Dynamic regulation and control method and system for mine ventilation

Abstract

The invention relates to a dynamic regulation and control method and system for mine ventilation. The method comprises the following steps: acquiring multi-source data in a mine, and preprocessing the multi-source data to obtain a standardized data set; calculating the risk of each region in the mine based on the standardized data set to obtain a risk prediction matrix; based on the standardized data set and the risk prediction matrix, the air volume demand of each area is calculated, and an air volume demand table is obtained; based on the air volume demand table, an optimization proposition corresponding to the air volume demand is constructed and solved, and a solution set of the optimization proposition is obtained; mapping the solution set based on a preset rule to obtain a feasible allocation scheme; and based on the feasible allocation scheme, generating an equipment cooperation instruction, and obtaining a security instruction set. According to the method, dynamic factors can be considered, the air volume is dynamically adjusted, multi-fan cooperation is achieved, and the effects of real-time sensing and autonomous dynamic decision making of mine ventilation are achieved.

Description

Technical Field

[0001] This invention relates to the field of mine ventilation technology, and in particular to a method and system for dynamic control of mine ventilation. Background Technology

[0002] Mine ventilation systems are core infrastructure for ensuring safe underground operations. Their core function is to precisely deliver fresh surface air to working areas such as mining faces and chambers, while simultaneously expelling toxic and harmful gases such as methane and dust, as well as residual heat, creating a safe and controllable environment for personnel and equipment operation. With the continuous increase in mining depth and the expansion of mining areas, the complexity and dynamism of underground working conditions have significantly increased, making traditional mine ventilation control models inadequate for the safety and energy efficiency requirements of modern mines.

[0003] Traditional ventilation control essentially relies on a passive management logic of "static design + manual experience": engineers draw static ventilation network design diagrams based on initial mine exploration data, and pre-set a global fixed air volume distribution scheme based on this. Daily control is achieved solely through manual adjustment of key parameters such as main fan frequency and damper opening. Data acquisition depends on regular manual inspection records, mainly collecting basic indicators such as gas concentration and wind speed. In emergencies such as a sudden gas outburst after blasting or sudden equipment failure, emergency operations must be carried out by notifying on-site personnel via telephone from the dispatch center. Control cycles are often measured in hours or even days.

[0004] This model reveals multiple fatal flaws under complex dynamic operating conditions:

[0005] First, there is a lack of dynamic adaptability. The continuous advancement of the underground mining face leads to real-time changes in the roadway topology. Blasting operations cause a large amount of gas to surge out in a short period of time. The start-up and shutdown of equipment such as tunneling machines and ventilation fans cause local airflow turbulence. These dynamic factors make the preset fixed air volume scheme quickly fail. The lag in manual response can easily lead to safety hazards such as gas accumulation and excessive dust in high-risk areas.

[0006] Secondly, the precision of the control is insufficient. The globally uniform fixed air volume distribution model lacks consideration for regional differences: high-risk areas face safety threats due to insufficient air volume supply, while low-risk areas suffer serious energy waste due to excessive air volume. At the same time, deep tunnels often have blind spots in air supply because the wind resistance increases with depth.

[0007] Third, the risk prediction and coordination capabilities are weak. Traditional methods can only monitor current environmental parameters and cannot predict potential risks based on multi-dimensional data such as geological conditions and equipment status. Moreover, the control of equipment such as main fans, local fans, and dampers is independent of each other and lacks a coordinated linkage mechanism, which can easily lead to system instability and further reduce ventilation efficiency.

[0008] More importantly, the traditional model fails to integrate key data such as geological stress distribution, equipment load rate, and real-time personnel location: potential risks such as decreased rock stability caused by changes in geological stress and abnormal fluctuations in borehole gas pressure cannot be quantitatively assessed; changes in air volume demand caused by the superposition of waste heat generated by the high-load operation of the tunneling machine and high ambient temperature are ignored; and oxygen supply in densely populated areas lacks accurate data support. These shortcomings collectively result in traditional ventilation control systems lacking real-time perception, risk prediction, and autonomous dynamic decision-making capabilities, making it difficult to meet the core needs of safe and efficient mining in modern mines. Summary of the Invention

[0009] In view of this, the purpose of the present invention is to provide a method and system for dynamic control of mine ventilation, which solves the problems of traditional mine ventilation relying on fixed schemes and manual scheduling, which cannot cope with the risk response lag, energy waste and insufficient equipment coordination caused by dynamic changes underground.

[0010] The technical solution adopted by this invention to solve its technical problem is:

[0011] A method and system for dynamic control of mine ventilation are provided, including the following steps:

[0012] 101. Acquire multi-source data within the mine and preprocess the multi-source data to obtain a standardized dataset; the multi-source data includes at least one of sensor data, equipment status data, and personnel positioning data;

[0013] 102. Based on the standardized dataset, calculate the environmental feature vector, production disturbance characteristics, and geological risk parameters of each area in the mine to obtain the risk prediction matrix;

[0014] 103. Based on the standardized dataset and the risk prediction matrix, calculate the air volume demand for each region to obtain an air volume demand table;

[0015] 104. Based on the air volume demand table, construct and solve the optimization proposition corresponding to the air volume demand to obtain the solution set of the optimization proposition; map the solution set based on preset rules to obtain a feasible allocation scheme;

[0016] 105. Based on the feasible allocation scheme, generate equipment coordination instructions to obtain a safety instruction set; the safety instruction set is used to instruct the wind power equipment in the mine to perform operations.

[0017] Preferably, step 102 includes:

[0018] Based on the standardized dataset, the gas concentration slope and the dust concentration change rate are calculated; and the gas concentration slope and the dust concentration change rate are integrated into an environmental feature vector.

