Intelligent coal mine ventilation optimization regulation and control method and system based on digital twinning

Abstract

The invention discloses a coal mine intelligent ventilation optimization regulation and control method and system based on digital twinning, and belongs to the technical field of coal mine safety ventilation. The method comprises the following steps: determining a three-dimensional space topological relation of a coal mine ventilation system based on mine structure data, and mapping real-time ventilation state data to corresponding roadway nodes and branch paths to obtain a twin ventilation state; and performing simulation calculation based on a three-dimensional space topological relation to obtain a theoretical ventilation distribution state, comparing the twin ventilation state with the theoretical ventilation distribution state, determining an air volume deviation region and a gas concentration abnormal region, and obtaining ventilation deviation characteristics to perform virtual rehearsal. Determining an adjustment scheme meeting the ventilation demand constraint and the energy consumption constraint, and generating an optimization regulation and control instruction; and the entity ventilation facility is controlled to execute the adjustment action, and the three-dimensional space topological relation is updated through feedback data. Accurate regulation and dynamic optimization of the ventilation system are achieved, and the safety and the energy-saving effect of coal mine ventilation are improved.

Description

Technical Field

[0001] This invention relates to the field of coal mine safety ventilation technology, and in particular to a method and system for optimizing and controlling intelligent ventilation in coal mines based on digital twins. Background Technology

[0002] In coal mine safety production, the ventilation system, as the core system for ensuring the safety of underground workers and maintaining normal production order, directly affects the mine's safety level through its operational efficiency and control precision. Traditional coal mine ventilation management mainly relies on manual experience for airflow allocation and adjustment, obtaining ventilation parameters through regular inspections and manual measurements, and then conducting manual analysis and decision-making based on ventilation network calculation theory. However, with the continuous increase in coal mining depth and the increasing complexity of roadway layouts, the dynamic and nonlinear characteristics of the ventilation system are becoming increasingly prominent, making traditional ventilation management methods insufficient to meet the safety production needs of modern mines.

[0003] In recent years, with the development of IoT, sensor, and computer simulation technologies, coal mine ventilation systems have gradually transformed towards intelligence and automation. Some mines have already deployed ventilation parameter monitoring systems capable of collecting key parameters such as wind speed, wind pressure, and gas concentration in real time, and using computer software to perform ventilation network calculations, providing data support for ventilation management. Meanwhile, digital twin technology, as an emerging technology that deeply integrates physical entities with virtual models, has been widely applied in industrial manufacturing, urban management, and other fields, providing a new technological path for the intelligent management of coal mine ventilation systems. However, existing coal mine ventilation management technologies still have many shortcomings. First, there is a lack of effective spatial mapping mechanisms between the ventilation data collected by existing monitoring systems and the ventilation network model. Real-time monitoring data often exists in discrete point form, making it difficult to accurately reflect the spatial distribution of the entire ventilation system. This leads to a disconnect between the monitoring data and the theoretical calculation model, failing to truly depict the actual operating conditions of the ventilation system and affecting the accuracy and timeliness of ventilation anomaly identification. Existing ventilation control methods are primarily based on static ventilation network calculations. When adjusting ventilation facilities, there is a lack of mechanisms for pre-simulation verification of the adjustment effects. Adjustment plans often rely on the experience and judgment of engineers, making it impossible to accurately predict the ventilation effects after adjustment before implementation. This easily leads to problems such as insufficient airflow or energy waste in localized areas due to improper adjustment. The adjustments are often arbitrary and lack precision. In existing ventilation management systems, monitoring, analysis, decision-making, and execution are relatively independent, lacking a closed-loop feedback mechanism. Post-adjustment effect data cannot be promptly fed back into the simulation model for correction, causing a gradual accumulation of discrepancies between the theoretical model and the actual ventilation system. This leads to a continuous decline in the accuracy of simulation calculations, hindering dynamic optimization and continuous improvement of the ventilation system and limiting the level of intelligent ventilation management. Summary of the Invention

[0004] The embodiments of the present invention provide a method and system for optimizing and controlling intelligent ventilation in coal mines based on digital twins, which can at least solve some of the problems existing in the prior art.

[0005] A first aspect of this invention provides a method for optimizing and controlling intelligent ventilation in coal mines based on digital twins, comprising:

[0006] Acquire real-time ventilation status data and mine structure data in underground coal mines;

[0007] Based on the mine structure data, the three-dimensional spatial topology of the coal mine ventilation system is determined, and the real-time ventilation status data is mapped to the corresponding roadway nodes and branch paths according to spatial coordinates to obtain twin ventilation status.

[0008] Simulation calculations are performed based on the aforementioned three-dimensional spatial topological relationships to obtain the theoretical ventilation distribution state. The twin ventilation state is then compared with the theoretical ventilation distribution state to identify areas of airflow deviation and abnormal gas concentration, thereby obtaining ventilation deviation characteristics.

[0009] Based on the ventilation deviation characteristics, the adjustment state of the ventilation facilities is virtually simulated in the three-dimensional spatial topology to determine the adjustment scheme that meets the ventilation demand constraints and energy consumption constraints, and an optimized control command is generated based on the adjustment scheme.

[0010] Based on the optimized control command, the corresponding physical ventilation facility is controlled to perform adjustment actions, and real-time ventilation status data after adjustment is collected as feedback data. The feedback data is used to update the three-dimensional spatial topology to maintain the consistency between simulation calculation and actual ventilation conditions.

[0011] Based on the mine structure data, the three-dimensional spatial topology of the coal mine ventilation system is determined, and the real-time ventilation status data is mapped to the corresponding roadway nodes and branch paths according to spatial coordinates to obtain twin ventilation status, including:

[0012] Based on the roadway topology data in the mine structure data, the roadway connectivity is determined, and based on the ventilation facility location data in the mine structure data, the node locations and branch connections of the airflow path are determined, thus obtaining the three-dimensional spatial topology.

[0013] Extract the spatial coordinate information of each measuring point in the real-time ventilation status data, and calculate the spatial distance between the measuring point data and each roadway node in the three-dimensional spatial topology for the spatial coordinate information of each measuring point data.

[0014] Based on the spatial distance, a roadway node matching each measurement point data is selected, and the roadway node identifier of the matching roadway node is used as the roadway node identifier to which the corresponding measurement point data belongs;

[0015] Based on the roadway node identifier, the air volume distribution data, air pressure distribution data and gas concentration distribution data of the corresponding measuring point are associated with the branch paths between the roadway node and the adjacent nodes, respectively.

[0016] Based on the airflow continuity constraint, the real-time ventilation status data on the branch paths between adjacent nodes are interpolated to obtain the twin ventilation status corresponding to all nodes and branch paths in the three-dimensional spatial topology.

[0017] Simulation calculations are performed based on the aforementioned three-dimensional spatial topological relationships to obtain the theoretical ventilation distribution state. The twin ventilation state is then compared with the theoretical ventilation distribution state to identify areas of airflow deviation and abnormal gas concentration, thus obtaining ventilation deviation characteristics, including:

[0018] The roadway nodes in the three-dimensional spatial topology are taken as network nodes, and the branch paths are taken as network branches. The ventilation network is solved using the ventilation network solution equation to obtain the theoretical wind pressure value of each network node and the theoretical air volume value of each network branch, which are taken as the theoretical ventilation distribution state.

[0019] Based on the measured air volume values ​​of each roadway node in the twin ventilation state and the theoretical air volume values ​​of the corresponding network nodes in the theoretical ventilation distribution state, the air volume deviation area is determined.

[0020] The measured gas concentration values ​​of each roadway node in the twin ventilation state are compared with the ventilation safety concentration limit node by node to identify roadway nodes whose measured gas concentration values ​​exceed the ventilation safety concentration limit. Based on the spatially continuous distribution of the roadway nodes exceeding the limit and the nodes within their coverage area, the gas concentration abnormal area is determined.

[0021] The ventilation deviation characteristics are obtained by associating and combining the airflow deviation area, the gas concentration abnormal area, and the spatial range of the area.

[0022] Based on the measured air volume values ​​of each roadway node in the twin ventilation state and the theoretical air volume values ​​of the corresponding network nodes in the theoretical ventilation distribution state, the air volume deviation area is determined, including:

[0023] Based on the mapping relationship between the roadway node identifier in the twin ventilation state and the network node number in the theoretical ventilation distribution state, the measured air volume value of the roadway node with the same spatial location and the theoretical air volume value of the corresponding network node are determined.

[0024] The difference between the measured air volume value of each roadway node at the same spatial location and the theoretical air volume value of the corresponding network node is calculated to obtain the air volume deviation value of each roadway node;

[0025] The airflow deviation value is processed to obtain a relative deviation rate. The relative deviation rate is compared with a preset airflow deviation judgment criterion, and roadway nodes whose relative deviation rate exceeds the airflow deviation judgment criterion are identified as deviation nodes.

[0026] Based on the spatial connectivity between the deviation nodes in the three-dimensional spatial topology, deviation nodes with direct branch path connections are clustered into the same deviation region group;

[0027] Extract the node identifier, airflow deviation value, relative deviation rate, and branch path identifier connecting the deviation nodes from all deviation nodes in each deviation region group to obtain the airflow deviation region.

[0028] Based on the ventilation deviation characteristics, a virtual simulation of the ventilation facility's adjustment state is performed within the three-dimensional spatial topology to determine an adjustment scheme that satisfies ventilation demand constraints and energy consumption constraints. Based on this adjustment scheme, optimized control commands are generated, including:

[0029] Based on the ventilation deviation characteristics, the target area to be adjusted is determined in the three-dimensional spatial topology. Based on the connectivity path of the target area in the three-dimensional spatial topology, the candidate adjustment objects are determined as ventilation facilities that affect the ventilation state of the target area.

