A comprehensive management system for smart mine based on internet of things

Abstract

The application discloses a kind of based on internet of things wisdom mine comprehensive management system, system includes data access module, sensor health state judging module, credible data screening module, default risk level judging module, risk correction module, risk propagation partition module, ventilation control strategy generation module, execution and update module, ventilation incentive back-reading module and closed-loop control module.The application introduces sensor health state online hierarchical evaluation into blast post-ventilation control process, modifies risk level according to sensor credibility, and realizes partition linkage control and closed-loop self-checking by combining air flow propagation path.The system can still consider safety and energy saving under the condition of dynamic change of sensor network quality, avoids risk misjudgment caused by sensor anomaly, and improves the intelligent level and robustness of mine ventilation management.

Description

Technical Field

[0001] This invention relates to the field of mine safety and ventilation control technology, and in particular to a smart mine integrated management system based on the Internet of Things. Background Technology

[0002] After blasting operations in mines, a large amount of toxic and harmful gases (such as carbon monoxide and nitrogen oxides) and dust are generated in the working face and return airway. These gases must be diluted and discharged through the ventilation system. Existing mine ventilation management typically adopts the following two methods: full-power ventilation is carried out for a fixed period of time after blasting, and normal ventilation is resumed only when the gas concentration drops below the safe threshold; or the ventilation equipment is started, stopped, or its speed adjusted based on the comparison between the monitoring value of a single gas sensor and its alarm threshold.

[0003] Both of the above methods have the following drawbacks: While the first method is simple to implement, it does not take into account changes in the amount of explosive charge, the passage of time, and the actual gas diffusion state, leading to safety risks of large blasts with small ventilation or energy waste of small blasts with large ventilation. While the second method achieves concentration feedback control, it relies on the assumption that the sensor data adopted by the system is accurate and reliable. However, gas monitoring sensors and wind speed and direction monitoring sensors in mines operate under harsh conditions of high temperature, high humidity, high dust, and strong vibration for extended periods, making them highly susceptible to zero-point drift, sensitivity decay, delayed response, or even complete failure. If a sensor malfunctions and is not detected in time, the system may make incorrect judgments based on erroneous data. For example, a gas sensor may falsely report a high concentration due to drift, leading to unnecessary emergency shutdowns and personnel evacuations; or a sensor may fail to report a fault, resulting in a real safety accident.

[0004] Furthermore, existing ventilation control systems generally employ a binary approach of accepting or discarding sensor data, lacking a hierarchical management and fault-tolerant usage mechanism for sub-health data. When the status of individual sensors in the monitoring network declines, the system often fails to adaptively adjust its data usage strategy, resulting in the waste of useful information. Simultaneously, existing risk assessments are mostly static evaluations based on single-point monitoring values, failing to incorporate the propagation path and diffusion trend of gas along the airflow direction in the roadway, leading to delays in alarms and responses. Summary of the Invention

[0005] The purpose of this section is to outline some aspects of embodiments of the present invention and to briefly describe some preferred embodiments. Simplifications or omissions may be made in this section, as well as in the abstract and title of this application, to avoid obscuring the purpose of these documents; however, such simplifications or omissions should not be construed as limiting the scope of the invention.

[0006] Therefore, to solve the above-mentioned technical problems, the present invention provides the following technical solution: a smart mine integrated management system based on the Internet of Things, comprising a processor, a memory, and functional modules communicatively connected to the processor, characterized in that the functional modules include: The data access module receives blasting parameter data, gas monitoring data, wind speed and direction monitoring data, ventilation equipment operation data, and basic roadway ventilation network data. It then establishes a mapping between monitoring points and roadway numbers, airflow directions, and equipment numbers. The blasting parameter data reflects the initial intensity of the risk source and serves as the starting point for subsequent risk level prediction. The gas monitoring data reflects the diffusion results of toxic and harmful gases in the actual space after blasting. The wind speed and direction monitoring data reflects the real-time state of underground airflow and is a key basis for determining the path along which the risk propagates. The ventilation equipment operation data reflects the system's current control capabilities and available resources. The basic roadway ventilation network data provides topological constraints on the mine space, enabling all types of data to be understood and used within a unified spatial framework. The sensor health status judgment module is used to output sensor health status markers based on gas monitoring data and wind speed and direction monitoring data; the reliable data filtering module is used to output valid gas concentration values ​​and low reliability data markers based on the sensor health status markers. The default risk level judgment module is used to output the default risk level based on the charge amount and the time after blasting; the risk correction module is used to correct the default risk level based on sensor health status markers, low confidence data markers, and monitoring blind zone markers, and output the comprehensive risk level. The risk propagation zoning module is used to divide risk propagation zones based on basic data of the roadway ventilation network and wind speed and direction monitoring data, and output the zone control priority and the comprehensive risk level after zone correction; The ventilation control strategy generation module is used to generate control commands based on the comprehensive risk level and zone control priority after zoning correction. The control commands include the main ventilation fan frequency setpoint and damper opening setpoint. The execution and update module and the ventilation excitation readback module are used to execute the control commands and update the sensor health status markers, monitoring blind zone markers and comprehensive risk levels. The closed-loop control module is used to control the above modules to execute sequentially in a loop to form closed-loop control.

[0007] As a preferred embodiment of the IoT-based smart mine integrated management system described in this invention, the data access module establishes a mapping relationship between blasting parameter data, gas monitoring data, wind speed and direction monitoring data, ventilation equipment operation data, and basic roadway ventilation network data according to the monitoring point location and roadway number, airflow direction, and equipment number, generating a data index table for risk assessment and zoning calculation in subsequent steps. The data index table projects the above heterogeneous data into the same spatial topology, and its essential function is to establish the correspondence between data and spatial location, enabling the processor to quickly locate the roadway location corresponding to a certain monitoring data, the upstream and downstream relationship of that location on the airflow path, and the ventilation control equipment associated with that location in subsequent steps.

[0008] As a preferred embodiment of the IoT-based smart mine integrated management system described in this invention, in the underground mining environment, gas monitoring sensors and wind speed and direction monitoring sensors are exposed to harsh conditions of high temperature, high humidity, dust, and vibration for extended periods, and their performance may degrade or fail in various forms. Therefore, the sensor health status judgment module identifies the health status of the sensors according to the following rules and outputs the corresponding sensor health status flag: The sensor is marked as healthy when its output value is consistent with that of a health sensor in the same monitoring area in the direction of change and the fluctuation amplitude does not exceed the preset fluctuation threshold. The sensor is marked as sub-healthy when its response time exceeds the preset response time threshold, or the deviation between the output value and the health sensor exceeds the preset deviation threshold, or the output value fluctuates continuously when the environment is stable. The sensor is marked as malfunctioning when it experiences data interruption, continuous loss of heart rate, a continuously fixed output value, or an output value that exceeds the range. By classifying sensor status into three levels—healthy, sub-healthy, and malfunctioning—the system achieves hierarchical management of data source quality, preventing abnormal data from directly participating in subsequent risk calculations. This prevents misjudgments of the entire system due to a single sensor failure and provides a reliable basis for subsequent data screening and risk correction.

