation control. The ventilation parameters include wind speed, wind pressure, gas concentration, temperature, and dust concentration.

[0052] The timing checker performs the following check process: For the i-th sensor, a sliding window of length L is preset to store its most recent L historical reading sequences. ;

[0053] Calculate the mean of the readings within the sliding window. and standard deviation ;

[0054] Current sampling time readings The standardized deviation is calculated by comparing it with the mean:

[0055]

[0056] in To prevent the division into zero small constants;

[0057] like Greater than the preset timing transition threshold If a significant change occurs in the i-th sensor, the sensor is marked as being in a state to be verified, and the sign of the change direction is recorded. ;

[0058] If the standardized deviation is less than or equal to the timing jump threshold, the sensor remains in normal condition and the spatial verifier is not triggered.

[0059] The verification process performed by the spatial verifier for sensors marked as to be verified includes:

[0060] Step A: Determine the ventilation network node where the sensor to be verified is located, denoted as node j. Node j is connected to m branches, and each branch is equipped with a wind speed sensor. The measured air volume values ​​are Q1, Q2...Qm, where the inflow node is positive and the outflow node is negative.

[0061] Step B: Calculate the algebraic sum of airflow at node j, denoted as the total node residual. :

[0062]

[0063] Step C: [The text appears to be incomplete and contains several grammatical errors. A more accurate translation would require the full context With respect to the preset spatial tolerance threshold Comparison:

[0064] like The spatial consistency was determined to be satisfactory.

[0065] like The space consistency is determined to be violated.

[0066] As a further technical solution, the verification process performed by the spatial verifier on the sensors marked as to be verified also includes:

[0067] Step D: When spatial consistency is violated, calculate the deviation of each branch's airflow from its historical baseline value:

[0068]

[0069] in This is the historical baseline value for the air volume of the k-th branch;

[0070] Step E: Compare the deviations of each branch. Total residual at nodes Relationship:

[0071] The deviation of each branch The absolute value of the node residuals The comparison is based on the following criteria:

[0072] If there exists a unique branch p that satisfies the following three conditions:

[0073] Condition one: ,and ,in It is a preset, tiny positive number;

[0074] Condition 2: The sensor on branch p is currently in a state of pending verification;

[0075] Condition 3: For all All have ;

[0076] If the sensor in branch p is determined to be the only source of failure, its status is updated to fault, and subsequent verification is terminated.

[0077] If the above conditions are not met, but multiple branches exist... Greater than or equal to If so, the status of the sensors to be verified on these branches is updated to suspicious, and the coupling checker is triggered.

[0078] It should also be noted that if spatial consistency passes, but the algebraic sum of airflow at the upstream or downstream node of the branch where the sensor to be verified is located also exceeds its corresponding spatial tolerance threshold, and the direction consistency is violated (i.e., the algebraic sum of each node is either positive or negative), then it is determined that a real physical event has occurred, the relevant sensor status is updated to a physical event, and the coupling checker is not triggered.

[0079] The verification process performed by the coupling checker for sensor groups marked as suspicious includes:

[0080] Specifically, for the mining face area, wind speed sensor readings are collected at the same time. Compared with gas concentration sensor readings Calculate the air volume at the working face , where A is the cross-sectional area of ​​the tunnel;

[0081] Based on the principle of material conservation, a gas emission estimation model is established:

[0082]

[0083] Simultaneously, based on historical data regression or monitoring values ​​of gas emission from neighboring areas, the expected gas emission from the current working face is obtained. ;

[0084] Calculate the relative deviation:

[0085]

[0086] To prevent the division into zero small constants;

[0087] Will With preset physical coupling threshold Comparison:

[0088] like If the physical coupling relationship is determined to be valid, the state of the suspicious sensor is restored to normal, and this event is recorded as a false alarm of the timing rule for subsequent adaptive adjustment of the timing threshold.

[0089] like Further, the self-diagnostic status information of the sensor is read, including the sensor's internal temperature, power supply voltage, and self-test code.

[0090] If the self-diagnostic status shows an abnormality, it is determined to be a sensor malfunction, and the status is updated to malfunction.

[0091] If the self-diagnosis status shows normal, it is determined to be a real physical event, the status is updated to physical event, and the event is reported to the central platform layer. The central platform feeds back the results of manual review after the physical event is reported to the dynamic threshold adjuster.