[0019] Based on the standardized dataset, features of the operating status of production equipment are extracted to obtain production disturbance features;

[0020] Based on the standardized dataset, geological conditions are quantified to obtain geological risk parameters;

[0021] The environmental feature vector, the production disturbance feature, and the geological risk parameter are input into a pre-trained risk prediction engine to obtain the predicted gas emission rate and dust risk value; the predicted gas emission rate and the dust risk value are integrated to obtain the risk prediction matrix.

[0022] Preferably, the standardized dataset further includes geological stress distribution; step 102, "quantifying geological conditions and obtaining geological risk parameters," includes: calculating the geological stress coefficient based on the geological stress distribution using the following formula:

[0023]

[0024] Where, k stress σ is the geological stress coefficient. max This represents the maximum geological stress.

[0025] Based on the borehole gas pressure values ​​and pressure monitoring times in the standardized dataset, the geological pressure change rate is calculated using the following formula:

[0026]

[0027] Where, r p ΔP is the geological pressure change rate, ΔP is the borehole gas pressure gradient, and Δt is the pressure monitoring time interval; by integrating the geological stress coefficient and the geological pressure change rate, the geological risk parameter is obtained.

[0028] Preferably, the standardized dataset further includes the tunneling machine load rate and ambient temperature; step 103, "calculating the air volume demand for each area to obtain an air volume demand table," includes:

[0029] Based on the standardized dataset, the gas dilution air volume, dust control air volume, and personnel oxygen supply air volume are calculated; and the maximum value among the gas dilution air volume, dust control air volume, and personnel oxygen supply air volume is taken to determine the basic air volume requirement; based on the risk prediction matrix, the air volume redundancy coefficient is calculated.

[0030] Based on the tunneling machine load rate and the ambient temperature, the heat compensation air volume is calculated using the following formula:

[0031]

[0032] ΔQ thermal =ΔQ L +ΔQT

[0033] Where, ΔQ thermal For heat compensation air volume, ΔQ L For load thermal compensation, ΔQ T For high temperature compensation, L is the tunneling machine load rate, Q base The base air volume is T, where T is the ambient temperature.

[0034] Based on the basic air volume requirement, the air volume redundancy coefficient, and the heat compensation air volume, the air volume requirement is calculated using the following formula, and the air volume requirements of each region are integrated to obtain the air volume requirement table:

[0035] Q req =Q base ×k redund +ΔQ thermal

[0036] Among them, Q req For air volume requirements, Q base Basic air volume, k redund ΔQ is the air volume redundancy factor. thermal Air volume for heat compensation.

[0037] Preferably, "calculating the air volume redundancy coefficient based on the risk prediction matrix" includes: extracting the predicted gas outburst from the risk prediction matrix and generating the gas risk redundancy coefficient using the following formula:

[0038]

[0039] Where, k gas δ is the gas risk redundancy coefficient. q q represents the outflow deviation rate. pred For the predicted gas outflow, q CH4 This represents the actual gas emission rate.

[0040] The dust risk value is mapped to the corresponding dust risk redundancy coefficient using a preset mapping rule.

[0041] The maximum value of the gas risk redundancy coefficient and the dust risk redundancy coefficient in the region is determined as the basic redundancy coefficient.

[0042] Based on the mining task in the area, the values ​​of the basic redundancy coefficient that are not within the preset threshold are adjusted to be within the preset threshold to obtain the adjusted redundancy coefficient.

[0043] The adjusted redundancy coefficient and the basic redundancy coefficient within the preset threshold are integrated to obtain the air volume redundancy coefficient.

[0044] Preferably, step 104, "constructing an optimization proposition corresponding to the air volume requirement," includes:

[0045] Based on the tunnel topology database, a dynamic ventilation network diagram is constructed with tunnel intersections as nodes and tunnel branches as edges; the edges of the dynamic ventilation network diagram are attached with wind resistance and initial air volume attributes.

[0046] Based on the dynamic ventilation network diagram, the objective function is constructed using the following formula:

[0047]

[0048] Where k is the ventilation zone number, K is the total number of ventilation zones, and w k As the risk level weight for the partition, Q k Q represents the actual air volume of zone k. req,k Let λ be the required air volume for zone k, λ be the energy consumption penalty coefficient, E be the set of ventilation network branches, (i,j) be the branch from node i to node j, and R be the value of R. ij Let Q be the wind resistance from node i to node j. ij Let β be the airflow from node i to node j, β be the node balance penalty coefficient, V be the set of nodes in the ventilation network, and Q be the airflow from node i to node j. in Q represents the total airflow into node v. out The total air volume flowing out of node v;

[0049] Based on the air volume demand table and the dynamic ventilation network diagram, safety hard constraints and demand soft constraints are constructed to obtain a constraint set; the safety hard constraints include the upper limit of wind speed and the limit of fan power, and the demand soft constraints include the zone air volume deviation range; the objective function and the constraint set are integrated to obtain the optimization proposition.

[0050] Preferably, step 105, "generating device coordination instructions to obtain a security instruction set," includes:

[0051] Query the preset fan characteristic curve library, solve for the fan efficiency frequency that satisfies the feasible allocation scheme, and obtain the main fan frequency command;

[0052] Based on the feasible allocation scheme, the local fan speed of each region is calculated to obtain a speed command set; the speed is calculated based on the region's air volume demand, fan characteristic constant, and impeller diameter.

[0053] The target air volume between different nodes is extracted from the feasible allocation scheme, and the damper opening angle is calculated based on the target air volume to obtain the damper opening instruction set; the opening angle is calculated based on the damper drag coefficient and the pressure difference between nodes.