[0009] As a preferred embodiment of the IoT-based smart mine integrated management system described in this invention, in actual deployment, key monitoring points are typically equipped with a main sensor and a redundant verification sensor; the main sensor is the default data source for control decisions, and the redundant verification sensor is used to provide alternative data when the main sensor malfunctions; the two can be installed in different locations or use different measurement principles to reduce the probability of simultaneous failure due to the same reason. The trusted data filtering module outputs valid gas concentration values ​​according to the following rules. Its core lies in adopting a differentiated usage strategy of selecting the best, downgrading, and labeling data based on the sensor's trust level, as detailed below: When the main sensor's health status is marked as healthy, its data is completely reliable, and the monitoring value collected by the main sensor is directly used as a valid value. When the main sensor's health status is marked as sub-healthy and there is a redundant verification sensor marked as healthy, meaning the data has potential reliability issues but is not completely invalid, the monitoring value collected by the redundant verification sensor is used as a more reliable valid value. When the main sensor's health status is marked as sub-healthy and there is no redundant verification sensor, the monitoring value collected by the main sensor is still used, but a low-reliability data label is added to this monitoring value. That is, the system does not discard this data in this case, but marks it as low-reliability data and continues to use it, thereby achieving a balance between data continuity and data reliability. When the main sensor's health status is marked as invalid, its data is no longer used in calculations, and the monitoring point is marked as a monitoring blind zone. The purpose of the low-reliability data marker is to trigger an upward adjustment of the risk level in subsequent risk correction steps, thereby offsetting data uncertainty with safety redundancy. The purpose of the monitoring blind zone marker is to clearly inform the system that the area currently lacks effective monitoring, triggering a more conservative risk assessment strategy during risk correction, and being included in the self-verification process in subsequent ventilation incentive readback steps. Under dynamic changes in sensor network quality, the system can maintain data availability to the maximum extent while avoiding low-quality data from misleading judgment results, thereby improving the system's fault tolerance.

[0010] As a preferred embodiment of the IoT-based smart mine integrated management system described in this invention, the default risk level judgment module queries a preset comparison relationship based on the combination relationship between the charge amount and the time interval after blasting, and outputs a safe, low-risk, medium-risk, or high-risk level. The charge quantity is a core parameter for measuring the scale of a blast, and its value directly determines the total amount of toxic and harmful gases generated at the moment of blasting. The post-blast time reflects the process of gas diffusion and dilution in the underground roadway. The combination of these two parameters follows the basic empirical rules in the field of mine safety: the larger the charge quantity and the shorter the post-blast time, the higher the concentration of toxic and harmful gases, and the greater the risk. As time goes by, under normal ventilation conditions, the gas concentration gradually decreases, and the risk decreases accordingly. The preset reference relationship solidifies the above empirical rules into a rule table that the system can directly query. Through the preset reference relationship, the system can obtain an initial risk level within milliseconds after a blast without performing complex online fluid dynamics calculations. This provides a fast, intuitive, and interpretable benchmark value for subsequent detailed risk assessment, significantly reducing the system's computational burden and response latency.

[0011] As a preferred embodiment of the IoT-based smart mine integrated management system described in this invention, the risk correction module corrects the default risk level according to the following rules: When any gas monitoring sensor at a key monitoring point is marked as faulty, it means that the system's ability to sense that area has a substantial gap. Even if the readings of other sensors are normal at this time, it cannot be ruled out that dangerous gas has accumulated at the location of the faulty sensor. Therefore, the system will directly raise the overall risk level to high risk and deal with the lack of information with the most conservative strategy. When there are no faulty sensors and two or more gas monitoring sensors are marked as sub-healthy, it indicates that the overall quality of the monitoring network has declined. Although the data deviation of a single sub-healthy sensor may still be within an acceptable range, the cumulative effect of multiple sub-healthy sensors may lead to an underestimation of risk. Therefore, the system will raise the default risk level by one level, for example, from low risk to medium risk.

[0012] When the effective gas concentration value corresponding to a certain sensor is marked with low confidence data, it also means that the data will be adopted in the next round of risk propagation calculation, but its reliability is questionable. The system will raise the risk level corresponding to the data by one level to exchange the safety margin for the preservation of data continuity. When the number of monitoring blind spots reaches a preset threshold, such as exceeding one-third of the total number of key monitoring points, it indicates that the sensor network has failed on a large scale and the overall perception capability of the system is seriously insufficient. At this time, the system no longer relies on the remaining local monitoring results, but directly raises the risk level to high risk. The aforementioned corrective rules can prevent the system from underestimating risks due to missing information when sensor network quality deteriorates or data is incomplete, thereby improving the overall safety level.

[0013] As a preferred embodiment of the IoT-based smart mine integrated management system described in this invention, in mine roadways, toxic and harmful gases generated by blasting are not uniformly distributed in space, but gradually diffuse along the ventilation airflow direction; therefore, if an abnormal gas concentration is detected at a monitoring point, it not only means that the point itself is at risk, but may also indicate that the roadways downstream of it will be affected in the future; existing technologies usually trigger alarms and controls based on the concentration exceeding the standard signal of a single sensor, and this point-based judgment method cannot identify the direction of risk propagation and the scope of impact; Therefore, the risk propagation zoning module summarized in this invention determines the current effective airflow direction by comparing the design airflow direction in the basic data of the roadway ventilation network with the real-time wind direction value in the wind speed and direction monitoring data; that is, it compares the design airflow direction in the basic data of the roadway ventilation network with the real-time wind direction value in the wind speed and direction monitoring data to determine the current effective airflow direction; when the real-time wind direction is basically consistent with the design direction, the design direction is used; when local backflow or abnormal wind direction occurs, correction is made based on the real-time data of most sensors.

[0014] The risk propagation zone is divided along the current effective airflow direction. The risk propagation status is judged based on the changing trend of effective gas concentration values ​​in adjacent zones, and the risk level of the downstream risk propagation zone is upgraded by one level. That is, the mine space is divided into multiple continuously arranged risk propagation zones along the current effective airflow direction, such as blasting operation area, upstream zone, midstream zone, downstream zone and personnel operation area, and an airflow propagation sequence is formed according to the airflow path. Based on the changing trend of effective gas concentration between adjacent zones, the risk propagation index is calculated. When the index continuously exceeds the preset threshold and the concentration shows an increasing trend, it is determined that the risk is propagating along the airflow direction, and the risk level of the downstream zone is upgraded by one level.

[0015] As a preferred embodiment of the IoT-based smart mine integrated management system described in this invention, in traditional control schemes, once a gas concentration exceeds the standard, the system typically activates maximum power ventilation directly. While this control method ensures safety, in many cases, moderate ventilation enhancement is sufficient to eliminate the risk, resulting in significant energy waste. In contrast, this invention uses a ventilation control strategy generation module to preset multi-level ventilation strategies, matching different levels of control response to different risk scenarios, as detailed below: When the overall risk level is safe, the system implements a routine economic ventilation strategy to maintain basic ventilation needs with minimal energy consumption; when the risk level is low, a slightly enhanced ventilation strategy is implemented, with enhanced exhaust only in localized areas upstream of the risk transmission, and the main ventilation fan maintaining a low operating frequency; when the risk level is medium, a global enhanced ventilation strategy is implemented, increasing the frequency of the main ventilation fan and adjusting the dampers in relevant areas accordingly; when the risk level is high, a maximum power emergency ventilation strategy is implemented, adjusting all fans and dampers to maximum capacity and triggering a personnel evacuation alarm. The specific control parameters corresponding to each strategy include the main ventilation fan frequency setting value and the damper opening setting value, which can be preset and stored in the system according to the rated parameters of the mine's main ventilation fan and the ventilation network calculation results; thus, under the premise of ensuring safety, it realizes the refined management of ventilation energy consumption and avoids over-response.