[0030] For each candidate adjustment object, multiple sets of virtual adjustment parameters are determined. These virtual adjustment parameters are applied to the boundary conditions of the corresponding ventilation facilities in the three-dimensional spatial topology. Virtual pre-simulation calculations are performed using the ventilation network solution equation to obtain the pre-simulation ventilation distribution state corresponding to each set of virtual adjustment parameters.

[0031] Extract the pre-simulated air volume and pre-simulated gas concentration values ​​of the target area from each pre-simulated ventilation distribution state, and screen out candidate adjustment parameter groups that meet ventilation demand constraints and ventilation safety concentration constraints based on the pre-simulated air volume and pre-simulated gas concentration values;

[0032] Calculate the total energy consumption of the ventilation system corresponding to each candidate set of adjustment parameters, and select a candidate set of adjustment parameters as the optimal set of adjustment parameters based on the total energy consumption of the ventilation system to obtain the adjustment scheme;

[0033] The optimal adjustment parameter set of each candidate adjustment object in the adjustment scheme is converted into the corresponding action command of the ventilation facility, and the optimized control command is generated.

[0034] For each candidate adjustment object, multiple sets of virtual adjustment parameters are determined. These virtual adjustment parameters are applied to the boundary conditions of the corresponding ventilation facilities in the three-dimensional spatial topology. Virtual pre-simulation calculations are performed using the ventilation network solution equation to obtain the pre-simulated ventilation distribution state corresponding to each set of virtual adjustment parameters, including:

[0035] Based on the facility type and current regulation status parameters of each candidate regulation object, determine the physical dimensions and value boundaries of the regulation parameters;

[0036] Based on the value boundary and the current adjustment state parameter, multiple virtual adjustment parameter values ​​are generated within the value boundary, and the virtual adjustment parameter values ​​are arranged in order to obtain a virtual adjustment parameter sequence.

[0037] Each virtual adjustment parameter value in the virtual adjustment parameter sequence is combined with the identifier of the candidate adjustment object to obtain multiple sets of virtual adjustment parameters;

[0038] In the three-dimensional spatial topology, the location of the ventilation facility corresponding to each candidate adjustment object is determined, and the boundary condition values ​​of the corresponding ventilation facility locations in the three-dimensional spatial topology are updated based on each set of virtual adjustment parameters.

[0039] For each set of virtual control parameters updated in three-dimensional space topology, the theoretical wind pressure values ​​of each network node and the theoretical air volume values ​​of each network branch corresponding to that set of virtual control parameters are obtained by solving the ventilation network equations.

[0040] The theoretical wind pressure values ​​of network nodes and the theoretical air volume values ​​of network branches corresponding to each set of virtual adjustment parameters are combined to determine the pre-simulated ventilation distribution state corresponding to each set of virtual adjustment parameters.

[0041] Based on the optimized control command, the corresponding physical ventilation facility is controlled to perform adjustment actions, and real-time ventilation status data after adjustment is collected as feedback data. The feedback data is used to update the three-dimensional spatial topology to maintain consistency between simulation calculations and actual ventilation conditions, including:

[0042] Based on the identifiers of each candidate control object in the optimized control command, the communication interface for querying the corresponding physical ventilation facility is determined;

[0043] The system sends control commands to the corresponding physical ventilation facilities through the communication interface, drives the physical ventilation facilities to perform adjustment actions, receives adjustment completion status signals from the physical ventilation facilities, and records the actual adjustment parameter values ​​of each physical ventilation facility.

[0044] After the physical ventilation system completes its adjustment, real-time ventilation status data at each monitoring point is collected via a real-time monitoring sensor network as the feedback data.

[0045] The actual adjustment response characteristics are obtained by comparing the feedback data with the actual adjustment parameter values.

[0046] Based on the actual adjustment response characteristics and the actual adjustment parameter values, the wind resistance characteristic parameters of the network nodes or branches where the ventilation facilities are located in the three-dimensional spatial topology are corrected, and the gas concentration measurement values ​​in the feedback data are updated to the gas concentration attribute values ​​of the corresponding roadway nodes in the three-dimensional spatial topology to maintain the consistency between the simulation calculation and the actual ventilation conditions.

[0047] A second aspect of this invention provides a digital twin-based intelligent ventilation optimization and control system for coal mines, comprising:

[0048] The first unit is used to acquire real-time ventilation status data and mine structure data in underground coal mines;

[0049] The second unit is used to determine the three-dimensional spatial topology of the coal mine ventilation system based on the mine structure data, and to map the real-time ventilation status data to the corresponding roadway nodes and branch paths according to spatial coordinates to obtain twin ventilation status;

[0050] The third unit is used to perform simulation calculations based on the three-dimensional spatial topological relationship to obtain the theoretical ventilation distribution state. The twin ventilation state is compared with the theoretical ventilation distribution state to determine the air volume deviation area and the gas concentration abnormal area, thereby obtaining the ventilation deviation characteristics.

[0051] The fourth unit is used to virtually simulate the adjustment state of the ventilation facilities in the three-dimensional spatial topology based on the ventilation deviation characteristics, determine the adjustment scheme that meets the ventilation demand constraints and energy consumption constraints, and generate optimized control instructions based on the adjustment scheme;

[0052] The fifth unit is used to control the corresponding physical ventilation facilities to perform adjustment actions based on the optimized control instructions, and to collect real-time ventilation status data after adjustment as feedback data. The feedback data is used to update the three-dimensional spatial topology to maintain the consistency between simulation calculations and actual ventilation conditions.

[0053] A third aspect of the present invention provides an electronic device, comprising:

[0054] processor;

[0055] Memory used to store processor-executable instructions;

[0056] The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0057] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0058] This invention constructs a digital twin model of a coal mine ventilation system, achieving a precise mapping between real-time ventilation status data and three-dimensional spatial topology. This model can intuitively reflect the actual operating status of the underground ventilation system, providing a visualized digital foundation for ventilation management and solving the problems of scattered information and unintuitive status perception in traditional ventilation management.

[0059] This invention, by comparing the twin ventilation state with the theoretical ventilation distribution state, can promptly identify areas of airflow deviation and abnormal gas concentration. Based on the three-dimensional spatial topological relationship, it can perform virtual pre-simulation of ventilation facility adjustment, evaluate the adjustment effect before actual control, and ensure that the adjustment scheme simultaneously meets ventilation demand constraints and energy consumption constraints. This effectively avoids the safety risks and energy waste that may be caused by blind adjustment, and improves the scientificity and accuracy of ventilation control.

[0060] This invention establishes a closed-loop control mechanism from data acquisition, deviation identification, scheme optimization to execution feedback. By using the adjusted real-time ventilation status data as feedback to continuously update the three-dimensional spatial topology, it maintains the dynamic consistency between the digital twin model and the actual ventilation system, realizes the adaptive optimization and adjustment of the ventilation system, and improves the intelligent management level and operating efficiency of the coal mine ventilation system. Attached Figure Description

[0061] Figure 1 This is a flowchart illustrating the intelligent ventilation optimization and control method for coal mines based on digital twins, according to an embodiment of the present invention.

[0062] Figure 2 This is a schematic diagram of the process for determining the airflow deviation region according to an embodiment of the present invention. Detailed Implementation

[0063] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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.

[0064] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0065] Figure 1 This is a schematic flowchart of a digital twin-based intelligent ventilation optimization and control method for coal mines, according to an embodiment of the present invention. Figure 1 As shown, the method includes:

[0066] Acquire real-time ventilation status data and mine structure data in underground coal mines;

[0067] Based on the mine structure data, the three-dimensional spatial topology of the coal mine ventilation system is determined, and the real-time ventilation status data is mapped to the corresponding roadway nodes and branch paths according to spatial coordinates to obtain twin ventilation status.

[0068] Simulation calculations are performed based on the aforementioned three-dimensional spatial topological relationships to obtain the theoretical ventilation distribution state. The twin ventilation state is then compared with the theoretical ventilation distribution state to identify areas of airflow deviation and abnormal gas concentration, thereby obtaining ventilation deviation characteristics.

[0069] Based on the ventilation deviation characteristics, the adjustment state of the ventilation facilities is virtually simulated in the three-dimensional spatial topology to determine the adjustment scheme that meets the ventilation demand constraints and energy consumption constraints, and an optimized control command is generated based on the adjustment scheme.

[0070] Based on the optimized control command, the corresponding physical ventilation facility is controlled to perform adjustment actions, and real-time ventilation status data after adjustment is collected as feedback data. The feedback data is used to update the three-dimensional spatial topology to maintain the consistency between simulation calculation and actual ventilation conditions.

[0071] In one optional implementation, the three-dimensional spatial topology of the coal mine ventilation system is determined based on the mine structure data, and the real-time ventilation status data is mapped to the corresponding roadway nodes and branch paths according to spatial coordinates to obtain twin ventilation status, including:

[0072] Based on the roadway topology data in the mine structure data, the roadway connectivity is determined, and based on the ventilation facility location data in the mine structure data, the node locations and branch connections of the airflow path are determined, thus obtaining the three-dimensional spatial topology.

[0073] Extract the spatial coordinate information of each measuring point in the real-time ventilation status data, and calculate the spatial distance between the measuring point data and each roadway node in the three-dimensional spatial topology for the spatial coordinate information of each measuring point data.

[0074] Based on the spatial distance, a roadway node matching each measurement point data is selected, and the roadway node identifier of the matching roadway node is used as the roadway node identifier to which the corresponding measurement point data belongs;

[0075] Based on the roadway node identifier, the air volume distribution data, air pressure distribution data and gas concentration distribution data of the corresponding measuring point are associated with the branch paths between the roadway node and the adjacent nodes, respectively.

[0076] Based on the airflow continuity constraint, the real-time ventilation status data on the branch paths between adjacent nodes are interpolated to obtain the twin ventilation status corresponding to all nodes and branch paths in the three-dimensional spatial topology.