[0054] The conflicting instructions in the main fan frequency command, the speed command set, and the damper opening command set are removed, and the remaining instructions are integrated into the safety command set; the conflicting instructions are combinations of equipment parameters that cause system instability.

[0055] A mine ventilation dynamic control system, characterized in that: the system is used to execute a mine ventilation dynamic control method as described in any one of the above claims, comprising:

[0056] The preprocessing module is used to perform step 101, which involves acquiring multi-source data from the mine and preprocessing it to obtain a standardized dataset.

[0057] The prediction module is used to perform step 102, which calculates environmental feature vectors, production disturbance features, and geological risk parameters based on a standardized dataset to obtain a risk prediction matrix.

[0058] The demand module is used to execute step 103, which calculates the air volume demand for each region based on the standardized dataset and risk prediction matrix, and obtains the air volume demand table.

[0059] The solution module is used to execute step 104, which constructs and solves optimization problems based on the air volume demand table, and maps the solution set to obtain feasible allocation schemes.

[0060] The instruction module is used to execute step 105, which generates device coordination instructions based on feasible allocation schemes to obtain a security instruction set.

[0061] A computer device includes a memory and a processor, the memory storing a computer program, characterized in that the processor executes the computer program to implement the steps of the dynamic control method for mine ventilation described in any of the preceding claims.

[0062] A computer-readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the steps of the dynamic control method for mine ventilation described in any of the preceding claims.

[0063] The beneficial effects of this invention are:

[0064] This invention provides a dynamic control method and system for mine ventilation. By integrating multi-source information such as sensor data, equipment status data, and personnel positioning data, and performing standardized preprocessing, it breaks through the limitations of traditional single environmental data. It constructs a three-dimensional risk prediction matrix by combining environmental feature vectors, production disturbance characteristics, and geological risk parameters, achieving advanced prediction and regional quantification of gas and dust risks. This avoids the lag and subjectivity of manual experience-based judgment, reducing safety hazards such as gas accumulation and excessive dust in high-risk areas from the source. Regarding energy utilization, it abandons the traditional global fixed air volume distribution mode and achieves dynamic air distribution through a precise air volume model. This avoids insufficient air volume in high-risk areas and reduces energy waste in low-risk areas. Combined with the energy consumption penalty constraint in multi-objective optimization problems, it significantly improves ventilation efficiency. The system improves the energy efficiency of the ventilation system and constructs a multi-objective optimization model based on a dynamic ventilation network diagram, which includes air volume deviation, energy consumption, and node balance. This model simultaneously generates coordinated commands for main fan frequency, local fan speed, and damper opening. Conflict checks eliminate contradictory operations, resolving the system disorder issues that are easily caused by traditional independent multi-device adjustments. This enables real-time dynamic control at the minute level, adapting to dynamic underground conditions such as face advancement and equipment start-up and shutdown. Regarding intelligent decision-making, a closed-loop design encompassing "data preprocessing - risk prediction - air volume calculation - optimized allocation - command generation" transforms ventilation control from "passive manual response" to "data and model-driven autonomous decision-making." This allows for precise adaptation to complex underground dynamics without frequent manual intervention, significantly improving the scientific and efficient nature of mine ventilation management. Attached Figure Description

[0065] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0066] Figure 1 A schematic flowchart of a dynamic control method for mine ventilation provided as an embodiment of the present invention;

[0067] Figure 2 This is a schematic diagram of a dynamic control system for mine ventilation provided in one embodiment of the present invention. Detailed Implementation

[0068] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0069] In one embodiment, such as Figure 1 As shown, a dynamic control method for mine ventilation is provided. This embodiment illustrates the method applied to a terminal. It should be noted that this method can also be applied to a server, or to a system including both a terminal and a server, and is implemented through interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0070] Step 101: Obtain multi-source data from within the mine and preprocess the multi-source data to obtain a standardized dataset; the multi-source data includes at least one of sensor data, equipment status data, and personnel positioning data.

[0071] Multi-source data comprises a collection of heterogeneous data sources collected in real-time within the mine, including sensor data (time-series monitoring values ​​of environmental parameters such as gas concentration, dust concentration, temperature, and humidity); equipment status data (operating parameters such as fan speed, damper opening, and tunneling machine load rate); and personnel location data (real-time location coordinates and distribution density of underground workers). The standardized dataset is a structured data set obtained by cleaning, aligning, and normalizing the multi-source data. Terminals collect multi-source data in real-time via IoT devices, removing outliers, filling in missing values, unifying data from different frequencies to the same timestamp to achieve time synchronization of multi-source data, and performing minimum-maximum scaling on numerical data to convert it to...

[0072] The range is [0,1]. One-hot encoding is performed on categorical data to obtain standardized data.

[0073] Step 102: Based on the standardized dataset, calculate the risk of each area in the mine to obtain the risk prediction matrix.

[0074] Specifically, the risk prediction matrix is ​​a two-dimensional matrix, where rows represent region IDs (unique codes) and columns represent risk types, used to quantify the dynamic risk value of each region. Optionally, the terminal calculates environmental feature vectors, extracts production disturbance features, quantifies geological risks, inputs the features into an LSTM neural network, and outputs predicted gas outbursts and dust risk values. The predicted gas outburst is an estimate of the gas release in the future period; the dust risk value is the probability that dust diffusion will lead to reduced visibility or an explosion. The risk values ​​of all regions are integrated and organized into a matrix with a row and column structure.

[0075] Step 103: Based on the standardized dataset and risk prediction matrix, calculate the air volume demand for each region to obtain the air volume demand table.

[0076] Specifically, the air volume demand table is a list representing the air volume demand of each area. For example, the terminal calculates the gas dilution air volume, dust control air volume, and personnel oxygen supply air volume separately, takes the maximum value of the three as the base air volume, adds dynamic compensation, calculates the total demand of a single area, and integrates the results of all areas to generate the air volume demand table.