[0016] As a preferred embodiment of the IoT-based smart mine integrated management system described in this invention, the ventilation excitation feedback module will inevitably change the wind speed and gas concentration distribution in the underground roadway after the ventilation system's operating status changes, for example, when the frequency of the main ventilation fan increases from 35 Hz to 48 Hz. At this point, the monitoring points previously marked as monitoring blind zones are taken as blind zones to be verified. New data collected by the sensors corresponding to the blind zones to be verified is received. When the direction of concentration change of the sensor is consistent with the direction of concentration change of the adjacent healthy sensor and the relative deviation of the concentration change does not exceed the preset threshold, the sensor health status mark of the sensor is restored; otherwise, its sensor health status mark is maintained or reduced.

[0017] As a preferred embodiment of the IoT-based smart mine integrated management system described in this invention, the following points are addressed: Due to the massive volume of IoT-sensed data and high real-time requirements in underground mines, the communication bandwidth between underground and the surface is limited and subject to latency. Uploading all data processing tasks to the surface monitoring center would not only consume significant communication resources but also potentially affect the real-time performance of control due to transmission delays. Furthermore, the computing resources of the underground edge computing gateway are limited, making it unsuitable for executing complex global analysis tasks. This invention employs a layered deployment strategy to resolve these contradictions: the sensor health status judgment module and the trusted data filtering module are deployed within the underground edge computing gateway. The processing logic of these two modules is based on preset threshold comparisons and rule-based judgments, requiring minimal computation but demanding high real-time performance, making them suitable for completion at the edge. Deploying them underground allows for immediate preliminary processing and trustworthiness assessment of the collected data, reducing the amount of raw data uploaded. The default risk level judgment module, risk correction module, risk propagation zoning module, and ventilation control strategy generation module are deployed and executed on the SCADA server in the ground monitoring center. These modules involve global risk analysis, spatial propagation calculation, and multi-level strategy matching, which have high computational complexity. They also need to comprehensively consider the ventilation network topology of the entire mine and the data of multiple zones, making it suitable to utilize the stronger computing power of the ground server to complete them.

[0018] The execution and update module and the ventilation excitation readback module communicate with the underground main ventilation fan frequency converter and damper controller through the mine industrial ring network, and are responsible for issuing control commands and transmitting status data. Through the above-mentioned layered deployment, the system achieves an organic combination of rapid data preprocessing at the edge and comprehensive global analysis at the ground, which enables the system to improve its overall computing power while ensuring real-time control, reducing the transmission pressure on the communication network, and improving the system's stability and scalability in complex mining environments.

[0019] The beneficial effects of this invention are: 1. This invention introduces a sensor health status judgment module to perform multi-dimensional online evaluation of the output behavior of each sensor and classify them into healthy, sub-healthy, and failed. This allows the system to know the credibility of the data before making control decisions. When the quality of the sensor network deteriorates, the system actively adjusts the risk level through the risk correction module and selects redundant data sources through the trusted data filtering module, thus avoiding misjudgments and malfunctions caused by a single sensor failure.

[0020] 2. This invention, through the cooperation of a default risk level judgment module and a risk correction module, combines information from three dimensions—explosive charge quantity, time window, and sensor network health status—to output a comprehensive risk level. Then, a risk propagation zoning module judges the risk diffusion trend along the airflow direction. Finally, a ventilation control strategy generation module matches a ventilation control scheme adapted to the current risk situation from multiple preset strategies. This hierarchical, progressive control method can adopt different levels of ventilation measures based on the actual risk level between the two extreme states of safety and high risk, changing the single mode of full-power ventilation in traditional schemes where exceeding limits results in reduced ventilation power consumption while ensuring safety.

[0021] 3. This invention uses a ventilation excitation readback module to observe the response direction, response amplitude, and response time of each sensor by using known environmental excitations such as fan frequency step jumps, and compares them with adjacent healthy sensors, thereby realizing the dynamic recovery of monitoring blind spots or confirmation of permanent failure. This closed-loop feedback mechanism enables the system to continuously self-correct and self-evolve its understanding of the sensor health status. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a system architecture diagram of the present invention.

[0023] Figure 2 This is an overall workflow diagram of Embodiment 1 of the present invention.

[0024] Figure 3 This is a flowchart of the overall workflow of step S1 in Embodiment 1 of the present invention.

[0025] Figure 4 This is a flowchart of the overall workflow of step S2 in Embodiment 1 of the present invention.

[0026] Figure 5This is a flowchart of the overall workflow of step S3 in Embodiment 1 of the present invention.

[0027] Figure 6 This is a flowchart of the overall workflow of steps S4 to S6 in Embodiment 1 of the present invention. Detailed Implementation

[0028] 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.

[0029] 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.

[0030] 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 an embodiment" appearing in different places in this specification does not all refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0031] Example 1 Reference Figure 1 The first embodiment of the present invention provides a smart mine integrated management system based on the Internet of Things, the system including a processor, a memory and multiple functional modules that are communicatively connected to the processor; The above functional modules include: The data access module is used to receive blasting parameter data, gas monitoring data, wind speed and direction monitoring data, ventilation equipment operation data, and basic data of the roadway ventilation network through an industrial Ethernet interface or wireless IoT gateway, and write the above multi-source data into the memory. The sensor health status judgment module is used to output sensor health status flags based on the direction of change, fluctuation amplitude, response time, and heartbeat status of gas monitoring data and wind speed and direction monitoring data. The reliable data filtering module is used to select, downgrade, or process missing labels on gas monitoring data based on sensor health status labels, and outputs valid gas concentration values ​​and low-reliability data labels. The default risk level judgment module is used to query the post-blast risk level comparison relationship stored in the memory based on the charge amount in the blasting parameter data and the time interval between the current time and the blasting moment, and output the default risk level. The risk correction module is used to correct the default risk level and output a comprehensive risk level based on sensor health status markers, low-reliability data markers, and monitoring blind zone markers. The risk propagation zoning module is used to divide the blasting operation area, key nodes of the return air roadway, and personnel operation area into multiple risk propagation zones based on the basic data of the roadway ventilation network and wind speed and direction monitoring data, and output the zone control priority and the comprehensive risk level after zone correction. The ventilation control strategy generation module is used to generate the main ventilation fan frequency setting value, damper opening setting value and alarm information based on the comprehensive risk level after zoning correction, sensor health status markers and zoning control priority. The execution and update module is used to send the main ventilation fan frequency setting value and damper opening setting value to the main ventilation fan inverter and damper controller, and update the sensor health status flag, monitoring blind zone flag and comprehensive risk level after ventilation control; The ventilation excitation readback module is used to verify the response consistency of monitoring points marked as blind spots to be verified, and to restore, maintain or reduce the state of the corresponding points based on the verification results. The closed-loop control module is used to control multiple functional modules to execute in a loop in the order of data access, health judgment, reliable screening, risk correction, zone linkage, ventilation execution, readback verification, and status update. Specifically, the data access module is an IoT acquisition interface circuit deployed in an edge computing gateway underground; the sensor health status judgment module and the trusted data filtering module are executed by the processor of the edge computing gateway; the default risk level judgment module, the risk correction module, the risk propagation zoning module, and the ventilation control strategy generation module are executed by the SCADA server of the ground monitoring center; the execution and update module and the ventilation excitation readback module are communicatively connected to the main ventilation fan frequency converter and the damper controller through the mine industrial ring network. Preferably, the edge computing gateway and the SCADA server are connected via a mining industrial ring network communication to achieve data collaboration between underground data acquisition, surface judgment, and equipment execution.

[0032] In one optional embodiment, the system further includes a human-machine interface and a remote alarm interface. The human-machine interface is used for manually inputting blasting parameters or verifying the output results of the steps. The remote alarm interface is used for sending alarm information and evacuation prompts to the mine integrated management and control platform and the underground personnel positioning terminal.