[0077] In establishing the three-dimensional spatial topology of a coal mine ventilation system, the first step is to read the roadway topology data from the mine structure data. This data is stored in a data table, where each record contains attribute information such as roadway number, starting coordinates, ending coordinates, roadway cross-sectional area, and roadway length. Roadway numbers use a hierarchical coding method; for example, a main haulage roadway is numbered H001, and its branch roadways are numbered H001-01, H001-02, etc. The data table is parsed to extract the connection relationships between adjacent roadways. Roadway connectivity is determined by identifying the coincidence point between the starting and ending coordinates of the roadways. When the ending coordinates of roadway H001 are X-coordinate 800 meters, Y-coordinate 200 meters, and Z-coordinate -450 meters, and the starting coordinates of roadway H002 are exactly the same, it is determined that the two roadways are connected at that spatial location, and a connection record is established in the data structure.

[0078] The system continues reading ventilation facility location data, which records the specific installation locations of ventilation facilities such as dampers, air windows, regulating air windows, and air bridges. Each ventilation facility record includes information such as facility number, facility type, installation coordinates, affiliated roadway number, and facility status. For example, a damper facility with the number FM-023 is installed at coordinates X: 450 meters, Y: 180 meters, Z: -420 meters, and belongs to roadway H003; its current status is closed. Based on the locations of these ventilation facilities, the original roadway nodes are subdivided, creating a new topology node at each ventilation facility installation location. Roadway H003, which originally connected directly from point A to point B, is divided into three parts after the installation of damper FM-023: the node from point A to the damper location, the damper node, and the node from the damper location to point B. This subdivision ensures that the airflow path accurately reflects the ventilation facility's control over airflow.

[0079] The established three-dimensional spatial topology is stored in a graph structure. The node data structure includes fields such as node identifier, three-dimensional spatial coordinates, node type, and a list of connections. Node types are categorized into roadway intersection nodes, ventilation facility nodes, and working face nodes. The branch path data structure includes attributes such as path identifier, starting node identifier, ending node identifier, path length, roadway cross-sectional area, and wind resistance coefficient. A specific topological node record is identified as node N-156, with spatial coordinates of X coordinate 620 meters, Y coordinate 240 meters, and Z coordinate -380 meters. The node type is a roadway intersection node, and the connection list contains four adjacent nodes: N-155, N-157, N-201, and N-202.

[0080] Real-time ventilation status data is collected from sensors distributed throughout the mine. Each sensor point generates a data record containing information such as the point number, collection timestamp, spatial coordinates, wind speed, wind pressure, oxygen concentration, carbon monoxide concentration, and methane concentration. For example, sensor number S-089 is located at X coordinate 615 meters, Y coordinate 238 meters, and Z coordinate -382 meters. At a certain moment, it collects the following data: wind speed 2.8 meters per second, wind pressure 1080 Pascals, oxygen concentration 19.5%, carbon monoxide concentration 0.0008%, and methane concentration 0.35%.

[0081] For the spatial coordinates of measuring point S-089, all nodes in the three-dimensional spatial topology are traversed, and the spatial distances from the measuring point to each node are calculated. The calculation method is to calculate the coordinate differences between the measuring point and the node coordinates in the X, Y, and Z directions, square each of the three differences, add the three squares together, and take the square root to obtain the spatial straight-line distance. The distance from measuring point S-089 to node N-156 is calculated as follows: X-direction difference 5 meters, Y-direction difference 2 meters, Z-direction difference 2 meters, sum of squares 29, and square root, the distance is approximately 5.39 meters. After calculating the distances from the measuring point to all nodes, the node with the smallest distance is selected as the matching node. The distance of 5.39 meters between measuring point S-089 and node N-156 is the minimum value among all calculated distances; therefore, node identifier N-156 is selected as the roadway node identifier to which the measuring point belongs.

[0082] After establishing the association between the measuring point data and the node identifier, the various ventilation parameters of the measuring point are assigned to the corresponding topology. The wind speed data of measuring point S-089 (2.8 m / s) is multiplied by the roadway cross-sectional area data. Assuming the roadway cross-sectional area at this location is 12 square meters, the calculated airflow at this location is 33.6 cubic meters per second. This airflow data is associated with node N-156 and also assigned to the branch paths adjacent to this node. Node N-156 connects four branch paths, and the applicable branch paths are determined based on the airflow direction. By analyzing the wind pressure gradient direction, it is determined that the airflow flows from node N-155 to node N-156, and then branches from node N-156 to nodes N-157 and N-201. Therefore, the data of measuring point S-089 is associated with paths N-155 to N-156, N-156 to N-157, and N-156 to N-201.

[0083] For branch paths where sensor measurement points are not directly deployed, interpolation is used to supplement ventilation status data. Path P-078 connects nodes N-156 and N-157, with a path length of 85 meters. At the starting node N-156, the airflow data for measurement point S-089 is 33.6 cubic meters per second, and at the ending node N-157, the airflow data for measurement point S-105 is 32.8 cubic meters per second. Multiple virtual interpolation points are established along the path, one every 10 meters, for a total of 8 interpolation points. The airflow value of the interpolation point 10 meters from the starting point is calculated by weighting the distances from that point to both ends of the path. The weight for the 10-meter distance from the starting point is 75 meters, and the weight for the 75-meter distance from the ending point is 10 meters. The airflow value is calculated as the airflow difference of 0.8 cubic meters per second multiplied by the distance ratio 10, divided by 85, and then the starting point airflow is subtracted from this product, resulting in an interpolated airflow of 33.51 cubic meters per second.

[0084] The interpolation calculation must satisfy the airflow continuity constraint, meaning the total inflow airflow at any node should equal the total outflow airflow. Node N-156 has one inflow branch path with an airflow of 33.6 cubic meters per second and two outlet branch paths. The calculation checks if the sum of the airflows of these two outlet paths equals the inflow airflow. When the sum of the outlet path airflows at node N-156 is detected to be 32.1 cubic meters per second, differing from the inflow airflow of 33.6 cubic meters per second by 1.5 cubic meters per second, the difference is adjusted according to the initial airflow ratio of each outlet path. The initial airflow of outlet path P-078 is 20.3 cubic meters per second, and the initial airflow of outlet path P-079 is 11.8 cubic meters per second, with an airflow ratio of approximately 1.72:1. The air volume difference of 1.5 cubic meters per second is allocated proportionally, increasing the air volume of path P-078 by 0.95 cubic meters per second and path P-079 by 0.55 cubic meters per second. After adjustment, the total air volume of the two outlet paths reaches 33.6 cubic meters per second, which satisfies the continuity constraint.

[0085] The interpolation calculation of gas concentration distribution data takes into account gas diffusion and mixing characteristics. The gas concentration at the starting point of path P-078 is 0.35%, and the gas concentration at the ending point is 0.42%. The gas concentration at the intermediate interpolation point is affected not only by the concentrations at both ends but also by whether there is a gas emission source in this section of the roadway. Reading the mine geological data, a gas emission was identified near the working face of this path, with an emission rate of 0.8 cubic meters per minute. When calculating the gas concentration at the interpolation point 40 meters from the starting point, in addition to calculating the base concentration by distance weighting, the contribution of the accumulated emission gas in this section to the concentration is also added. The emission gas volume divided by the ventilation volume in this section yields a concentration increment of 0.04%, which is then added to the base interpolation concentration to obtain the final gas concentration value at this point.

[0086] The interpolation of wind pressure distribution data needs to consider frictional losses along the route. The wind pressure at the starting point of path P-078 is 1080 Pascals, and the wind pressure at the ending point is 1065 Pascals. The pressure difference of 15 Pascals is caused by the frictional resistance of this section of the tunnel. The theoretical pressure drop is calculated based on the tunnel perimeter, tunnel length, wind speed, and frictional resistance coefficient. The tunnel perimeter is 14 meters, the length is 85 meters, the average wind speed is 2.75 meters per second, and the frictional resistance coefficient is 0.018 kg per cubic meter. Substituting these parameters into the resistance calculation method, the theoretical pressure drop is 14.8 Pascals, which is basically consistent with the measured pressure difference of 15 Pascals. The wind pressure value at the interpolation point is obtained by subtracting the cumulative resistance loss from the wind pressure at the starting point segment by segment. The cumulative resistance loss at the interpolation point 40 meters away from the starting point is the total resistance loss of 15 Pascal multiplied by the distance ratio of 40 and divided by 85, which gives 7 Pascal. The wind pressure at this point is the starting wind pressure of 1080 Pascal minus 7 Pascal, and the calculated result is 1073 Pascal.

[0087] After completing the data mapping and interpolation calculations for all nodes and branch paths, a complete twin ventilation status dataset is generated. This dataset, indexed by timestamps, records the airflow, air pressure, and concentrations of various gases at each node in the three-dimensional spatial topology at a given moment, as well as the detailed distribution of airflow parameters on each branch path. This twin ventilation status data enables real-time mapping of the physical ventilation system in digital space, providing a data foundation for the analysis, early warning, and optimized control of the ventilation system.

[0088] In one optional implementation, simulation calculations are performed based on the three-dimensional spatial topological relationship to obtain the theoretical ventilation distribution state. The twin ventilation state is compared with the theoretical ventilation distribution state to determine the airflow deviation region and the gas concentration abnormal region, thereby obtaining ventilation deviation characteristics, including:

[0089] The roadway nodes in the three-dimensional spatial topology are taken as network nodes, and the branch paths are taken as network branches. The ventilation network is solved using the ventilation network solution equation to obtain the theoretical wind pressure value of each network node and the theoretical air volume value of each network branch, which are taken as the theoretical ventilation distribution state.

[0090] Based on the measured air volume values ​​of each roadway node in the twin ventilation state and the theoretical air volume values ​​of the corresponding network nodes in the theoretical ventilation distribution state, the air volume deviation area is determined.

[0091] The measured gas concentration values ​​of each roadway node in the twin ventilation state are compared with the ventilation safety concentration limit node by node to identify roadway nodes whose measured gas concentration values ​​exceed the ventilation safety concentration limit. Based on the spatially continuous distribution of the roadway nodes exceeding the limit and the nodes within their coverage area, the gas concentration abnormal area is determined.