[0092] The verification engine also includes a dynamic threshold adjuster, which is used to adaptively update the temporal jump threshold, spatial tolerance threshold and physical coupling threshold based on statistical information of historical verification results.

[0093] The dynamic threshold adjuster performs the following adjustment process:

[0094] Define the false alarm rate of the i-th sensor within the time window T:

[0095]

[0096] in This represents the number of times the timing consistency verification module triggers the verification process. This refers to the number of times that are ultimately determined to be normal by the physical coupling verification module.

[0097] like Greater than the preset false alarm rate limit Then increase the timing jump threshold of the sensor:

[0098]

[0099] in To adjust the step size coefficient, For the target false alarm rate, For the updated timing transition threshold, This is the old timing transition threshold.

[0100] like If the false alarm rate is below the preset lower limit and the false alarm rate increases, then reduce the false alarm rate. ;

[0101] The spatial tolerance threshold Adjustments were made based on the total continuous normal operating time of the entire mine.

[0102] Define continuous fault-free operation time This duration represents the length of time the system has been operating normally since the last detected real fault or physical event. When the first duration threshold is exceeded, it indicates that the system is in a long-term stable operating condition, and the contraction coefficient is gradually reduced according to the preset value. When a real fault is detected (i.e., the sensor is judged to be in a fault state) or a real physical event (such as damper operation, gas outburst, etc.), it indicates a significant change in operating conditions or a sensor malfunction. To avoid misjudgment due to overly strict thresholds under non-steady-state conditions, the threshold will be... Immediately restore to the initial tolerance threshold;

[0103] The physical coupling threshold Adaptive settings are based on the historical standard deviation of gas emission rates at the working face:

[0104]

[0105] in This represents the standard deviation of gas emission from the working face within a historical time window. It reflects the natural fluctuation range of gas emission; a larger standard deviation indicates higher instability in gas emission. This is a proportionality coefficient used to control the contribution of the standard deviation to the threshold; it can be obtained through experimental fitting. The base threshold represents the minimum permissible relative deviation under stable gas emission conditions. Through the above adaptive setting, when the gas emission at the working face fluctuates significantly, the physical coupling threshold automatically increases to avoid misjudging normal gas emission fluctuations as sensor malfunctions or abnormal events; when the gas emission is stable, the threshold automatically decreases to improve the sensitivity of sensor anomaly detection.

[0106] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A mine ventilation intelligent management system based on the Internet of Things, characterized in that, The system includes: The sensing layer includes sensors used downhole to collect ventilation parameters; An edge computing module is deployed in each pre-set verification zone underground. Each verification zone is configured with one or more edge computing gateways. The edge computing gateways are communicatively connected to sensors within their jurisdiction to perform distributed data verification. The edge computing gateway has a built-in verification engine, which includes: A timing checker is used to perform univariate timing analysis on the current readings of each sensor at fixed intervals to determine whether there are significant jumps. The spatial verifier is triggered when the timing verifier determines that there is a significant jump, and performs multi-sensor joint verification on the node or loop where the jump sensor is located based on the ventilation network topology. The coupling checker is triggered when the spatial checker cannot uniquely identify the abnormal sensor, and performs in-depth verification based on the material and energy conservation relationships between cross-type sensors.

2. The intelligent mine ventilation management system based on the Internet of Things according to claim 1, characterized in that, The system also includes: The data transmission module, including an industrial Ethernet ring network and a wireless communication network, is used to establish data connection relationships between the sensing layer, the central platform, and the execution module. The central platform communicates with the edge computing module to receive verified ventilation parameters and perform real-time calculations of the ventilation network, safety situation assessments, and intelligent control decisions. The execution module, including the main ventilation fan, local ventilation fan, automatic damper and automatic window, is used to receive control instructions from the central platform and edge computing module and execute ventilation control.

3. The intelligent mine ventilation management system based on the Internet of Things according to claim 2, characterized in that, The ventilation parameters include wind speed, wind pressure, gas concentration, temperature, and dust concentration.

4. The intelligent mine ventilation management system based on the Internet of Things according to claim 3, characterized in that, The timing checker performs the following check process: For each sensor, a sliding window is preset to store a predetermined number of its most recent historical reading sequences, and the mean and standard deviation of the readings within the sliding window are calculated. The reading at the current sampling time is compared with the mean, and the standardization deviation is calculated. If the standardization deviation is greater than the preset time jump threshold, the sensor is determined to have undergone a significant jump, the sensor is marked as a state to be verified, and the jump direction is recorded. If the standardized deviation is less than or equal to the timing jump threshold, the sensor remains in normal condition and the spatial verifier is not triggered.