[0033] This embodiment, through the above-mentioned modular setup, can integrate blasting parameter acquisition, sensor health assessment, reliable data filtering, risk propagation correction, zone linkage control, and readback verification into a single system, making it easy to deploy in a mine integrated management and control platform, SCADA system, or edge control terminal.

[0034] Example 2 Reference Figures 2-6 This is the second embodiment of the present invention, which differs from the first embodiment in that: the present invention also provides an implementation method for an IoT-based smart mine integrated management system as described in Embodiment 1 above, the specific steps of which are as follows: S1. The data access module collects multi-source IoT sensing data from underground mines to construct the raw data input layer; like Figure 3 As shown, the data access module receives and parses multi-source IoT sensing data from the blasting operation area and its associated return airway network through an industrial Ethernet interface or a wireless IoT gateway at a preset sampling period. The multi-source IoT sensing data includes at least: Blasting parameter data, including but not limited to: the charge quantity, blasting time, blasting operation orientation code, and the roadway number corresponding to the blasting operation; the blasting parameter data can be automatically pushed by the blasting design system or manually entered by the dispatcher through the human-machine interface; Gas monitoring data, which is collected by multiple gas monitoring sensors deployed in the blasting operation area, key nodes of the return airway, and personnel operation area; the gas monitoring data includes, but is not limited to: methane concentration monitoring values, carbon monoxide concentration monitoring values, and oxygen concentration monitoring values. Wind speed and direction monitoring data, which is collected by multiple wind speed and direction monitoring sensors deployed in each main ventilation tunnel, and includes, but is not limited to: wind speed value and wind direction value; The ventilation equipment operation data is fed back by the main ventilation fan frequency converter and the damper controller; the ventilation equipment operation data includes, but is not limited to: the current operating frequency of the main ventilation fan, the current opening degree of the damper, and the start / stop status of the main ventilation fan. The basic data of the roadway ventilation network includes, but is not limited to: the three-dimensional spatial connection relationship of each roadway, the design airflow direction and the wind resistance coefficient; the basic data of the roadway ventilation network is static basic data, which is pre-stored in the memory and used to determine the airflow propagation sequence later.

[0035] Preferably, the gas monitoring data and wind speed and direction monitoring data are accompanied by timestamps, unique sensor identifiers, sampling sequence numbers, and heartbeat status information, so as to determine the sensor health status in subsequent steps; Preferably, the data access module associates and maps multi-source IoT sensing data in the following order: blasting operation area, key nodes of return air roadway, personnel operation area, ventilation equipment status, and basic data of roadway ventilation network, forming a data index table for subsequent judgment. This data index table records at least the correspondence between each monitoring point and the corresponding roadway number, airflow direction, and equipment number. For example, the data index table may include the following fields: monitoring point number, monitoring point type, roadway number, zone number, airflow direction, associated equipment number, data status flag, and update timestamp. Through the data index table, the processor can associate blasting parameter data, gas monitoring data, wind speed and direction monitoring data, ventilation equipment operation data, and roadway ventilation network basic data under the same index framework, thereby avoiding isolated storage and isolated judgment between different types of data.

[0036] Step S2: Filter reliable data based on sensor health status markers and determine the risk situation of gas diffusion after the explosion; like Figure 4 As shown, after obtaining the multi-source heterogeneous data and data index table output from step S1, the processor in the system executes the following sub-steps to complete the progressive judgment process from raw data to risk status: S21. The sensor health status is judged and processed by the sensor health status judgment module; Specifically, the sensor health status judgment module independently performs health status judgment processing on each gas monitoring sensor and each wind speed and direction monitoring sensor, and outputs the judgment results in the form of sensor health status labels; the sensor health status labels include at least: healthy, sub-healthy, and failed; Furthermore, the processor marks the sensor's health status as healthy when both of the following conditions are met simultaneously; Condition 1: Within a first preset number of sampling periods N1, the direction of change of the sensor output value is the same as the direction of change of the output value of adjacent sensors in the same monitoring area that are determined to be healthy; where the direction of change refers to the increasing or decreasing trend of the output value between two adjacent sampling periods. Condition 2: Within N1 consecutive sampling periods, the fluctuation range of the sensor output value does not exceed a preset fluctuation threshold; the fluctuation range refers to the difference between the maximum and minimum values ​​of the output value within that time period; wherein, the first preset number N1 should be greater than the number of sampling periods corresponding to the typical response time of the sensor output value; for example, if the sampling period is 10 seconds and the typical response stabilization time of the sensor after environmental changes is 30 to 50 seconds, then N1 can be set to 5; The preset fluctuation threshold can be set according to the sensor's range and accuracy level; preferably, the preset fluctuation threshold is set to 2% to 5% of the sensor's full-scale value; for example, for a methane sensor with a full-scale value of 4% volume concentration, the preset fluctuation threshold can be set to 0.08% to 0.2% volume concentration. Furthermore, when the sensor is able to continuously output monitoring data, but at least one of the following conditions occurs, the processor marks the sensor's health status as sub-healthy: Response hysteresis: The response time of the current sensor to the same environmental change exceeds a preset response duration threshold; the response time refers to the time elapsed from the moment when the environmental parameter changes detectably (starting from the moment when the output value of the adjacent health sensor changes beyond the preset fluctuation threshold) to the moment when the change in the output value of the current sensor reaches a preset proportion (e.g., 70%) of the change in the adjacent health sensor; the preset response duration threshold can be set according to the sensor's nominal response time (T90) plus an engineering margin; for example, for a gas sensor with a nominal T90 of 20 seconds, considering the harsh working conditions downhole, the preset response duration threshold can be set to 30 to 60 seconds.

[0037] Excessive Deviation: The deviation between the sensor output value and the health sensor output value in the same monitoring area exceeds a preset deviation threshold, but the output value has not exceeded the physically reasonable range defined by the sensor range (i.e., it is still between the lower limit and the upper limit of the range); the deviation threshold can be set according to the maximum allowable measurement consistency error of the same model sensor under the same environment; for example, the deviation threshold can be set to 10% to 15% of the health sensor output value, or set to 5% of the full range value; for example, for a methane sensor with a range of 4% volume concentration, when the health sensor reading is 1%, the deviation threshold is 0.10% to 0.15%.

[0038] Abnormal fluctuation: During a period when the environment remains stable (based on the judgment that the change in the output value of adjacent health sensors does not exceed the preset fluctuation threshold), the absolute difference of the output value of the sensor between consecutive adjacent sampling periods is greater than the fluctuation threshold; wherein, the stable environment means that the change in the output value of the health sensor in the same monitoring area does not exceed the fluctuation threshold.

[0039] Furthermore, the processor marks the sensor's health status as failed when the sensor exhibits at least one of the following conditions: Data interruption: The sensor has no valid data output within a second preset number N2 sampling periods; the second preset number N2 is used to distinguish between the sensor's instantaneous communication jitter and a true loss of connection; in order to balance the timeliness of fault identification and the fault tolerance rate of misjudgment, N2 can be set as the maximum tolerable instantaneous interruption duration divided by the sampling period; for example, if the maximum tolerable instantaneous interruption duration is 30 seconds and the sampling period is 10 seconds, then N2=3.

[0040] Heartbeat Abnormality: The heartbeat packet information of the sensor is continuously lost for a third preset number of sampling cycles, N3. Since the transmission frequency of heartbeat packets is usually higher than the data sampling frequency, the loss of heartbeat packets can reflect communication link abnormalities more quickly than data interruption. Therefore, N3 can be less than N2. For example, if the heartbeat packet is sent once every 5 seconds and the sampling period is 10 seconds, and it is expected that the communication abnormality can be identified within 20 seconds, then N3=2.