[0092] The ventilation deviation characteristics are obtained by associating and combining the airflow deviation area, the gas concentration abnormal area, and the spatial range of the area.

[0093] During simulation calculations, the three-dimensional spatial topology of the mine ventilation system is converted into a computable network model. Specifically, each roadway node is mapped to a network node, and the branch paths connecting two roadway nodes are mapped to network branches. For example, in a mine ventilation system, the main intake shaft is designated as network node N001. This node connects to the mining area intake airway via the main haulage roadway, and this section of roadway is designated as network branch B001. The entrance to the mining area intake airway is designated as network node N002. Following this method, all roadway nodes and branch paths of the entire ventilation system are mapped and converted one by one.

[0094] After constructing the network model, the solution is performed based on the ventilation network equations, which are grounded in the laws of airflow continuity and energy conservation in ventilation systems. At each network node, the sum of the airflow volumes flowing into that node equals the sum of the airflow volumes flowing out of that node. In any closed loop, the algebraic sum of the air pressures along the loop is zero. The resistance characteristics of each network branch are also considered; the branch pressure drop equals the product of the branch resistance and the square of the airflow volume. By establishing a system of equations encompassing all network nodes and branches, an iterative solution is used to obtain a convergent solution. The convergence criterion is set as the rate of change of airflow between two consecutive iterations being less than one-thousandth.

[0095] After iterative calculations, the theoretical wind pressure values ​​for each network node and the theoretical airflow values ​​for each network branch are obtained. For example, the theoretical wind pressure value for network node N002 is 1580 Pa, the theoretical airflow value for network branch B001 is 280 cubic meters per second, the theoretical wind pressure value for network node N015 is 1320 Pa, and the theoretical airflow value for network branch B014 is 65 cubic meters per second. These calculation results constitute a complete theoretical ventilation distribution state, reflecting the ventilation state that should be achieved under the existing ventilation system topology and equipment configuration.

[0096] The real-time collected twin ventilation status data is compared and analyzed with the theoretical ventilation distribution. For each roadway node, the measured air volume value in the twin ventilation status is extracted, and the theoretical air volume value of the corresponding network node in the theoretical ventilation distribution status is also obtained. The deviation between the two is calculated, which is equal to the measured air volume value minus the theoretical air volume value. The deviation rate is also calculated, which is equal to the deviation divided by the theoretical air volume value and then multiplied by 100%. For example, if the measured air volume value of roadway node D025 is 58 cubic meters per second, and the theoretical air volume value of the corresponding network node is 65 cubic meters per second, then the deviation is -7 cubic meters per second, and the deviation rate is -10.8%.

[0097] The airflow deviation judgment threshold is preset, for example, 15%. When the absolute value of the airflow deviation rate of a certain roadway node exceeds this threshold, the node is marked as an airflow deviation node. After traversing all roadway nodes to complete the deviation judgment, all airflow deviation nodes are identified. Based on the connection relationship in the three-dimensional spatial topology, spatially adjacent or connected airflow deviation nodes are found, and these nodes and their respective roadway segments are combined to form an airflow deviation region. For example, if the airflow deviation rates of roadway nodes D025, D026, and D027 are -10.8%, -16.3%, and -14.2% respectively, and these three nodes are spatially continuous in the same mining area's return airway, then these three nodes and their respective mining area return airway segments are designated as the airflow deviation region R01.

[0098] To identify abnormal gas concentrations, measured gas concentration values ​​for each roadway node were extracted from the twin ventilation data. The main monitored gas types included methane, carbon monoxide, and carbon dioxide. Concentration limits for each gas were set according to ventilation safety standards; for example, the methane concentration limit in the return airway of the mining area was 1%, the methane concentration limit in the mining face was 1%, and the carbon monoxide concentration limit was 24 / 10000. The measured gas concentration value for each roadway node was compared with the corresponding location and type concentration limit.

[0099] When the measured gas concentration at a roadway node exceeds the ventilation safety concentration limit, the node is marked as a gas concentration exceeding the limit. For example, if the measured methane concentration at roadway node D032 is 1.3%, exceeding the 1% concentration limit for that location, then node D032 is marked as a gas concentration exceeding the limit. After traversing all roadway nodes and completing the concentration comparison, the distribution of all gas concentration exceeding the limit nodes is obtained.

[0100] The system identifies continuously distributed nodes with excessive gas concentrations in three-dimensional space. Continuity is determined by a direct roadway connection between the nodes and a distance within a set range, set at 50 meters. Two nodes with excessive gas concentrations are considered spatially continuous if the roadway distance between them is less than 50 meters. Multiple spatially continuous nodes with excessive concentrations are identified as a cluster, and the influence coverage area of ​​this cluster is determined based on the airflow direction. The coverage area includes the location of the excessive node and all nodes within a certain distance downwind of it, with a coverage distance set at 100 meters. For example, if nodes D032, D033, and D034 all have excessive methane concentrations and are spatially continuous, these three nodes, along with nodes D035 and D036 within 100 meters downwind of them, together constitute the gas concentration anomaly area A01.

[0101] After obtaining the airflow deviation area and the gas concentration anomaly area, a correlation and combination analysis is performed to determine whether the two types of areas overlap or are adjacent in space. When the airflow deviation area and the gas concentration anomaly area intersect spatially, or the boundary distance between the two areas is less than 30 meters, they are considered to be correlated. The spatial coordinates of the correlated area, the involved roadway node numbers, the numerical characteristics of the airflow deviation, and the type and degree of gas concentration exceeding the limit are recorded. For example, the airflow deviation area R01 and the gas concentration anomaly area A01 partially overlap, with the overlapping area including nodes D025 to D034. The airflow deviation rate in this area is -15% to -10.8%, and the methane concentration exceeds the limit by 0.2% to 0.3%. This comprehensive information is integrated to form a ventilation deviation feature, which fully describes the location, nature, severity, and spatial distribution of the problem area in the ventilation system, providing accurate data support for subsequent ventilation system control and safety early warning.

[0102] In one optional implementation, the airflow deviation area is determined by calculating the measured airflow value of each roadway node in the twin ventilation state and the theoretical airflow value of the corresponding network node in the theoretical ventilation distribution state, including:

[0103] Based on the mapping relationship between the roadway node identifier in the twin ventilation state and the network node number in the theoretical ventilation distribution state, the measured air volume value of the roadway node with the same spatial location and the theoretical air volume value of the corresponding network node are determined.

[0104] The difference between the measured air volume value of each roadway node at the same spatial location and the theoretical air volume value of the corresponding network node is calculated to obtain the air volume deviation value of each roadway node;

[0105] The airflow deviation value is processed to obtain a relative deviation rate. The relative deviation rate is compared with a preset airflow deviation judgment criterion, and roadway nodes whose relative deviation rate exceeds the airflow deviation judgment criterion are identified as deviation nodes.

[0106] Based on the spatial connectivity between the deviation nodes in the three-dimensional spatial topology, deviation nodes with direct branch path connections are clustered into the same deviation region group;

[0107] Extract the node identifier, airflow deviation value, relative deviation rate, and branch path identifier connecting the deviation nodes from all deviation nodes in each deviation region group to obtain the airflow deviation region.

[0108] Figure 2 This is a schematic diagram illustrating the process of determining the airflow deviation region according to an embodiment of the present invention. Figure 2As shown, in determining the airflow deviation area, it is necessary to systematically compare and analyze the measured airflow values ​​of each roadway node in the twin ventilation state with the theoretical airflow values ​​of the corresponding network nodes in the theoretical ventilation distribution state. This process relies on a pre-established mapping table between roadway node identifiers and network node numbers. This mapping table records the unique identifier of each roadway node and its corresponding network node number, ensuring that nodes with the same spatial location can be accurately matched. Taking a mine ventilation system as an example, the identifier of roadway node A-101 corresponds to node number N-056 in the theoretical ventilation network. Both are located at the intersection of the third level transport roadway and the return airway in the east wing of the mine. The spatial coordinates of this location are X coordinate 2350 meters, Y coordinate 1820 meters, and Z coordinate -580 meters.

[0109] After establishing the mapping relationship, the measured air volume data and the theoretical air volume data of the corresponding network node are extracted one by one. For roadway node A-101, the measured air volume value is calculated to be 42.3 cubic meters per second using real-time data collected by the wind speed and wind pressure sensors installed at that location, combined with the roadway cross-sectional area. In the theoretical ventilation distribution state, the theoretical air volume value of network node N-056, determined according to the ventilation resistance calculation and air volume distribution principle, is 38.5 cubic meters per second. Similarly, roadway node A-102 corresponds to network node N-057, with a measured air volume value of 55.8 cubic meters per second and a theoretical air volume value of 60.2 cubic meters per second; roadway node A-103 corresponds to network node N-058, with a measured air volume value of 28.6 cubic meters per second and a theoretical air volume value of 30.1 cubic meters per second.

[0110] After acquiring the airflow data for all paired nodes, the difference between the measured airflow value and the theoretical airflow value of the corresponding network node at each spatial location is calculated. For node A-101, the measured airflow value of 42.3 cubic meters per second is subtracted from the theoretical airflow value of 38.5 cubic meters per second, resulting in an airflow deviation of 3.8 cubic meters per second, indicating that the actual ventilation volume of this node is higher than the theoretical design value. For node A-102, the measured airflow value of 55.8 cubic meters per second is subtracted from the theoretical airflow value of 60.2 cubic meters per second, resulting in an airflow deviation of -4.4 cubic meters per second, indicating that the actual ventilation volume of this node is lower than the theoretical design value. The calculated airflow deviation value for node A-103 is -1.5 cubic meters per second. By traversing all mapped nodes, a complete airflow deviation value dataset is formed.