[0077] Step 104: Based on the air volume demand table, construct and solve the optimization proposition corresponding to the air volume demand to obtain the solution set of the optimization proposition; map the solution set according to the preset rules to obtain a feasible allocation scheme.

[0078] Specifically, the optimization proposition is a mathematical model whose objective is to minimize airflow deviation, energy consumption, and node imbalance. The solution set is the set of all feasible solutions to the optimization proposition, which is the mathematical solution space. A feasible allocation scheme is the optimal solution that satisfies engineering constraints. For example, a dynamic ventilation network diagram is constructed at the terminal, and an objective function is defined, which includes an airflow deviation term: the weighted absolute difference between the actual airflow and the required airflow; an energy consumption term: the product of branch resistance and the square of the airflow; and a node balance term: the square of the difference between inflow and outflow airflow. Constraints are set, the solution set is solved, and the optimal solution is selected from the solution set according to preset rules (such as prioritizing the lowest energy consumption), transforming the mathematical solution into an engineering-executable airflow allocation scheme.

[0079] Step 105: Based on the feasible allocation scheme, generate equipment coordination instructions to obtain a safety instruction set; the safety instruction set is used to instruct the wind power equipment in the mine to perform operations.

[0080] The equipment coordination instructions are a set of parameters for controlling ventilation equipment, including fan frequency, fan speed, and damper opening. The safety instruction set is the final instruction package after conflict detection. The terminal calculates the main fan frequency, local fan speed, and damper opening, converts them into instructions, detects conflicting instructions, deletes conflicting instructions, and iterates again to solve the problem. The verified instructions are then categorized and packaged according to equipment type to obtain the safety instruction set.

[0081] This embodiment provides a method for dynamic control of mine ventilation. It acquires multi-source data within the mine and preprocesses this data to obtain a standardized dataset. The multi-source data includes at least one of sensor data, equipment status data, and personnel positioning data. Based on the standardized dataset, the risk of each area in the mine is calculated to obtain a risk prediction matrix. Based on the standardized dataset and the risk prediction matrix, the air volume demand of each area is calculated to obtain an air volume demand table. Based on the air volume demand table, optimization problems corresponding to the air volume demand are constructed and solved to obtain a solution set. The solution set is mapped based on preset rules to obtain a feasible allocation scheme. Based on the feasible allocation scheme, equipment coordination instructions are generated to obtain a safety instruction set. The safety instruction set is used to instruct the ventilation equipment in the mine to perform operations. Through the above technical means, the method achieves dynamic adjustment of air volume through multi-fan coordination, considering the effects of real-time perception and autonomous dynamic decision-making in mine ventilation.

[0082] In one embodiment, based on a standardized dataset, the risk of each area in the mine is calculated to obtain a risk prediction matrix, including:

[0083] Based on a standardized dataset, the slope of gas concentration and the rate of change of dust concentration are calculated; and the slope of gas concentration and the rate of change of dust concentration are integrated into an environmental feature vector.

[0084] Among them, the gas concentration slope is the rate of change of gas concentration per unit time, reflecting the gas accumulation trend; a positive value indicates an increase in concentration, and a negative value indicates a decrease. The dust concentration change rate is the rate of change of dust concentration per unit time, reflecting the dynamics of dust diffusion. The environmental feature vector is a two-dimensional vector integrating the gas concentration slope and the dust concentration change rate, characterizing the dynamic changes in environmental risk. The terminal uses the least squares method to fit a linear trend to the time-series gas concentration data in the standardized dataset; the slope of this fit is the gas concentration slope. For the dust concentration data, the average difference between adjacent time points is calculated. These two scalars are combined to form a two-dimensional vector, resulting in the environmental feature vector.

[0085] Based on a standardized dataset, features of the operating status of production equipment are extracted to obtain production disturbance features.

[0086] Specifically, production disturbance characteristics are a set of indicators reflecting abnormal equipment operating states. For example, production disturbance characteristics may include the variance of tunneling machine load fluctuations, the harmonic distortion rate of fan current, and the frequency of equipment start-up and shutdown events. Optionally, the terminal calculates statistical characteristics on the equipment status data: mean, variance, and peak value; performs a Fourier transform on the current signal to extract the harmonic distortion rate; counts the number of equipment start-ups and shutdowns per unit time; if the frequency is greater than 3 times per hour, it is marked as a disturbance event; and integrates the statistical characteristics, harmonic distortion rate, and disturbance events to obtain the production disturbance characteristics.

[0087] Based on standardized datasets, geological conditions are quantified to obtain geological risk parameters.

[0088] Specifically, geological risk parameters are comprehensive indicators that integrate geological stress and gas pressure. The terminal obtains the maximum geological stress in the standardized dataset, calculates the geological stress coefficient, obtains the borehole gas pressure gradient and monitoring time interval, calculates the geological pressure change rate, integrates the parameters, and obtains the geological risk parameters.

[0089] By inputting environmental feature vectors, production disturbance characteristics, and geological risk parameters into a pre-trained risk prediction engine, the predicted gas outburst volume and dust risk value are obtained.

[0090] Among them, the predicted gas emission rate is the forecast value of gas release in the future period. The dust risk value is the probability of dust-related disasters. The terminal inputs environmental feature vectors, production disturbance characteristics, and geological risk parameters into the model. The model processes these parameters through a multi-layer fully connected network and outputs the predicted gas emission rate and the dust risk value.

[0091] By integrating the predicted gas emission volume and dust risk value, a risk prediction matrix is ​​obtained.