[0041] Output freeze: The sensor output value remains fixed for a fourth preset number of sampling periods N4, and the fixed output value changes in the opposite direction to the change direction of other health sensors in the same period, or the proportion of the change direction consistency is lower than the direction consistency threshold (e.g., lower than 50%). The fourth preset number N4 should be greater than the maximum number of continuous periods when the sensor outputs a naturally constant value when the environment is stable, so as to avoid misjudging normal stability as freeze. For example, N4 can be set to 5.

[0042] Exceeding limits: The sensor output value is less than the lower limit of its range or greater than the upper limit of its range; at this time, the processor determines that the sensor's own circuit or sensing element has suffered a serious malfunction.

[0043] Specifically, the relationship between the above sampling periods and time is as follows: if the sampling period is set to 10 seconds, then N1=5 corresponds to 50 seconds, N2=3 corresponds to 30 seconds, N3=2 corresponds to 20 seconds, and N4=5 corresponds to 50 seconds. The above threshold parameters are all preset in memory in combination with the dynamic response characteristics of the sensor, the typical timeout characteristics of the communication link, and the safety redundancy requirements of underground mines.

[0044] In a preferred embodiment, the processor further enhances the health status judgment by combining specific events; for example, when the wind speed and direction monitoring sensor at a key monitoring point undergoes a step change in the operating frequency of the main ventilation fan, and the wind speed change is less than the response magnitude threshold within a preset response time threshold, the processor directly marks it as failed, regardless of its previous state. The step change refers to the change in the operating frequency of the main ventilator exceeding a preset step frequency change threshold within one sampling period. This step frequency change threshold ensures that the frequency change generates a detectable wind speed change, and its value is determined based on the following: when the frequency change of the main ventilator exceeds this threshold, the change in wind speed at the fan outlet is greater than the resolution of the wind speed and direction monitoring sensor. For example, for a main ventilator with a rated frequency of 50 Hz, the step frequency change threshold can be set to 10% of the rated frequency, i.e., 5 Hz. The response magnitude threshold can be set to 20% of the steady-state wind speed difference before and after the step change. All of the above thresholds can be determined through on-site calibration based on the rated parameters of the main ventilator, duct resistance characteristics, and sensor resolution.

[0045] It should be noted that this step introduces an assessment of the sensor data quality within the post-blast ventilation control process. Unlike existing technologies that assume all sensor data is accurate and reliable, this step proactively identifies abnormal sensor states by analyzing the direction of change, fluctuation amplitude, response time, and correlation with adjacent sensors. This judgment process is entirely based on pre-defined logical rules, providing crucial prior information about data reliability for subsequent decision-making.

[0046] Step S22: Filter the trusted data using the trusted data filtering module; Specifically, after obtaining the sensor health status markers of each sensor output from step S21, the gas monitoring data is filtered for reliable data based on these markers to determine the effective gas concentration values ​​at each monitoring point. Each monitoring point includes critical and non-critical monitoring points. Critical monitoring points include at least the blasting face, the intersection of the main return airway, and the first airflow arrival point corresponding to the personnel work area. The monitoring data from these critical monitoring points directly participate in the correction of the comprehensive risk level, the determination of the priority of zoning control, the identification of monitoring blind spots, and the verification of ventilation excitation readback. Non-critical monitoring points refer to other monitoring points besides the critical monitoring points. Their monitoring data is used to assist in the trend comparison of sensor health status, the consistency verification of airflow direction, and to provide redundant references, but does not directly participate in the correction of the comprehensive risk level.

[0047] The rules for selecting reliable data for key monitoring points are as follows: a. When the sensor health status of the main sensor at a key monitoring point is marked as healthy, the processor directly uses the monitoring value collected by the main sensor as the effective gas concentration value at that monitoring point. b. When the sensor health status of the main sensor at a key monitoring point is marked as sub-healthy, the processor checks whether the monitoring point is equipped with a redundant verification sensor that is marked as healthy. If a redundant verification sensor exists, the processor directly uses the monitoring value collected by the redundant verification sensor as the effective gas concentration value for that point. If it does not exist, the processor still uses the monitoring value collected by the main sensor as the effective gas concentration value, but at the same time adds a low-confidence data mark to the effective gas concentration value. The low-confidence data mark indicates that the reliability of the effective gas concentration value is lower than the normal level.

[0048] c. When the sensor health status of the main sensor at a key monitoring point is marked as failed, the processor marks the effective gas concentration value of that monitoring point as missing data and marks the detection area where the monitoring point is located as a monitoring blind zone.

[0049] In this embodiment, the monitoring blind zone indicates that the data of the current monitoring point is in an unavailable state. For monitoring points marked as monitoring blind zones, they are not included in the calculation of effective gas concentration values ​​when calculating the partition concentration in step S3, and the unavailable state can be further converted into a blind zone to be verified for recovery verification in the subsequent step S5.

[0050] It should be noted that this step establishes a mechanism for automatically optimizing and degrading data sources, enabling the system to have stronger fault tolerance and data availability when the quality of the sensor network changes dynamically.

[0051] Step S23: The default risk level of gas diffusion after blasting is judged and processed by the default risk level judgment module. Specifically, the default risk level judgment module takes the blasting time in the blasting parameter data as the zero time starting point, and queries the post-blasting risk level comparison relationship stored in the system memory according to the charge amount and the time interval between the current time and the blasting time to obtain the corresponding default risk level. The default risk level is one of safe, low risk, medium risk or high risk. The relationship between the blasting risk levels is as follows: The default risk level is low when the charge amount is less than or equal to the first charge amount threshold and the current time is within the first preset time after blasting; the default risk level is safe when the charge amount is less than or equal to the first charge amount threshold and the current time is between the first and second preset time after blasting; the default risk level is medium when the first charge amount threshold is less than or equal to the second charge amount threshold and the current time is within the second preset time after blasting; the default risk level is low when the first charge amount threshold is less than or equal to the second charge amount threshold and the current time is between the second and third preset time; the default risk level is high when the charge amount is greater than the second charge amount threshold and the current time is within the third preset time after blasting; and the default risk level is medium when the charge amount is greater than the second charge amount threshold and the current time is between the third and fourth preset time after blasting.

[0052] The first and second charge thresholds are constants preset based on the mine ventilation network calculation results and blasting operation procedures. The basis for their setting is that when the charge does not exceed the first charge threshold, the amount of toxic and harmful gases generated by the blasting can be diluted to below a safe concentration within a second preset time under normal ventilation conditions. When the charge exceeds the second charge threshold, high-intensity ventilation must be activated to ensure safety. For example, the first charge threshold can be set to 50 kg and the second charge threshold can be set to 200 kg.

[0053] The first, second, third, and fourth preset durations are used to characterize the typical gas diffusion stages within different time windows after blasting. For example, the first preset duration can be set to 15 minutes, the second preset duration to 30 minutes, the third preset duration to 45 minutes, and the fourth preset duration to 90 minutes. When the mine has strong ventilation capacity, the above time thresholds can be appropriately shortened. When the mine has high wind resistance or a large blasting scale, the above time thresholds can be appropriately extended.

[0054] It should be noted that this step provides a rapid and intuitive preliminary assessment of the risk of gas diffusion after blasting. It solidifies industry-standard experience regarding charge quantity, timing, and risk into automatically executable rules, reducing the consumption of edge computing resources, and the results are highly interpretable and deterministic.