[0111] To eliminate the comparability of deviation values ​​between different roadway nodes due to varying theoretical airflow benchmarks, absolute deviation values ​​need to be converted into relative deviation rates. The relative deviation rate is calculated by dividing the airflow deviation value by the corresponding theoretical airflow value and then multiplying by 100%. For node A-101, the airflow deviation value of 3.8 cubic meters per second divided by the theoretical airflow value of 38.5 cubic meters per second yields a quotient of 0.0987, which, multiplied by 100%, results in a relative deviation rate of 9.87%. The relative deviation rate for node A-102 is calculated as -4.4 divided by 60.2 and then multiplied by 100%, resulting in -7.31%. The relative deviation rate for node A-103 is -1.5 divided by 30.1 and then multiplied by 100%, resulting in -4.98%. This normalization process makes the degree of deviation between nodes of different airflow levels comparable.

[0112] After determining the relative deviation rate, the relative deviation rate of each node is compared one by one with the preset airflow deviation judgment criterion. This criterion is set according to the mine safety regulations and ventilation management requirements, and is usually set at ±8% as the acceptable deviation range threshold. When the absolute value of the relative deviation rate of a node exceeds 8%, the node is marked as a deviation node. In the above case, the relative deviation rate of node A-101, 9.87%, exceeds the 8% threshold and is marked as a deviation node. The absolute value of the relative deviation rate of node A-102, -7.31%, is 7.31%, which does not exceed the threshold and is not a deviation node. Further inspection revealed that the relative deviation rate of node A-105 is 12.3%, the relative deviation rate of node A-108 is -10.6%, and the relative deviation rate of node A-112 is 15.8%, and these nodes were all identified as deviation nodes.

[0113] After identifying all deviation nodes, it is necessary to analyze the spatial connectivity between these deviation nodes based on the three-dimensional spatial topology. The three-dimensional spatial topology records the connection between nodes in the tunnel network through branch paths, including whether there are direct tunnel branches connecting the nodes and the path identifiers of the branches. For deviation nodes A-101 and A-105, querying the three-dimensional spatial topology reveals that there is a direct connection between them through branch path P-231, which is a 120-meter-long transport tunnel segment. It is also found that deviation nodes A-105 and A-108 are connected by branch path P-245, with a path length of 85 meters. Deviation node A-112 does not have a direct branch path connection with other identified deviation nodes; its nearest neighbor is a non-deviation node.

[0114] Based on spatial connectivity, clustering was used to group deviation nodes with direct branch paths into the same deviation region group. Deviation nodes A-101, A-105, and A-108, forming a continuous ventilation path through branch paths P-231 and P-245, were clustered into deviation region group G-01. This region group covers a major ventilation path on the third level of the east wing, containing three deviation nodes and two connecting branches. Deviation node A-112, being spatially isolated, constitutes a separate deviation region group G-02. If subsequent analysis reveals a branch path P-287 connecting deviation nodes A-115 and A-112, then A-115 will be incorporated into deviation region group G-02.

[0115] After clustering the deviation region groups, detailed feature information for each group was extracted. For deviation region group G-01, all deviation nodes were identified as A-101, A-105, and A-108, with airflow deviation values ​​of 3.8 m / s, 4.7 m / s, and -6.4 m / s, respectively, and relative deviation rates of 9.87%, 12.3%, and -10.6%. The branch paths connecting these deviation nodes were identified as P-231 and P-245, corresponding to specific tunnel segments. For deviation region group G-02, the identifier for deviation node A-112 was extracted, with an airflow deviation value of 9.5 m / s and a relative deviation rate of 15.8%. No branch path information is currently available for this region group. By integrating the node identifiers, airflow deviation values, relative deviation rates, and branch path identifiers, structured airflow deviation region data was formed, providing precise spatial positioning and quantitative basis for subsequent optimization and adjustment of the ventilation system.

[0116] In one optional implementation, based on the ventilation deviation characteristics, the adjustment state of the ventilation facility is virtually simulated in the three-dimensional spatial topology to determine an adjustment scheme that meets ventilation demand constraints and energy consumption constraints. Based on the adjustment scheme, optimized control instructions are generated, including:

[0117] Based on the ventilation deviation characteristics, the target area to be adjusted is determined in the three-dimensional spatial topology. Based on the connectivity path of the target area in the three-dimensional spatial topology, the candidate adjustment objects are determined as ventilation facilities that affect the ventilation state of the target area.

[0118] For each candidate adjustment object, multiple sets of virtual adjustment parameters are determined. These virtual adjustment parameters are applied to the boundary conditions of the corresponding ventilation facilities in the three-dimensional spatial topology. Virtual pre-simulation calculations are performed using the ventilation network solution equation to obtain the pre-simulation ventilation distribution state corresponding to each set of virtual adjustment parameters.

[0119] Extract the pre-simulated air volume and pre-simulated gas concentration values ​​of the target area from each pre-simulated ventilation distribution state, and screen out candidate adjustment parameter groups that meet ventilation demand constraints and ventilation safety concentration constraints based on the pre-simulated air volume and pre-simulated gas concentration values;

[0120] Calculate the total energy consumption of the ventilation system corresponding to each candidate set of adjustment parameters, and select a candidate set of adjustment parameters as the optimal set of adjustment parameters based on the total energy consumption of the ventilation system to obtain the adjustment scheme;

[0121] The optimal adjustment parameter set of each candidate adjustment object in the adjustment scheme is converted into the corresponding action command of the ventilation facility, and the optimized control command is generated.

[0122] During the automatic optimization and control of the mine ventilation system, when a ventilation deviation characteristic of insufficient airflow is detected in a certain area, the abnormal area is immediately located in the established three-dimensional spatial topology model. For example, assuming that the actual airflow monitored at a coal face at the -450-meter level in a mine is 18 cubic meters per second, while the safe operating requirement for this area is 25 cubic meters per second, this coal face is identified as the target area to be adjusted. The control system traces the upstream ventilation path of this target area in the three-dimensional spatial topology, identifying all ventilation nodes from the main ventilation room to the working face, including main duct branches, regulating windows, local ventilators, and ventilation doors. Through topology analysis, candidate adjustment objects affecting the ventilation status of this target area are identified, including the third regulating window 300 meters away from the target area, the local ventilator dedicated to this area, and the auxiliary ventilation door 150 meters upstream.

[0123] For the candidate regulating window No. 3, five sets of virtual regulating parameters were generated, with window openings set to 30%, 45%, 60%, 75%, and 90%, respectively. The corresponding ventilation resistance coefficients for each set of virtual regulating parameters were set to 8.5, 6.2, 4.1, 2.8, and 1.9 Pascals per cubic meter, respectively. These virtual regulating parameters were then applied sequentially to the boundary conditions of the corresponding ventilation facility nodes in the three-dimensional spatial topology model. When the window opening was set to 60%, the ventilation resistance coefficient of that node was modified to 4.1, while keeping the boundary conditions of other ventilation facilities unchanged. The ventilation network solution engine was then invoked, performing iterative calculations based on the principles of node airflow balance and loop air pressure balance. During the calculation, the entire ventilation network was divided into 127 nodes and 156 branches. An air volume conservation equation was established for each node, requiring that the total air volume flowing into the node equals the total air volume flowing out of the node. An air pressure balance equation was established for each independent loop, requiring that the algebraic sum of the products of the ventilation resistance of each branch in the loop and the square of the air volume be zero.

[0124] After the virtual simulation calculation was completed, the simulated airflow value for the target area was 22 cubic meters per second when the ventilation window opening was 60%, with a methane concentration of 0.65% and a carbon monoxide concentration of 15 ppm. Further calculations were performed for other opening combinations. When the opening was 75%, the simulated airflow for the target area reached 26 cubic meters per second, with the methane concentration decreasing to 0.48% and the carbon monoxide concentration decreasing to 11 ppm. The simulated ventilation distribution corresponding to all five sets of virtual adjustment parameters was extracted, and data extraction and safety assessment were conducted for the target area. According to the ventilation requirements of the mine, the minimum airflow requirement for the coal face is 25 cubic meters per second, with a methane concentration below 0.8% and a carbon monoxide concentration below 20 ppm. Comparison of the simulated data revealed that the simulated airflow was only 19 cubic meters per second when the opening was 30%, and 21 cubic meters per second when the opening was 45%, both failing to meet the minimum airflow requirements. These two sets of parameters were therefore discarded. The pre-simulated air volumes corresponding to the three sets of parameters with opening degrees of 60%, 75%, and 90% were 22 cubic meters per second, 26 cubic meters per second, and 28 cubic meters per second, respectively. Among them, the pre-simulated air volume with an opening degree of 60% still did not meet the standard. In the end, only the two sets of parameters with opening degrees of 75% and 90% met the ventilation demand constraints and gas concentration safety constraints, and were determined as candidate adjustment parameter groups.

[0125] For the local ventilation fan as a candidate for regulation, four sets of virtual regulation parameters were set: fan speeds of 800 rpm, 950 rpm, 1100 rpm, and 1250 rpm. In the virtual simulation calculation, different speeds corresponded to different fan characteristic curves, and the boundary conditions of this facility node were updated according to the relationship between fan pressure and air volume. When the speed was 1100 rpm, the virtual simulation showed that the air volume in the target area reached 27 cubic meters per second, and the methane concentration was 0.52%, meeting the safety constraints. This parameter set was recorded, and other speed combinations were further evaluated.

[0126] After obtaining the candidate adjustment parameter sets for all candidate adjustment objects, the total energy consumption of the ventilation system corresponding to each parameter combination scheme is calculated. For the combination scheme of adjusting the third air window opening to 75% and the local ventilation fan speed to 1100 rpm, the operating power of the main ventilation fan in the entire mine is 420 kW, the total power of all local ventilation fans is 185 kW, and the energy consumption of the compressed air power damper is 12 kW. The total energy consumption of the ventilation system for this scheme is 617 kW. When the air window opening is adjusted to 90% and the local ventilation fan speed is reduced to 950 rpm, the power of the main ventilation fan drops to 405 kW, the total power of the local ventilation fans drops to 158 kW, and the energy consumption of the compressed air power equipment is 12 kW. The total energy consumption of this scheme is 575 kW. Comparing the energy consumption values ​​of all candidate adjustment parameter groups that meet the safety constraints, the parameter combination with the lowest energy consumption value is selected as the optimal adjustment parameter group. The adjustment scheme is determined as follows: the opening of the third adjustment window is set to 90%, the speed of the local ventilation fan is set to 950 revolutions per minute, and the auxiliary damper is kept in its current open state.