[0092] In this risk prediction matrix, rows represent regions, columns represent risk types, and elements are risk values. For each region, the terminal normalizes the predicted gas outflow into a gas risk value, while the dust risk value is directly taken from the model output. A matrix is ​​created, updated every 5 minutes, and historical matrix data is continuously stored to obtain the risk prediction matrix.

[0093] This embodiment achieves early warning through dynamic perception and real-time monitoring, reduces false alarm rate, avoids major risk omissions, supports accurate calculation of subsequent air volume demand modules, and improves the pertinence and rationality of ventilation control.

[0094] In one embodiment, the standardized dataset also includes geological stress distribution;

[0095] Based on a standardized dataset, geological conditions are quantified to obtain geological risk parameters, including:

[0096] Based on the geological stress distribution, the geological stress coefficient is calculated using the following formula:

[0097]

[0098] Where, k stress σ is the geological stress coefficient. max This represents the maximum geological stress.

[0099] Specifically, the geological stress coefficient is a normalized index (range 0-1) that quantifies rock strata stability; a higher value indicates a higher risk of rockburst. The maximum geological stress is the maximum principal stress value of the surrounding rock in the tunnel extracted from the standardized dataset, derived from monitoring data from a stress sensor array. The terminal reads geological stress distribution data from the standardized dataset, locates the maximum principal stress value in the current area, and calculates the geological stress coefficient using a formula. The baseline value is 12 MPa (Megapascal), corresponding to the critical stress point of the rock strata; the coefficient is 0.4, controlling the steepness of the curve to ensure a significant gradient change within the 10 MPa to 14 MPa range.

[0100] Based on the borehole gas pressure values ​​and pressure monitoring times in the standardized dataset, the geological pressure change rate is calculated using the following formula:

[0101]

[0102] Where, r p ΔP represents the rate of change of geological pressure, ΔP represents the borehole gas pressure gradient (in MPa), and Δt represents the pressure monitoring time interval (in h / min).

[0103] Specifically, the geological pressure change rate is the instantaneous rate of change in borehole gas pressure, reflecting the risk of sudden gas outbursts. The borehole gas pressure gradient is the difference in borehole gas pressure between adjacent monitoring periods. The pressure monitoring time interval is the gas pressure sampling interval defined in the standardized dataset. The terminal obtains the time-series borehole gas pressure sequence from the standardized dataset, reads the preset time interval, calculates the difference between adjacent data points, takes the maximum gradient value, solves for the change rate, and obtains the geological pressure change rate.

[0104] By integrating the geological stress coefficient and the geological pressure change rate, geological risk parameters are obtained.

[0105] The geological risk parameter is a two-dimensional vector that integrates stress and gas pressure, used as input to the risk prediction engine. The terminal directly splices the geological stress coefficient and the rate of change of geological pressure to obtain the geological risk parameter.

[0106] This embodiment transforms raw geological data into machine-understandable quantitative indicators through multi-parameter collaborative early warning, providing key inputs for the risk prediction engine and solving the problem of delayed response to geological mutations.

[0107] In one embodiment, the standardized dataset also includes tunneling machine load rate and ambient temperature;

[0108] Based on a standardized dataset and a risk prediction matrix, the air volume demand for each region is calculated, resulting in an air volume demand table, including:

[0109] Based on a standardized dataset, the gas dilution air volume, dust control air volume, and personnel oxygen supply air volume are calculated; and the maximum value among the gas dilution air volume, dust control air volume, and personnel oxygen supply air volume is taken to determine the basic air volume requirement.

[0110] Among them, the gas dilution air volume is the minimum air volume required to reduce the gas concentration to a safe threshold. The dust control air volume is the air volume required to suppress dust diffusion and is positively correlated with dust particle size and density. The personnel oxygen supply air volume is the minimum air volume required to ensure personnel breathing. The basic air volume requirement is the regional benchmark air volume, which is the maximum value among the gas dilution air volume, dust control air volume, and personnel oxygen supply air volume. The terminal extracts real-time parameters such as gas concentration, dust concentration, and personnel distribution density from a standardized dataset. Based on the real-time gas emission, it dynamically calculates the gas dilution air volume, calculates the dust control air volume by referring to a preset table based on the dust type (coal dust / rock dust), and calculates the personnel oxygen supply air volume according to the number of personnel in the area and the per capita demand. The maximum value is taken as the basic air volume requirement.

[0111] Calculate the air volume redundancy coefficient based on the risk prediction matrix.

[0112] Specifically, the air volume redundancy coefficient is a risk-driven amplification factor with a value of no less than 1, used to cope with fluctuations in predicted risks. The terminal extracts the predicted gas emission volume and dust risk value corresponding to the area, generates a gas risk redundancy coefficient, maps the dust risk value to the dust risk redundancy coefficient, takes the maximum value of the two as the basic redundancy coefficient, and calibrates the basic redundancy coefficient based on the mining task to obtain the air volume redundancy coefficient.

[0113] Based on the tunneling machine load rate and ambient temperature, the heat compensation air volume is calculated using the following formula:

[0114]

[0115] ΔQ thermal =ΔQ L +ΔQ T

[0116] Where, ΔQ thermal For heat compensation air volume, ΔQ L For load thermal compensation, ΔQ T For high temperature compensation, L is the tunneling machine load rate, Q base The base air volume is T, where T is the ambient temperature.

[0117] Specifically, thermal compensation airflow is the additional airflow required due to equipment heating or high-temperature environments. Load thermal compensation is the compensation airflow required for heat dissipation when the tunneling machine is operating under high load. High-temperature compensation is the additional airflow required when the ambient temperature exceeds the limit. For example, the terminal calculates the tunneling machine load compensation based on a piecewise function. No compensation is given when the load rate is ≤70%, linearly increasing when the load rate is between 70% and 90%, and fixed compensation is performed to prevent overload when the load rate is >90%. Temperature triggers high-temperature compensation, and the total thermal compensation is calculated.