[0055] Step S24: Correct the comprehensive risk level based on the sensor health status through the risk correction module, and output the final comprehensive risk situation result. Specifically as follows: A1. After obtaining the default risk level output in step S23, combine the sensor health status markers of each key monitoring point output in step S21 with the monitoring blind zone markers and low-confidence data markers output in sub-step S22, and correct the default risk level according to the following correction rules to obtain the final comprehensive risk level: a) When any gas monitoring sensor at a key monitoring point is marked as faulty, the risk level of that key monitoring point is directly and forcibly set to high risk; b) When no sensor at a key monitoring point has been marked as faulty, but two or more gas monitoring sensors at key monitoring points are marked as sub-optimal, the processor raises the default risk level of that key monitoring point by one level, i.e., from safe to low risk, from low risk to medium risk, and from medium risk to high risk; c) When an effective gas concentration value used in risk assessment is marked as low-confidence data, the processor raises the corresponding risk level by one level; d) When the number of blind spots at key monitoring points is greater than or equal to the threshold for the number of blind spots, the processor will forcibly set the overall risk level to high risk. For example, the threshold for the number of blind spots can be preset according to the total number of key monitoring points and the mine safety redundancy requirements, preferably one-third of the total number of key monitoring points, so that when a concentrated failure occurs at key monitoring points, the system no longer relies on local monitoring results, but directly raises the risk level to high risk to enhance the system's safety redundancy in the event of a decline in sensor network quality. e. When none of the conditions of the above correction rules one to four are triggered, the processor will directly use the default risk level as the overall risk level.

[0056] A2. After completing the above-mentioned default risk level correction, the risk correction module outputs the final comprehensive risk situation result, which includes at least: comprehensive risk level, the reason for triggering the change of comprehensive risk level, the effective gas concentration value of each key monitoring point, and the monitoring blind zone marker.

[0057] It should be noted that this step uses the health status of the sensor network as an independent variable to correct the risk assessment results, and establishes a clear and interpretable rule system; it outputs the multi-dimensional intermediate judgment results in a structured manner, providing the operators in the monitoring center with a complete decision-making traceability chain.

[0058] Step S3: The risk propagation zoning module performs coordinated control processing based on the zoning and hierarchical classification of the airflow propagation sequence in the alleyway; like Figure 5 As shown, specifically, after obtaining the basic data and data index table of the roadway ventilation network output in step S1 and the comprehensive risk situation result output in step S2, the risk propagation zoning module performs the following sub-steps; S31. Determine the current effective airflow direction; The processor first reads the design airflow direction from the basic data of the roadway ventilation network, and combines it with the real-time wind direction values ​​from the wind speed and direction monitoring data of all monitoring points to determine the current effective airflow direction: when the consistency ratio between the real-time wind direction values ​​of most wind speed and direction monitoring sensors on the same airflow path and the design airflow direction is greater than a preset direction consistency threshold (preferably 70%), the processor determines the design airflow direction as the current effective airflow direction; when the consistency ratio is less than the direction consistency threshold, the processor re-determines the current effective airflow direction based on the real-time wind direction values ​​of most sensors.

[0059] S32. Divide the risk transmission zones and construct the airflow transmission sequence; After determining the current effective airflow direction, the processor, based on the roadway number and airflow direction information recorded in the data index table, divides the blasting operation area, key nodes of the return airway, and personnel operation area into multiple continuously arranged risk propagation zones along the current effective airflow direction. Each risk propagation zone is a continuous roadway spatial unit along the airflow direction, used to describe the spatial propagation path of gas risk. Each risk propagation zone includes at least: a blasting operation area zone, an upstream risk propagation zone, a midstream risk propagation zone, a downstream risk propagation zone, and a personnel operation area zone. The processor arranges each risk propagation zone according to the airflow path sequence based on the current effective airflow direction, forming an airflow propagation sequence. For example, if the zone numbers are Z1, Z2, Z3, Z4 and Z5, the airflow propagation sequence is represented as: Z1→Z2→Z3→Z4→Z5, where Z1 is the blasting operation area, Z2 is the upstream zone, Z3 is the midstream zone, Z4 is the downstream zone, and Z5 is the personnel operation area. When the real-time wind direction value of a certain zone is inconsistent with the current effective wind flow direction, the zone is marked as a wind direction abnormal zone, and the risk level of the zone is upgraded by one level in subsequent risk calculations.

[0060] S33. Calculate the risk transmission index and determine the transmission trend; Based on the effective gas concentration values ​​of each key monitoring point output by the reliable data filtering module, and according to the zoning relationship, the effective gas concentration values ​​of multiple key monitoring points within the same risk transmission zone are aggregated and calculated to obtain the effective gas concentration value of that zone. Preferably, the aggregation calculation method is an arithmetic mean; Calculate the risk transmission index based on the effective gas concentration values ​​of adjacent risk transmission zones. The calculation formula is: ;in, Let be the risk propagation index of the j-th risk propagation zone, which is a dimensionless quantity; Let j be the effective gas concentration value of the j-th risk propagation zone. This represents the effective gas concentration value for the previous risk transmission zone. The preset minimum positive number to prevent zeroing; for example, the preferred value is... ; The system determines that the risk propagates along the airflow direction when the following conditions are met: If the sampling period is greater than the preset propagation threshold for N5 consecutive sampling periods, and > For example, N5 is preferably 2 to 3; the preset propagation threshold can be determined according to the statistical distribution of the concentration change ratio of adjacent zones in historical blasting events, and the preset propagation threshold is preferably 0.25 to 0.30, for example, a value of 0.25; When it is determined that a risk is spreading along the wind direction, the processor will raise the risk level of the downstream risk propagation zone by one level.

[0061] S34. Determine the partition control priority and output the partition correction results; Based on the risk propagation assessment results, the priority of zone control is determined according to the following rules; the zone control priority includes at least three levels: priority local control, local control and global standby, and priority global control: Prioritized local control: When an upstream risk propagation zone is determined to be of medium or high risk, and its directly downstream adjacent risk propagation zone is still in a low-risk or safe state, the processor will prioritize local control as the zone control priority. At this time, the processor will prioritize directional enhanced control of the dampers corresponding to the upstream risk propagation zone, and will not increase the operating frequency of the main ventilation fan. The local control refers to the local airflow adjustment achieved by adjusting the damper opening corresponding to the damper controller.

[0062] Local control + global standby: When an upstream risk propagation zone is determined to be of medium or high risk, and its directly downstream adjacent risk propagation zone has been determined to be of low risk, but the risk propagation index of the downstream zone shows an upward trend within N5 consecutive sampling periods and has not yet exceeded the preset propagation threshold, the processor sets the zone control priority to local control and global standby. At this time, while the processor performs local enhanced control on the upstream zone, it presets the target frequency parameter of the main ventilator to a reserve frequency, but does not immediately issue it for execution. The reserve frequency is a frequency value that is slightly higher than the current operating frequency and not higher than the second enhanced frequency, which is used to quickly switch to global ventilation state when the risk spreads further, thereby taking into account both response speed and energy consumption control.

[0063] Priority global control: When multiple consecutive risk propagation zones simultaneously have risk propagation indices greater than the preset propagation threshold, and the effective gas concentration values ​​of each zone show an increasing trend along the airflow direction, the processor will upgrade the zone control priority to priority global control; at this time, the processor will start the global enhanced ventilation process and increase the operating frequency of the main ventilation fan to the second enhanced frequency or a higher level to achieve overall suppression of the risk propagation path.