[0127] The optimal adjustment parameter set in the adjustment scheme is converted into specific equipment action commands. For the third regulating window, the generated action command includes equipment address code 03A7, command type "opening adjustment," target opening value of 90%, and execution time set to 180 seconds. This command is sent to the window controller via fieldbus. For the local ventilation fan, the generated action command includes equipment address code 05B2, command type "variable frequency speed control," target speed of 950 rpm, and acceleration time set to 30 seconds. This command is sent to the variable frequency control cabinet via industrial Ethernet. The optimized control command integrates the action commands of all candidate regulating objects and sets execution priority and timing relationships to ensure that each ventilation facility completes the adjustment action in a predetermined order, achieving precise optimized control of the ventilation status of the target area.

[0128] In one optional implementation, multiple sets of virtual adjustment parameters are determined for each candidate adjustment object. These virtual adjustment parameters are applied to the boundary conditions of the corresponding ventilation facilities in the three-dimensional spatial topology. Virtual pre-simulation calculations are performed using the ventilation network solution equation to obtain the pre-simulated ventilation distribution state corresponding to each set of virtual adjustment parameters, including:

[0129] Based on the facility type and current regulation status parameters of each candidate regulation object, determine the physical dimensions and value boundaries of the regulation parameters;

[0130] Based on the value boundary and the current adjustment state parameter, multiple virtual adjustment parameter values ​​are generated within the value boundary, and the virtual adjustment parameter values ​​are arranged in order to obtain a virtual adjustment parameter sequence.

[0131] Each virtual adjustment parameter value in the virtual adjustment parameter sequence is combined with the identifier of the candidate adjustment object to obtain multiple sets of virtual adjustment parameters;

[0132] In the three-dimensional spatial topology, the location of the ventilation facility corresponding to each candidate adjustment object is determined, and the boundary condition values ​​of the corresponding ventilation facility locations in the three-dimensional spatial topology are updated based on each set of virtual adjustment parameters.

[0133] For each set of virtual control parameters updated in three-dimensional space topology, the theoretical wind pressure values ​​of each network node and the theoretical air volume values ​​of each network branch corresponding to that set of virtual control parameters are obtained by solving the ventilation network equations.

[0134] The theoretical wind pressure values ​​of network nodes and the theoretical air volume values ​​of network branches corresponding to each set of virtual adjustment parameters are combined to determine the pre-simulated ventilation distribution state corresponding to each set of virtual adjustment parameters.

[0135] When determining multiple sets of virtual adjustment parameters for each candidate control object, it is necessary to read the facility type attribute of the candidate control object. Facility types include ventilation equipment such as dampers, regulating windows, local ventilators, and main ventilators. For damper-type control objects, the physical dimension of its adjustment parameter is the opening percentage, with a value range between 0 and 100, where 0 represents a fully closed state and 100 represents a fully open state. When the current adjustment state parameter of a damper is read as 60% opening, this value is used as the baseline value. Combined with the mechanical structure constraints of the damper, the minimum adjustable opening of the damper is determined to be 10%, and the maximum opening is 100%, thus obtaining the value boundary of the damper's adjustment parameter as 10% to 100%. For local ventilator-type control objects, the physical dimension of its adjustment parameter is rotational speed, measured in revolutions per minute. When the rated speed of a local ventilation fan is 1450 rpm and the current operating speed is 1200 rpm, the speed range of the local ventilation fan is determined to be between 800 rpm and 1450 rpm, based on the speed regulation range of the frequency converter of the equipment.

[0136] When generating virtual control parameter values ​​based on the value boundaries and current control state parameters, an equal-interval sampling method is used to generate multiple discrete parameter values ​​within the value boundaries. For a damper with an opening of 60% and value boundaries ranging from 10% to 100%, 15 virtual control parameter values ​​are generated. The span of the value boundaries is calculated to be 90%, and dividing this span by 14 yields a sampling interval of approximately 6.43%. Starting from the minimum boundary value of 10%, the sampling interval is increased sequentially, resulting in virtual control parameter values ​​of 10%, 16.43%, 22.86%, 29.29%, 35.71%, 42.14%, 48.57%, 55%, 61.43%, 67.86%, 74.29%, 80.71%, 87.14%, 93.57%, and 100%. These 15 parameter values ​​are arranged in ascending order to form a virtual control parameter sequence. For a local ventilation fan with a rotation speed of 1200 rpm and a value range of 800 to 1450 rpm, 15 virtual adjustment parameter values ​​are generated, with a value range of 650 rpm and a sampling interval of approximately 46.43 rpm. The generated virtual adjustment parameter sequence is 800, 846.43, 892.86, 939.29, 985.71, 1032.14, 1078.57, 1125, 1171.43, 1217.86, 1264.29, 1310.71, 1357.14, 1403.57, and 1450 rpm.

[0137] When combining each virtual control parameter value in the virtual control parameter sequence with the identifier of the candidate control object, the unique identifier of the candidate control object is read. Assuming the identifier of a damper is FD-A-205, this damper corresponds to 15 virtual control parameter values. The identifier FD-A-205 is paired with parameter values ​​of 10%, 16.43%, 22.86%, etc., generating 15 sets of virtual control parameters, represented as FD-A-205 with 10%, FD-A-205 with 16.43%, FD-A-205 with 22.86%, etc. For a local ventilation fan with the identifier FJ-B-108, this identifier is paired with its 15 speed parameter values, generating 15 sets of virtual control parameters, such as FJ-B-108 with 800 rpm, FJ-B-108 with 846.43 rpm, etc.

[0138] When determining the location of ventilation facilities corresponding to each candidate adjustment object in the three-dimensional spatial topology, the node set and branch set of the three-dimensional spatial topology are traversed. For the damper with the identifier FD-A-205, the branch record containing this identifier is searched in the branch set, locating branch B267 between node N158 and node N159 of this damper. The current boundary condition value of branch B267 is read, which includes the wind resistance value and ventilation area. When the damper opening is 60%, the corresponding effective ventilation area is 2.4 square meters, and the wind resistance value is 0.15 kg per cubic meter. When the virtual adjustment parameter value of 10% is applied, according to the correspondence between damper opening and effective ventilation area, the effective ventilation area is calculated to become 0.4 square meters. At the same time, according to the inverse relationship between ventilation area and wind resistance, the wind resistance value is calculated to increase to 5.4 kg per cubic meter. The boundary condition values ​​of branch B267 are updated to a ventilation area of ​​0.4 square meters and a wind resistance value of 5.4 kg per cubic meter. When the virtual adjustment parameter value is set to 100%, the effective ventilation area is updated to 4 square meters, and the wind resistance value is updated to 0.09 kg per cubic meter.

[0139] For each set of updated virtual adjustment parameters in the three-dimensional spatial topology, iterative calculations are performed using the ventilation network solution equations. Taking the three-dimensional spatial topology after applying the virtual adjustment parameter FD-A-205 with a 10% adjustment as an example, all updated branch boundary conditions and node boundary conditions are read. Assuming the ventilation network contains 200 nodes and 280 branches, a set of node pressure balance equations and a set of branch airflow continuity equations are established. The node pressure balance equations describe the algebraic sum of the pressure drops in each branch at each node being equal to zero, while the branch airflow continuity equations describe the relationship that the total airflow into the node equals the total airflow out of the node. Initial iteration parameters are set, with the initial pressure values ​​of all nodes set to standard atmospheric pressure (101325 Pascals), and the initial airflow values ​​of all branches set to the estimated average airflow. The first iteration is performed, calculating the pressure drop of each branch based on its resistance value and assumed airflow value, and then adjusting the pressure values ​​of each node according to the node pressure balance equations. After the calculation is completed, it is checked whether the residual values ​​of all node pressure balance equations are less than the set threshold of 0.1 Pascals. When the residual values ​​of some nodes exceed the threshold after the first iteration, the updated node wind pressure values ​​are used to recalculate the airflow values ​​of each branch, and the second iteration begins. After multiple iterations, when the residual values ​​of the wind pressure balance equations for all nodes are less than 0.1 Pascals, and the residual values ​​of the airflow continuity equations for all branches are less than 0.01 cubic meters per second, the iteration is considered to have converged, and the theoretical wind pressure values ​​for each network node and the theoretical airflow values ​​for each network branch corresponding to the set of virtual adjustment parameters are output. For node N158, its theoretical wind pressure value is 101280 Pascals, and for branch B267, its theoretical airflow value is 8.5 cubic meters per second.

[0140] When combining the theoretical wind pressure values ​​of network nodes and the theoretical air volume values ​​of network branches corresponding to each set of virtual adjustment parameters, ventilation network calculations are performed for each of the 15 sets of virtual adjustment parameters, resulting in 15 sets of calculation results. Each set of calculation results includes wind pressure values ​​for 200 nodes and air volume values ​​for 280 branches. The first set of pre-simulated ventilation distribution states is formed by combining all node wind pressure values ​​and all branch air volume values ​​corresponding to the virtual adjustment parameter FD-A-205 at 10%, which describes the wind pressure and air volume distribution of the entire ventilation network when the damper FD-A-205 is adjusted to 10%. The second set of pre-simulated ventilation distribution states is formed by combining the calculation results corresponding to the virtual adjustment parameter FD-A-205 at 16.43%. In this way, 15 sets of pre-simulated ventilation distribution states are generated, each state corresponding to a specific virtual adjustment parameter value, providing complete pre-simulated data support for subsequent adjustment scheme evaluation.