[0118] Based on the basic air volume requirement, air volume redundancy factor, and heat compensation air volume, the air volume requirement is calculated using the following formula, and the air volume requirements of each area are integrated to obtain the air volume requirement table:

[0119] Q req =Q base ×k redund +ΔQ thermal

[0120] Among them, Q req For air volume requirements, Q base Basic air volume, k redund ΔQ is the air volume redundancy factor. thermal Air volume for heat compensation.

[0121] Specifically, the air volume demand table is a regional-level list of air volume demands. Optionally, the terminal calculates the demand using formulas, integrates the regional data, and updates the air volume demand table every 5 minutes, providing structured input for optimization tasks.

[0122] This embodiment uses a dynamic compensation mechanism and a thermal compensation formula to address the load / temperature effects ignored by traditional methods. It also achieves risk-adaptive airflow amplification through a redundancy coefficient, resulting in a more accurate and targeted mine ventilation control method.

[0123] In one embodiment, the airflow redundancy coefficient is calculated based on the risk prediction matrix, including:

[0124] Extract the predicted gas outflow from the risk prediction matrix and generate the gas risk redundancy coefficient using the following formula:

[0125]

[0126] Where, k gas δ is the gas risk redundancy coefficient. q q represents the outflow deviation rate. pred For the predicted gas outflow, q CH4 This represents the actual amount of gas emitted.

[0127] Specifically, the gas emission deviation rate is the relative deviation between the predicted gas emission and the actual gas emission. The gas risk redundancy coefficient is the output value of a piecewise function based on the deviation rate, used to amplify the base air volume to cope with gas risks. The terminal extracts the regional predicted gas emission from the risk prediction matrix, obtains the real-time gas emission from the standardized dataset, calculates the deviation rate, and maps it to the corresponding gas risk redundancy coefficient through a piecewise function.

[0128] By using preset mapping rules, dust risk values ​​are mapped to corresponding dust risk redundancy coefficients.

[0129] The dust risk redundancy coefficient is the result of converting dust risk values ​​to amplification factors, mapped using preset rules. The terminal extracts regional dust risk values ​​from the risk prediction matrix and maps them to the corresponding dust risk redundancy coefficient based on the preset mapping rules. For example, the mapping rules are designed based on the following: when the dust risk value is in a low-risk state (≤0.3), no compensation is given; when the dust risk value is in a medium-risk state (between 0.3 and 0.8), linear compensation is applied; and when the dust risk value is in a high-risk state (>0.8), exponential compensation is applied.

[0130] The maximum value of the gas risk redundancy coefficient and the dust risk redundancy coefficient in the region is taken as the basic redundancy coefficient.

[0131] The basic redundancy coefficient is the minimum airflow amplification factor required for the current area, obtained by taking the maximum value of the gas and dust coefficients. For the same area, the terminal takes the maximum value of the gas risk redundancy coefficient and the dust risk redundancy coefficient as the basic redundancy coefficient to ensure that the most dangerous factor receives sufficient airflow redundancy, while avoiding energy waste caused by double compensation.

[0132] Based on the mining task in a region, the values ​​of the basic redundancy coefficient that are not within the preset threshold are adjusted to the preset threshold to obtain the adjusted redundancy coefficient.

[0133] Specifically, the adjusted redundancy coefficient is the final redundancy value after being constrained by the mining task, and is forcibly limited to a preset threshold. For example, when the task is tunneling, the preset threshold is...

[0134] [1.2, 2.0] represents a high gas risk area; when the task is longwall mining, the preset threshold is...

[0135] [1.1, 1.6] represents a relatively stable region; when the task is roadway maintenance, the preset threshold is...

[0136] [1.0, 1.3] represents a low-risk area. If the basic redundancy coefficient is less than the lower threshold, it is raised to the lower threshold; if the basic redundancy coefficient is greater than the upper threshold, it is lowered to the upper threshold to prevent over-ventilation.

[0137] The adjusted redundancy coefficient and the basic redundancy coefficient within the preset threshold are integrated to obtain the air volume redundancy coefficient.

[0138] Specifically, the airflow redundancy coefficient is a set of calibrated coefficients for all regions, serving as the final input for airflow demand calculation. The terminal iterates through all regions, integrates the coefficients, and correlates them with the airflow demand table in real time, updating the airflow redundancy coefficient accordingly.

[0139] This embodiment reduces energy waste from traditional fixed redundancy coefficients through risk-driven dynamic redundancy, achieves accurate matching of actual threats through independent assessment of dual risks of gas and dust, and adaptively avoids unknown risks in mining scenarios.

[0140] In one embodiment, based on the airflow demand table, an optimization proposition corresponding to the airflow demand is constructed, including:

[0141] Based on the tunnel topology database, a dynamic ventilation network diagram is constructed with tunnel intersections as nodes and tunnel branches as edges.

[0142] The tunnel topology database stores structured data on the spatial relationships of mine tunnels, including attributes such as node coordinates, branch lengths, and cross-sectional dimensions. The dynamic ventilation network diagram is a directed graph, where the node set represents tunnel intersections, including 3-way and 4-way junctions; the edge set represents tunnel branches, with attributes including air resistance and initial airflow. The terminal extracts the tunnel intersection coordinates from the topology database, assigns a unique node ID to each intersection, defines the tunnels connecting adjacent nodes as directed edges with directions set according to the main airflow direction, updates edge attributes in real time, dynamically calculates air resistance based on support deformation, and takes the initial airflow from the nearest sensor reading, transforming the physical tunnels into a mathematically processable network model.