5. The intelligent mine ventilation management system based on the Internet of Things according to claim 4, characterized in that, The verification process performed by the spatial verifier for sensors marked as to be verified includes: Step A: Determine the ventilation network node where the sensor to be verified is located, denoted as node j. Node j is connected to m branches, and each branch is equipped with a wind speed sensor, measuring airflow values ​​Q1, Q2...Q m , where inflow nodes are positive and outflow nodes are negative; Step B: Calculate the algebraic sum of airflow at node j, denoted as the total node residual. : ； Step C: [The text appears to be incomplete and contains several grammatical errors. A more accurate translation would require the full context With respect to the preset spatial tolerance threshold Comparison: like The spatial consistency was determined to be satisfactory. like The space consistency is determined to be violated.

6. The intelligent mine ventilation management system based on the Internet of Things according to claim 5, characterized in that, The verification process performed by the spatial verifier for sensors marked as to be verified also includes: Step D: When spatial consistency is violated, calculate the deviation of each branch's airflow from its historical baseline value: ； in This is the historical baseline value for the air volume of the k-th branch; Step E: Compare the deviations of each branch. Total residual at nodes Relationship: The deviation of each branch The absolute value of the node residuals The comparison is based on the following criteria: If there exists a unique branch p that satisfies the following three conditions: Condition one: ,and ,in It is a preset, tiny positive number; Condition 2: The sensor on branch p is currently in a state of pending verification; Condition 3: For all All have ; If the sensor in branch p is determined to be the only source of failure, its status is updated to fault, and subsequent verification is terminated. If the above conditions are not met, but multiple branches exist... Greater than or equal to If so, the status of the sensors to be verified on these branches is updated to suspicious, and the coupling checker is triggered.

7. The intelligent mine ventilation management system based on the Internet of Things according to claim 6, characterized in that, The verification process performed by the coupling checker for sensor groups marked as suspicious includes: For the mining face area, the wind speed sensor readings and gas concentration sensor readings are collected at the same time to calculate the working face air volume, which is equal to the wind speed multiplied by the roadway cross-sectional area. Based on the principle of material conservation, the working face air volume is multiplied by the gas concentration to obtain the calculated gas emission rate. Based on historical data regression or the gas emission rate monitoring value of the adjacent area, the expected gas emission rate of the current working face is obtained. The relative deviation is calculated by dividing the absolute value of the difference between the calculated gas emission and the expected gas emission by the sum of the expected gas emission and a zero-prevention constant. The relative deviation is compared with a preset physical coupling threshold: If the relative deviation is less than or equal to the physical coupling threshold, the physical coupling relationship is determined to be established, the state of the suspicious sensor is restored to normal, and this event is recorded as a false alarm of the timing rule. If the relative deviation is greater than the physical coupling threshold, the self-diagnostic status information of the sensor is read again: if the self-diagnostic status shows an abnormality, it is determined to be a sensor fault, and the status is updated to fault; if the self-diagnostic status shows a normal status, it is determined to be a real physical event, the status is updated to physical event, and the event is reported to the central platform.

8. The intelligent mine ventilation management system based on the Internet of Things according to claim 7, characterized in that, The verification engine also includes a dynamic threshold adjuster, used to adaptively update the temporal jump threshold, spatial tolerance threshold, and physical coupling threshold based on statistical information from historical verification results; the dynamic threshold adjuster performs the following adjustment process: The false alarm rate of each sensor within a preset time window is obtained. If the false alarm rate is greater than the preset upper limit of the false alarm rate, the timing jump threshold of the sensor is increased. If the false alarm rate is less than the preset lower limit of the false alarm rate and the missed alarm rate increases, the timing jump threshold of the sensor is decreased. The spatial tolerance threshold is adjusted based on the continuous normal operation time of the entire mine: when the continuous fault-free operation time exceeds the first time threshold, the spatial tolerance threshold is reduced; when a real fault or physical event is detected, the spatial tolerance threshold is immediately restored to the initial tolerance value. The physical coupling threshold is adaptively set based on the historical standard deviation of gas emission from the working face.

Images