[0141] In one optional implementation, the corresponding physical ventilation facility is controlled to perform adjustment actions based on the optimized control command, and real-time ventilation status data after adjustment is collected as feedback data. The feedback data is then used to update the three-dimensional spatial topology to maintain consistency between simulation calculations and actual ventilation conditions, including:

[0142] Based on the identifiers of each candidate control object in the optimized control command, the communication interface for querying the corresponding physical ventilation facility is determined;

[0143] The system sends control commands to the corresponding physical ventilation facilities through the communication interface, drives the physical ventilation facilities to perform adjustment actions, receives adjustment completion status signals from the physical ventilation facilities, and records the actual adjustment parameter values ​​of each physical ventilation facility.

[0144] After the physical ventilation system completes its adjustment, real-time ventilation status data at each monitoring point is collected via a real-time monitoring sensor network as the feedback data.

[0145] The actual adjustment response characteristics are obtained by comparing the feedback data with the actual adjustment parameter values.

[0146] Based on the actual adjustment response characteristics and the actual adjustment parameter values, the wind resistance characteristic parameters of the network nodes or branches where the ventilation facilities are located in the three-dimensional spatial topology are corrected, and the gas concentration measurement values ​​in the feedback data are updated to the gas concentration attribute values ​​of the corresponding roadway nodes in the three-dimensional spatial topology to maintain the consistency between the simulation calculation and the actual ventilation conditions.

[0147] After generating the optimized control command, a query operation needs to be performed in the equipment management database based on the identification information of each candidate control object in the command. This database stores information such as the equipment number, equipment type, communication protocol type, and communication interface address of all physical ventilation facilities underground. Taking the No. 3 ventilation door in a certain mine as an example, its equipment identification is FM-003. The database retrieves information showing that this ventilation door uses the Modbus TCP communication protocol, with a communication interface address of 192.168.10.35 and port number 502. For ventilation equipment, such as the local ventilation fan numbered FJ-012, its communication interface uses the PROFINET protocol, with an interface address of 192.168.10.68 and port number 34962. A communication connection table is established to map and associate the identification of each controllable device with the corresponding communication interface parameters.

[0148] After establishing a communication connection, specific control commands are sent to each physical ventilation device through the acquired communication interface. For damper devices, the control command includes an opening angle parameter. For example, setting the opening angle of damper FM-003 to 45 degrees is encapsulated into a data frame conforming to the Modbus TCP protocol specification. The data frame contains the device address, function code, register address, and a hexadecimal representation of the angle value. For the local ventilation fan FJ-012, the control command includes a speed adjustment parameter, adjusting its speed from the current 1200 rpm to 1450 rpm. This parameter is encoded according to the PROFINET protocol format and sent over the network. Upon receiving the control command, the physical ventilation device activates its actuator to perform the adjustment. The electric actuator of the damper drives the damper blades to rotate to the target angle, and the frequency converter of the fan adjusts the motor power supply frequency to change the speed. After the device completes the adjustment, its controller generates an adjustment completion status signal and sends it back to the monitoring system through the communication interface. This status signal includes the device identifier, the adjustment completion timestamp, and the actual parameter value achieved. Upon receiving the feedback signal from the FM-003 damper, its actual opening angle was recorded as 44.7 degrees, and the completion time was 14:32:18 on a certain day. Similarly, the actual rotational speed of the FJ-012 fan was recorded as 1448 revolutions per minute, and the completion time was 14:32:22 on a certain day. These actual adjustment parameter values ​​were stored in the adjustment history data table.

[0149] After the physical ventilation system is adjusted, the real-time monitoring sensor network deployed in various underground roadways begins collecting new ventilation status data. This sensor network consists of wind speed sensors, wind pressure sensors, and gas concentration sensors, each collecting data according to a set sampling period. Taking a main transport roadway as an example, the wind speed sensor CS-205, deployed in this roadway, began collecting data 3 minutes after adjustment, measuring a wind speed of 3.2 meters per second at that point. The wind pressure sensor CP-205 measured a wind pressure of 680 Pascals at that point. The methane concentration sensor CG-205 measured a methane concentration of 0.35%. In the return air roadway, the wind speed sensor CS-312 measured a wind speed of 2.8 meters per second, and the carbon monoxide concentration sensor CG-312 measured a carbon monoxide concentration of 18 ppm. Each sensor transmits the collected data to the data acquisition server at the ground monitoring center via the underground industrial Ethernet ring network. The server performs timestamp verification and data validity verification on the received data, removes abnormal data that is outside the physical range, and stores the valid data as a feedback data set.

[0150] The ventilation status parameters extracted from the feedback data set were compared with the baseline status data before adjustment. Taking the roadway where the FM-003 damper is located as an example, before adjustment, the wind speed at this location was 2.5 m / s and the wind pressure was 520 Pascals. After adjustment, the wind speed changed to 3.2 m / s and the wind pressure changed to 680 Pascals. The calculated change in wind speed was 0.7 m / s and the change in wind pressure was 160 Pascals. Based on the actual adjustment parameter values, the opening angle of the FM-003 damper was adjusted from 30 degrees to 44.7 degrees, a change of 14.7 degrees. By analyzing the relationship between the change in angle and the changes in wind speed and wind pressure, the actual adjustment response characteristics of the damper were obtained. These response characteristics indicate that for every 1 degree increase in the damper opening angle, the wind speed at this location increases by approximately 0.048 m / s and the wind pressure increases by approximately 10.9 Pascals. For the FJ-012 fan, when its speed increases from 1200 rpm to 1448 rpm, the change in speed is 248 rpm. The corresponding downstream wind speed increases from 2.1 m / s to 2.8 m / s, a change in wind speed of 0.7 m / s. This shows that for every 100 rpm increase in the fan speed, the downstream wind speed increases by approximately 0.28 m / s.

[0151] Based on the calculated actual regulation response characteristics, the drag characteristic parameters of corresponding nodes or branches in the three-dimensional spatial topology model are corrected. In the topology model, the network branch number corresponding to the FM-003 damper is Branch-128, and the original drag coefficient parameter of this branch is 0.085 kg / m³. According to the actual regulation response characteristics, when the damper opening angle is 44.7 degrees, the equivalent drag coefficient of this branch should be corrected to 0.062 kg / m³. Locate the parameter record of Branch-128 in the model database and update the drag coefficient field value to 0.062. For the network node number corresponding to the FJ-012 fan, Node-076, the coefficients of the wind pressure-air volume relationship curve in the original fan characteristic parameters of this node need to be adjusted according to the actual operating characteristics. Update the fan characteristic curve parameters of this node to ensure that the fan performance used in the simulation calculation is consistent with the actual operating performance.

[0152] Simultaneously, the gas concentration measurements from the feedback data are updated to the attribute data of the corresponding roadway nodes in the three-dimensional spatial topology. For example, the methane concentration sensor numbered CG-205 corresponds to Node-205 in the topology model, and its methane concentration attribute value is updated from the original 0.42 percentage to the newly acquired 0.35 percentage. The carbon monoxide concentration sensor numbered CG-312 corresponds to Node-312, and its carbon monoxide concentration attribute value is updated to 18 ppm. Through this parameter correction and data update operation, the network structure parameters, equipment characteristic parameters, and environmental state parameters in the three-dimensional spatial topology model are synchronized with the actual ventilation conditions underground. This ensures that the model used in subsequent simulation calculations can accurately reflect the current operating status of the ventilation system, providing an accurate calculation basis for the next round of ventilation optimization and control.

[0153] The intelligent ventilation optimization and control system for coal mines based on digital twins, according to embodiments of the present invention, includes:

[0154] The first unit is used to acquire real-time ventilation status data and mine structure data in underground coal mines;

[0155] The second unit is used to determine the three-dimensional spatial topology of the coal mine ventilation system based on the mine structure data, and to map the real-time ventilation status data to the corresponding roadway nodes and branch paths according to spatial coordinates to obtain twin ventilation status;

[0156] The third unit is used to perform simulation calculations based on the three-dimensional spatial topological relationship to obtain the theoretical ventilation distribution state. The twin ventilation state is compared with the theoretical ventilation distribution state to determine the air volume deviation area and the gas concentration abnormal area, thereby obtaining the ventilation deviation characteristics.

[0157] The fourth unit is used to virtually simulate the adjustment state of the ventilation facilities in the three-dimensional spatial topology based on the ventilation deviation characteristics, determine the adjustment scheme that meets the ventilation demand constraints and energy consumption constraints, and generate optimized control instructions based on the adjustment scheme;

[0158] The fifth unit is used to control the corresponding physical ventilation facilities to perform adjustment actions based on the optimized control instructions, and to collect real-time ventilation status data after adjustment as feedback data. The feedback data is used to update the three-dimensional spatial topology to maintain the consistency between simulation calculations and actual ventilation conditions.

[0159] A third aspect of the present invention provides an electronic device, comprising:

[0160] processor;

[0161] Memory used to store processor-executable instructions;

[0162] The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0163] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0164] This invention can be a method, apparatus, system, and / or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for performing various aspects of the invention.

[0165] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for intelligent ventilation optimization and control in coal mines based on digital twins, characterized in that, include: Acquire real-time ventilation status data and mine structure data in underground coal mines; Based on the mine structure data, the three-dimensional spatial topology of the coal mine ventilation system is determined, and the real-time ventilation status data is mapped to the corresponding roadway nodes and branch paths according to spatial coordinates to obtain twin ventilation status. Simulation calculations are performed based on the aforementioned three-dimensional spatial topological relationships to obtain the theoretical ventilation distribution state. The twin ventilation state is then compared with the theoretical ventilation distribution state to identify areas of airflow deviation and abnormal gas concentration, thereby obtaining ventilation deviation characteristics. Based on the ventilation deviation characteristics, the adjustment state of the ventilation facilities is virtually simulated in the three-dimensional spatial topology to determine the adjustment scheme that meets the ventilation demand constraints and energy consumption constraints, and an optimized control command is generated based on the adjustment scheme. Based on the optimized control command, the corresponding physical ventilation facility is controlled to perform adjustment actions, and real-time ventilation status data after adjustment is collected as feedback data. The feedback data is used to update the three-dimensional spatial topology to maintain the consistency between simulation calculation and actual ventilation conditions.