[0143] Based on the dynamic ventilation network diagram, the objective function is constructed using the following formula:

[0144]

[0145] Where k is the ventilation zone number, K is the total number of ventilation zones, and w k As the risk level weight for the partition, Q k Q represents the actual air volume of zone k. req,k Let λ be the required air volume for zone k, λ be the energy consumption penalty coefficient, E be the set of ventilation network branches, (i,j) be the branch from node i to node j, and R be the value of R. ij Let Q be the wind resistance from node i to node j. ij Let β be the airflow from node i to node j, β be the node balance penalty coefficient, V be the set of nodes in the ventilation network, and Q be the airflow from node i to node j. in Q represents the total airflow into node v. out This represents the total airflow from node v.

[0146] Specifically, the objective function is a weighted sum of three objectives that needs to be minimized. These include an airflow deviation term, prioritizing accurate air supply to high-risk areas; an energy consumption term, suppressing high-airflow distribution in high-resistance roadways; and a node balance term, forcibly satisfying the fluid continuity equation. The terminal substitutes the parameters into the formula to obtain the objective function.

[0147] Based on the air volume demand table and dynamic ventilation network diagram, safety hard constraints and demand soft constraints are constructed to obtain the constraint set.

[0148] Specifically, hard safety constraints are inviolable physical limits; violating them will result in no optimization solution. Soft demand constraints are operational requirements that allow for small deviations and can be implemented through penalties via the objective function. For example, hard safety constraints may include wind speed constraints and fan power constraints; soft demand constraints may include zone airflow constraints and critical node wind pressure constraints. Hard constraints ensure the safe operation of the system, while soft constraints are handled flexibly through penalties via the objective function.

[0149] By integrating the objective function and the set of constraints, an optimization proposition is obtained.

[0150] The optimization problem is a mathematical description that includes an objective function and constraints. For example, the terminal obtains a dynamic network diagram, an air volume demand table, an objective function, and a set of constraints to obtain the optimal air volume allocation scheme, and then solves the optimization problem using solvers such as linear programming, quadratic programming, and heuristic algorithms.

[0151] This embodiment transforms mine ventilation control into a computable mathematical problem by constructing an optimization proposition, thereby reducing airflow deviation in high-risk areas, decreasing total power consumption of fans to reduce energy consumption, and accelerating response to emergencies.

[0152] In one embodiment, based on a feasible allocation scheme, device coordination instructions are generated to obtain a security instruction set, including:

[0153] Query the preset fan characteristic curve library, solve for the fan efficiency frequency that satisfies the feasible allocation scheme, and obtain the main fan frequency command.

[0154] The fan characteristic curve library is a database storing performance parameters of different fan models, including the mapping relationship between air volume, frequency, and efficiency. The main fan frequency command is the command value for controlling the speed of the main ventilation fan, used to adjust the total air intake of the mine. The terminal extracts the total required air volume of the entire mine from the feasible allocation plan, queries the characteristic curve of the current main fan model in the curve library, and selects the frequency point with the highest fan efficiency under the premise of meeting the total required air volume, generating the main fan frequency command.

[0155] Based on feasible allocation schemes, the local fan speeds of each region are calculated using the calculation formula. (where k is a characteristic constant and D is the impeller diameter), thus obtaining the speed command set.

[0156] Specifically, the speed command set is a set of commands to control the speed of local ventilation equipment. The terminal reads the required air volume values ​​for each area from the feasible allocation scheme, calculates the local fan speed based on the fan characteristic constant, fan impeller diameter, and base speed, and obtains the speed command set.

[0157] Extract the target air volume between different nodes from the feasible allocation scheme, and calculate the damper opening angle based on the target air volume using the calculation formula. (where k) P (where is the pressure differential coefficient) is used to obtain the damper opening command set.

[0158] Specifically, the damper opening instruction set is a set of instructions for controlling the opening angle of dampers between roadways. The terminal obtains the branch air volume between nodes from the feasible allocation scheme, calculates the damper opening angle based on the damper drag coefficient and the pressure difference between nodes, and packages it into the damper opening instruction set.

[0159] Conflicting commands in the main fan frequency command, speed command set, and damper opening command set are removed, and the remaining commands are integrated into a safety command set.

[0160] Conflicting instructions are those whose combinations of device control parameters cause system instability. The safety instruction set is the final executable instruction package after conflict detection. The terminal, based on a conflict detection rule base, applies preset physical rules to check each rule, eliminates conflicting rules, and integrates them into a safety instruction set to avoid system failures caused by device conflicts.

[0161] This embodiment reduces local ventilation energy consumption by calculating regional rotation speed on demand and using a multi-device conflict rule library, while improving the stability and accuracy of mine ventilation control.

[0162] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0163] Based on the same inventive concept, this application also provides a mine ventilation dynamic control system for implementing the aforementioned mine ventilation dynamic control method. The solution provided by this system is similar to the solution described in the above method; therefore, the specific limitations in one or more mine ventilation dynamic control system embodiments provided below can be found in the limitations of the mine ventilation dynamic control method described above, and will not be repeated here.

[0164] In one exemplary embodiment, such as Figure 2 As shown, a dynamic control system 800 for mine ventilation is provided, comprising:

[0165] The preprocessing module 801 is used to acquire multi-source data in the mine and preprocess the multi-source data to obtain a standardized dataset; the multi-source data includes at least one of sensor data, equipment status data, and personnel positioning data;

[0166] Prediction module 802 is used to calculate the risk of each area in the mine based on a standardized dataset and obtain a risk prediction matrix.

[0167] Demand module 803 is used to calculate the air volume demand of each region based on a standardized dataset and a risk prediction matrix, and to obtain an air volume demand table.