[0064] The final output of step S3 includes at least: the partition-corrected comprehensive risk level, the partition control priority, the risk propagation direction marker, and the downstream risk propagation partition marker; wherein, the partition-corrected comprehensive risk level refers to the partition-level risk level obtained by propagating the corresponding partition based on the comprehensive risk level output in step S2 and the risk propagation judgment result; the partition-corrected comprehensive risk level serves as one of the inputs to step S4, and the partition control priority serves as the basis for step S4 to select local control, standby control, or global control.

[0065] This step upgrades ventilation control from single-zone concentration triggering to zoned propagation sequence triggering, enabling the processor to not only determine the existence of a risk but also the propagation path of the risk. By setting three levels of zoned control priority, a progressive control system is constructed, from local treatment to local treatment and global preparation, and then to global response. This system can more accurately match different stages of risk spread, reduce unnecessary energy consumption, and avoid focusing only on the local area while ignoring the safety hazards of the propagation link.

[0066] S4. Based on the overall risk level after zoning correction, sensor health status, and zoning control priority, execute the corresponding level of ventilation control strategy; like Figure 6 As shown, after obtaining the comprehensive risk level and zone control priority after the zoning correction output in step S3, and the sensor health status markers of each key monitoring point output in step S2, the ventilation control strategy generation module automatically selects and executes the corresponding ventilation control strategy according to the preset ventilation control strategy comparison relationship, and generates the corresponding control command. The specific ventilation control strategies are as follows: When the overall risk level after zoning correction is safe, the processor executes the daily economic ventilation strategy; there are no restrictions on the priority of zoning control; there are no restrictions on the values ​​of the sensor health status markers at key monitoring points; the daily economic ventilation strategy is: adjusting the frequency of the main ventilation fan to the preset energy-saving operating frequency, and restoring each damper to the default production opening; when the following conditions are met simultaneously, the processor executes a slightly enhanced ventilation strategy: the overall risk level after zoning correction is low risk and the sensor health status markers at all key monitoring points are healthy, and the zoning control priority is priority local control; the slightly enhanced ventilation strategy is: prioritizing local enhanced exhaust ventilation for the dampers corresponding to the upstream risk propagation zone, while the main ventilation fan maintains its current operating frequency or maintains a low-amplitude adjustment.

[0067] The processor executes a moderately enhanced ventilation strategy when any of the following conditions are met simultaneously: the overall risk level after zoning correction is low risk, and at least one sensor at the key monitoring point has a sensor health status marked as sub-healthy or has an effective gas concentration value with an attached low-confidence data label, and the zoning control priority is local control with global standby; or the overall risk level after zoning correction is medium risk, and the zoning control priority is local control with global standby.

[0068] The above-mentioned moderately enhanced ventilation strategy is as follows: enhanced exhaust is performed on the dampers corresponding to the upstream risk transmission zone; at the same time, the target frequency parameter of the main ventilation fan is preset as a reserve frequency, which is a frequency value that is slightly higher than the current operating frequency and not higher than the second enhanced frequency, but the entire system is not immediately boosted; when the zone control priority is further increased in the next cycle, the reserve frequency is converted into the actual execution frequency.

[0069] The processor executes a global enhanced ventilation strategy when the following conditions are met simultaneously: if the overall risk level after zoning correction is medium risk, then the zoning control priority is the global control priority; the global enhanced ventilation strategy is: to increase the frequency of the main ventilation fan to the second enhanced frequency, and to switch the air doors in the relevant areas along the risk propagation path to the enhanced exhaust position, while sending a prompt message to the underground personnel positioning terminal to pay attention to the ventilation changes.

[0070] The processor executes the maximum power emergency ventilation strategy when the following conditions are met simultaneously: if the overall risk level after the partition correction is high risk, then the partition control priority is set to priority global control; the maximum power emergency ventilation strategy is: to increase the frequency of the main ventilation fan to the maximum allowable frequency, to switch all relevant air doors to the full exhaust position, to trigger a high-risk alarm to the mine integrated management and control platform, and to send an emergency evacuation command to the underground personnel positioning terminal to immediately evacuate to the fresh air flow area; Furthermore, when the combination of the overall risk level after partition correction and the partition control priority does not meet any of the above strategy triggering conditions, the processor will prioritize the ventilation control strategy corresponding to the higher overall risk level, following the principle of choosing the higher one.

[0071] The specific fan frequency and damper opening values ​​corresponding to the above strategies can be preset according to the rated parameters of the mine's main ventilation fan and the ventilation network calculation results. For example, the energy-saving operating frequency can be set to 35 Hz, the first enhanced frequency can be set to 42 Hz, the second enhanced frequency can be set to 48 Hz, and the maximum allowable frequency can be set to 50 Hz. When the main ventilation fan is at the energy-saving operating frequency, the air volume meets the daily production needs. When it is adjusted to the first enhanced frequency or the second enhanced frequency, different levels of airflow enhancement effects can be formed. When the maximum allowable frequency is reached, it is used for emergency ventilation.

[0072] In a preferred embodiment, the processor further prioritizes local enhanced control of the area upstream of the risk propagation zone based on the risk propagation zone determination result, and then decides whether to increase the frequency of the main ventilation fan based on the risk changes of the downstream risk propagation zone, thereby forming a progressive linkage control with local first and global last-line.

[0073] It should be noted that this step constructs a multi-level progressive ventilation control strategy that is linked to three dimensions: the comprehensive risk level after the partition correction, the sensor health status markers, and the partition control priority. Unlike the single mode of full-power ventilation when exceeding the limit in the existing technology, this invention can automatically select a progressive strategy from daily economic ventilation to maximum power emergency ventilation between the two extreme states of safety and high risk, based on the different levels of the comprehensive risk level, the health of the sensor network, and the stage of risk propagation along the roadway, to achieve a fine balance between safety redundancy and energy saving.

[0074] S5. Execute ventilation control commands and use the monitored data after control to update the health status of the sensors in reverse. like Figure 6 As shown, this step includes the following sub-steps: S51. Issue and execute control commands; The execution and update module uses an industrial Ethernet to send the control commands generated in step S4, including the main ventilation fan frequency setting value and the opening setting value of each damper, to the corresponding main ventilation fan frequency converter and damper controller, thereby completing the switching of the ventilation system's operating status. At the same time, if there are alarm or prompt messages, they are pushed to the mine integrated management and control platform and the underground personnel positioning terminal.

[0075] S52, read back the ventilation excitation and self-check the monitoring blind spot; After the processor completes the above-mentioned switching of the ventilation system's operating state, it does not immediately permanently remove the key monitoring points marked as failed in step S2, but temporarily marks them as blind areas to be verified, indicating that they have not yet been confirmed as permanently failed and still need further stimulation verification. After the operating status of the ventilation system changes, the ventilation excitation readback module continuously receives new data collected by various gas monitoring sensors and wind speed and direction monitoring sensors, and performs self-verification on the blind area to be verified using the following rules. The specific steps are as follows: When a step change occurs, if the wind speed and direction monitoring sensor located in the blind zone to be verified does not reach the response magnitude threshold within the preset response time threshold, then the sensor at that key monitoring point will continue to be marked as failed. If a gas monitoring sensor located in the blind zone to be verified shows a response trend consistent with that of an adjacent healthy sensor after enhanced ventilation, then the sensor is restored from the blind zone to a sub-healthy or healthy state. When a monitoring point in a certain blind zone fails to respond effectively after multiple verification cycles, it is identified as a permanent failure blind zone and switched to an adjacent redundant point or a neighboring reference point for compensation. For monitoring points marked as permanent failure blind zones, they can only re-participate in the health status assessment process after manual maintenance or equipment replacement.