2. The method according to claim 1, characterized in that, Based on the mine structure data, the three-dimensional spatial topology of the coal mine ventilation system is determined, and the real-time ventilation status data is mapped to the corresponding roadway nodes and branch paths according to spatial coordinates to obtain twin ventilation status, including: Based on the roadway topology data in the mine structure data, the roadway connectivity is determined, and based on the ventilation facility location data in the mine structure data, the node locations and branch connections of the airflow path are determined, thus obtaining the three-dimensional spatial topology. Extract the spatial coordinate information of each measuring point in the real-time ventilation status data, and calculate the spatial distance between the measuring point data and each roadway node in the three-dimensional spatial topology for the spatial coordinate information of each measuring point data. Based on the spatial distance, a roadway node matching each measurement point data is selected, and the roadway node identifier of the matching roadway node is used as the roadway node identifier to which the corresponding measurement point data belongs; Based on the roadway node identifier, the air volume distribution data, air pressure distribution data and gas concentration distribution data of the corresponding measuring point are associated with the branch paths between the roadway node and the adjacent nodes, respectively. Based on the airflow continuity constraint, the real-time ventilation status data on the branch paths between adjacent nodes are interpolated to obtain the twin ventilation status corresponding to all nodes and branch paths in the three-dimensional spatial topology.

3. The method according to claim 1, characterized in that, Simulation calculations are performed based on the aforementioned three-dimensional spatial topological relationships to obtain the theoretical ventilation distribution state. The twin ventilation state is then compared with the theoretical ventilation distribution state to identify areas of airflow deviation and abnormal gas concentration, thus obtaining ventilation deviation characteristics, including: The roadway nodes in the three-dimensional spatial topology are taken as network nodes, and the branch paths are taken as network branches. The ventilation network is solved using the ventilation network solution equation to obtain the theoretical wind pressure value of each network node and the theoretical air volume value of each network branch, which are taken as the theoretical ventilation distribution state. Based on the measured air volume values ​​of each roadway node in the twin ventilation state and the theoretical air volume values ​​of the corresponding network nodes in the theoretical ventilation distribution state, the air volume deviation area is determined. The measured gas concentration values ​​of each roadway node in the twin ventilation state are compared with the ventilation safety concentration limit node by node to identify roadway nodes whose measured gas concentration values ​​exceed the ventilation safety concentration limit. Based on the spatially continuous distribution of the roadway nodes exceeding the limit and the nodes within their coverage area, the gas concentration abnormal area is determined. The ventilation deviation characteristics are obtained by associating and combining the airflow deviation area, the gas concentration abnormal area, and the spatial range of the area.

4. The method according to claim 3, characterized in that, Based on the measured air volume values ​​of each roadway node in the twin ventilation state and the theoretical air volume values ​​of the corresponding network nodes in the theoretical ventilation distribution state, the air volume deviation area is determined, including: Based on the mapping relationship between the roadway node identifier in the twin ventilation state and the network node number in the theoretical ventilation distribution state, the measured air volume value of the roadway node with the same spatial location and the theoretical air volume value of the corresponding network node are determined. The difference between the measured air volume value of each roadway node at the same spatial location and the theoretical air volume value of the corresponding network node is calculated to obtain the air volume deviation value of each roadway node; The airflow deviation value is processed to obtain a relative deviation rate. The relative deviation rate is compared with a preset airflow deviation judgment criterion, and roadway nodes whose relative deviation rate exceeds the airflow deviation judgment criterion are identified as deviation nodes. Based on the spatial connectivity between the deviation nodes in the three-dimensional spatial topology, deviation nodes with direct branch path connections are clustered into the same deviation region group; Extract the node identifier, airflow deviation value, relative deviation rate, and branch path identifier connecting the deviation nodes from all deviation nodes in each deviation region group to obtain the airflow deviation region.

5. The method according to claim 1, characterized in that, Based on the ventilation deviation characteristics, a virtual simulation of the ventilation facility's adjustment state is performed within the three-dimensional spatial topology to determine an adjustment scheme that satisfies ventilation demand constraints and energy consumption constraints. Based on this adjustment scheme, optimized control commands are generated, including: Based on the ventilation deviation characteristics, the target area to be adjusted is determined in the three-dimensional spatial topology. Based on the connectivity path of the target area in the three-dimensional spatial topology, the candidate adjustment objects are determined as ventilation facilities that affect the ventilation state of the target area. For each candidate adjustment object, multiple sets of virtual adjustment parameters are determined. These virtual adjustment parameters are applied to the boundary conditions of the corresponding ventilation facilities in the three-dimensional spatial topology. Virtual pre-simulation calculations are performed using the ventilation network solution equation to obtain the pre-simulation ventilation distribution state corresponding to each set of virtual adjustment parameters. Extract the pre-simulated air volume and pre-simulated gas concentration values ​​of the target area from each pre-simulated ventilation distribution state, and screen out candidate adjustment parameter groups that meet ventilation demand constraints and ventilation safety concentration constraints based on the pre-simulated air volume and pre-simulated gas concentration values; Calculate the total energy consumption of the ventilation system corresponding to each candidate set of adjustment parameters, and select a candidate set of adjustment parameters as the optimal set of adjustment parameters based on the total energy consumption of the ventilation system to obtain the adjustment scheme; The optimal adjustment parameter set of each candidate adjustment object in the adjustment scheme is converted into the corresponding action command of the ventilation facility, and the optimized control command is generated.

6. The method according to claim 5, characterized in that, For each candidate adjustment object, multiple sets of virtual adjustment parameters are determined. These virtual adjustment parameters are applied to the boundary conditions of the corresponding ventilation facilities in the three-dimensional spatial topology. Virtual pre-simulation calculations are performed using the ventilation network solution equation to obtain the pre-simulated ventilation distribution state corresponding to each set of virtual adjustment parameters, including: Based on the facility type and current regulation status parameters of each candidate regulation object, determine the physical dimensions and value boundaries of the regulation parameters; Based on the value boundary and the current adjustment state parameter, multiple virtual adjustment parameter values ​​are generated within the value boundary, and the virtual adjustment parameter values ​​are arranged in order to obtain a virtual adjustment parameter sequence. Each virtual adjustment parameter value in the virtual adjustment parameter sequence is combined with the identifier of the candidate adjustment object to obtain multiple sets of virtual adjustment parameters; In the three-dimensional spatial topology, the location of the ventilation facility corresponding to each candidate adjustment object is determined, and the boundary condition values ​​of the corresponding ventilation facility locations in the three-dimensional spatial topology are updated based on each set of virtual adjustment parameters. For each set of virtual control parameters updated in three-dimensional space topology, the theoretical wind pressure values ​​of each network node and the theoretical air volume values ​​of each network branch corresponding to that set of virtual control parameters are obtained by solving the ventilation network equations. The theoretical wind pressure values ​​of network nodes and the theoretical air volume values ​​of network branches corresponding to each set of virtual adjustment parameters are combined to determine the pre-simulated ventilation distribution state corresponding to each set of virtual adjustment parameters.

7. The method according to claim 1, characterized in that, Based on the optimized control command, the corresponding physical ventilation facility is controlled to perform adjustment actions, and real-time ventilation status data after adjustment is collected as feedback data. The feedback data is used to update the three-dimensional spatial topology to maintain consistency between simulation calculations and actual ventilation conditions, including: Based on the identifiers of each candidate control object in the optimized control command, the communication interface for querying the corresponding physical ventilation facility is determined; The system sends control commands to the corresponding physical ventilation facilities through the communication interface, drives the physical ventilation facilities to perform adjustment actions, receives adjustment completion status signals from the physical ventilation facilities, and records the actual adjustment parameter values ​​of each physical ventilation facility. After the physical ventilation system completes its adjustment, real-time ventilation status data at each monitoring point is collected via a real-time monitoring sensor network as the feedback data. The actual adjustment response characteristics are obtained by comparing the feedback data with the actual adjustment parameter values. Based on the actual adjustment response characteristics and the actual adjustment parameter values, the wind resistance characteristic parameters of the network nodes or branches where the ventilation facilities are located in the three-dimensional spatial topology are corrected, and the gas concentration measurement values ​​in the feedback data are updated to the gas concentration attribute values ​​of the corresponding roadway nodes in the three-dimensional spatial topology to maintain the consistency between the simulation calculation and the actual ventilation conditions.

8. A coal mine intelligent ventilation optimization and control system based on digital twins, used to implement the method as described in any one of claims 1-7, characterized in that, include: The first unit is used to acquire real-time ventilation status data and mine structure data in underground coal mines; The second unit is used to determine the three-dimensional spatial topology of the coal mine ventilation system based on the mine structure data, and to map the real-time ventilation status data to the corresponding roadway nodes and branch paths according to spatial coordinates to obtain twin ventilation status; The third unit is used to perform simulation calculations based on the three-dimensional spatial topological relationship to obtain the theoretical ventilation distribution state. The twin ventilation state is compared with the theoretical ventilation distribution state to determine the air volume deviation area and the gas concentration abnormal area, thereby obtaining the ventilation deviation characteristics. The fourth unit is used to virtually simulate the adjustment state of the ventilation facilities in the three-dimensional spatial topology based on the ventilation deviation characteristics, determine the adjustment scheme that meets the ventilation demand constraints and energy consumption constraints, and generate optimized control instructions based on the adjustment scheme; The fifth unit is used to control the corresponding physical ventilation facilities to perform adjustment actions based on the optimized control instructions, and to collect real-time ventilation status data after adjustment as feedback data. The feedback data is used to update the three-dimensional spatial topology to maintain the consistency between simulation calculations and actual ventilation conditions.

9. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 7.

Images