[0076] Furthermore, consistent response trend means that within the current verification period, the direction of concentration change of the sensor to be verified is the same as the direction of concentration change of the adjacent health sensor, and the relative deviation between the concentration change of the sensor to be verified and the concentration change of the adjacent health sensor does not exceed 30%; the relative deviation The calculation formula is: ; in, This represents the concentration change of the sensor to be verified during the verification period. This represents the concentration change of adjacent health sensors. The preset minimum positive number to prevent zeroing is preferably set to a value of [value to be filled in]. When the stated At that time, the processor determines that its response trend is consistent with that of the adjacent health sensors.

[0077] S53, Closed-loop update; The execution and update module uses the new monitoring data after ventilation control to re-execute steps S21 to S24 to update the sensor health status marker, monitoring blind zone marker, and comprehensive risk level, and uses the updated results as the input for the next cycle, thereby forming a complete closed-loop control chain.

[0078] It should be noted that by observing the sensor's response to known and controllable environmental stimuli, this step enables the system to continuously and dynamically correct its perception of the health status of each sensor, and distinguish between permanent failure blind zones and recoverable blind zones to be verified, thus achieving the purpose of closed-loop self-verification of the system.

[0079] S6, execute in a loop; like Figure 6 As shown, the updated sensor health status marker, updated monitoring blind zone marker, and updated comprehensive risk level output from step S5 are used as inputs for the next step S2. The above steps are executed cyclically to form a continuously operating closed-loop system for comprehensive mine management.

[0080] It should be noted that this step is executed cyclically, enabling the system to continuously adjust the control strategy based on changes in the post-blasting environment, ventilation status, and sensor status. This ensures that ventilation management, risk assessment, and equipment status management in the mine remain dynamically consistent. Compared with the existing technology that does not verify after a single control, this achieves more stable online operation and stronger on-site adaptability.

[0081] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A smart mine integrated management system based on the Internet of Things, comprising a processor, a memory, and functional modules communicating with the processor, characterized in that, The functional modules include: The data access module is used to receive blasting parameter data, gas monitoring data, wind speed and direction monitoring data, ventilation equipment operation data and basic data of roadway ventilation network, and to establish a correlation mapping between the data and the roadway number, airflow direction and equipment number according to the monitoring point; The sensor health status judgment module is used to output sensor health status markers based on gas monitoring data and wind speed and direction monitoring data. The reliable data filtering module is used to output valid gas concentration values ​​and low-reliability data labels based on the sensor's health status. The default risk level assessment module is used to output a default risk level based on the amount of explosive charge and the time after blasting. The risk correction module is used to correct the default risk level based on sensor health status marking, low-reliability data marking, and monitoring blind zone marking, and output a comprehensive risk level. The risk propagation zoning module is used to divide risk propagation zones based on basic data of the roadway ventilation network and wind speed and direction monitoring data, and output the zone control priority and the comprehensive risk level after zone correction; The ventilation control strategy generation module is used to generate control instructions based on the comprehensive risk level and zone control priority after zoning correction. The control instructions include the main ventilation fan frequency setting value and the damper opening setting value. The execution and update module and the ventilation excitation readback module are used to execute control commands and update sensor health status markers, monitoring blind zone markers, and comprehensive risk levels; The closed-loop control module is used to control the above modules to execute sequentially and cyclically, forming a closed-loop control.

2. The smart mine integrated management system based on the Internet of Things as described in claim 1, characterized in that: The data access module establishes a mapping relationship between blasting parameter data, gas monitoring data, wind speed and direction monitoring data, ventilation equipment operation data, and basic roadway ventilation network data according to the monitoring point location and roadway number, airflow direction, and equipment number, and generates a data index table for risk assessment and zoning calculation in subsequent steps.

3. The smart mine integrated management system based on the Internet of Things as described in claim 2, characterized in that: The sensor health status determination module outputs sensor health status flags according to the following rules: When the sensor's output value is consistent with that of a health sensor in the same monitoring area in the direction of change and the fluctuation amplitude does not exceed the preset fluctuation threshold, it is marked as healthy; when the sensor's response time exceeds the preset response time threshold, or the deviation between the output value and the health sensor exceeds the preset deviation threshold, or the output value fluctuates continuously when the environment is stable, it is marked as sub-healthy; when the sensor experiences data interruption, continuous loss of heart rate, a continuously fixed output value, or an output value that exceeds the range, it is marked as faulty.

4. The smart mine integrated management system based on the Internet of Things as described in claim 3, characterized in that: The reliable data filtering module outputs the effective gas concentration value according to the following rules: when the main sensor's sensor health status is marked as healthy, the monitoring value collected by the main sensor is directly used; when the main sensor's sensor health status is marked as sub-healthy and there is a redundant verification sensor marked as healthy, the monitoring value collected by the redundant verification sensor is used; when the main sensor's sensor health status is marked as sub-healthy and there is no redundant verification sensor, the monitoring value collected by the main sensor is still used, but a low-reliability data label is added to the monitoring value; when the main sensor's sensor health status is marked as failed, the monitoring point is marked as a monitoring blind zone.

5. The smart mine integrated management system based on the Internet of Things as described in claim 4, characterized in that: The default risk level judgment module queries a preset comparison relationship based on the combination relationship between the charge amount and the time interval after blasting, and outputs a safe, low-risk, medium-risk, or high-risk level.

6. The smart mine integrated management system based on the Internet of Things as described in claim 5, characterized in that: The risk correction module corrects the default risk level according to the following rules: when any gas monitoring sensor at a key monitoring point is marked as faulty, the overall risk level is set to high risk; when there are no faulty sensors and two or more gas monitoring sensors are marked as sub-optimal, the default risk level is increased by one level; when an effective gas concentration value is marked with low-reliability data, the corresponding risk level is increased by one level; when the number of monitoring blind spots reaches a preset threshold, the overall risk level is set to high risk.

7. The smart mine integrated management system based on the Internet of Things as described in claim 6, characterized in that: The risk propagation zoning module determines the current effective airflow direction by matching the design airflow direction in the basic data of the roadway ventilation network with the real-time wind direction value in the wind speed and direction monitoring data. It then divides the risk propagation zones along the current effective airflow direction, judges the risk propagation status based on the changing trend of the effective gas concentration values ​​of adjacent zones, and raises the risk level of the downstream risk propagation zone by one level.

8. The smart mine integrated management system based on the Internet of Things as described in claim 7, characterized in that: The ventilation control strategy generation module selects the corresponding strategy from the daily economic ventilation strategy, mild enhanced ventilation strategy, moderate enhanced ventilation strategy, global enhanced ventilation strategy and maximum power emergency ventilation strategy according to the comprehensive risk level and zoning control priority after zoning correction, and outputs the corresponding control parameters.

9. The smart mine integrated management system based on the Internet of Things as described in claim 8, characterized in that: After the ventilation system's operating status changes, the ventilation excitation readback module takes the previously marked monitoring blind spots as the blind spots to be verified and receives new data collected by the sensors corresponding to the blind spots to be verified. When the direction of concentration change of the sensor is consistent with the direction of concentration change of the adjacent healthy sensor and the relative deviation of the concentration change does not exceed a preset threshold, the sensor's sensor health status mark is restored; otherwise, its sensor health status mark is maintained or reduced.

10. The smart mine integrated management system based on the Internet of Things as described in claim 9, characterized in that: The sensor health status judgment module and the trusted data filtering module are deployed and executed in the downhole edge computing gateway, while the default risk level judgment module, risk correction module, risk propagation zoning module and ventilation control strategy generation module are deployed and executed in the SCADA server of the ground monitoring center.

Images