[0168] The solution module 804 is used to construct and solve optimization problems corresponding to the air volume demand based on the air volume demand table, and obtain the solution set of the optimization problems; the solution set is then mapped to obtain feasible allocation schemes based on preset rules.

[0169] The instruction module 805 is used to generate equipment coordination instructions based on feasible allocation schemes to obtain a safety instruction set; the safety instruction set is used to instruct the wind power equipment in the mine to perform operations.

[0170] Furthermore, the prediction module 802 is also used to: calculate the gas concentration slope and dust concentration change rate based on the standardized dataset; integrate the gas concentration slope and dust concentration change rate into an environmental feature vector; extract the features of the operating status of production equipment based on the standardized dataset to obtain production disturbance features; quantify the geological conditions based on the standardized dataset to obtain geological risk parameters; input the environmental feature vector, production disturbance features and geological risk parameters into a pre-trained risk prediction engine to obtain the predicted gas emission and dust risk value; and integrate the predicted gas emission and dust risk value to obtain a risk prediction matrix.

[0171] Furthermore, the standardized dataset also includes geological stress distribution;

[0172] The prediction module 802 is also used to: calculate the geological stress coefficient based on the geological stress distribution using the following formula:

[0173]

[0174] Where, k stress σ is the geological stress coefficient. max This represents the maximum geological stress.

[0175] Based on the borehole gas pressure values ​​and pressure monitoring times in the standardized dataset, the geological pressure change rate is calculated using the following formula:

[0176]

[0177] Where, r p ΔP is the rate of change of geological pressure, ΔP is the gradient of borehole gas pressure, and Δt is the time interval of pressure monitoring.

[0178] By integrating the geological stress coefficient and the geological pressure change rate, geological risk parameters are obtained.

[0179] Furthermore, the standardized dataset also includes tunneling machine load rate and ambient temperature;

[0180] Module 803 is also used for: calculating the gas dilution air volume, dust control air volume, and personnel oxygen supply air volume based on a standardized dataset; determining the basic air volume requirement by taking the maximum value among these three air volumes; calculating the air volume redundancy coefficient based on the risk prediction matrix; and calculating the heat compensation air volume using the following formula based on the tunneling machine load rate and ambient temperature:

[0181]

[0182] ΔQ thermal =ΔQ L +ΔQ T

[0183] Where, ΔQ thermal For heat compensation air volume, ΔQ L For load thermal compensation, ΔQ T For high temperature compensation, L is the tunneling machine load rate, Q base The base air volume is T, where T is the ambient temperature.

[0184] Based on the basic air volume requirement, air volume redundancy factor, and heat compensation air volume, the air volume requirement is calculated using the following formula, and the air volume requirements of each area are integrated to obtain the air volume requirement table:

[0185] Q req =Q base ×k redund +ΔQ thermal

[0186] Among them, Q req For air volume requirements, Q baseBasic air volume, k redund ΔQ is the air volume redundancy factor. thermal Air volume for heat compensation.

[0187] Furthermore, the demand module 803 is also used for:

[0188] Extract the predicted gas outflow from the risk prediction matrix and generate the gas risk redundancy coefficient using the following formula:

[0189]

[0190] Where, k gas δ is the gas risk redundancy coefficient. q q represents the outflow deviation rate. pred For the predicted gas outflow, q CH4 This represents the actual gas emission rate.

[0191] By using preset mapping rules, dust risk values ​​are mapped to corresponding dust risk redundancy coefficients;

[0192] The maximum value of the gas risk redundancy coefficient and the dust risk redundancy coefficient in the region is determined as the basic redundancy coefficient. Based on the mining task of the region, the values ​​of the basic redundancy coefficient that are not within the preset threshold are adjusted to the preset threshold to obtain the adjusted redundancy coefficient. The adjusted redundancy coefficient and the basic redundancy coefficient that are within the preset threshold are integrated to obtain the air volume redundancy coefficient.

[0193] Furthermore, the solver module 804 is also used to: construct a dynamic ventilation network graph based on the tunnel topology database, using tunnel intersections as nodes and tunnel branches as edges; and construct the objective function based on the dynamic ventilation network graph using the following formula:

[0194]

[0195] Where k is the ventilation zone number, K is the total number of ventilation zones, and w k As the risk level weight for the partition, Q k Q represents the actual air volume of zone k. req,k Let λ be the required air volume for zone k, λ be the energy consumption penalty coefficient, E be the set of ventilation network branches, (i,j) be the branch from node i to node j, and R be the value of R. ij Let Q be the wind resistance from node i to node j. ij Let β be the airflow from node i to node j, β be the node balance penalty coefficient, V be the set of nodes in the ventilation network, and Q be the airflow from node i to node j. in Q represents the total airflow into node v. out The total air volume flowing out of node v;

[0196] Based on the air volume demand table and dynamic ventilation network diagram, safety hard constraints and demand soft constraints are constructed to obtain a constraint set; by integrating the objective function and the constraint set, the optimization proposition is obtained.

[0197] Furthermore, the instruction module 805 is also used to: query a preset fan characteristic curve library, solve for the fan efficiency frequency that satisfies the feasible allocation scheme, and obtain the main fan frequency instruction; calculate the local fan speed of each area based on the feasible allocation scheme, and obtain the speed instruction set; extract the target air volume between different nodes from the feasible allocation scheme, and calculate the damper opening angle based on the target air volume, and obtain the damper opening instruction set; remove conflicting instructions from the main fan frequency instruction, speed instruction set, and damper opening instruction set, and integrate the remaining instructions into a safety instruction set.

[0198] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the dynamic control method for mine ventilation as described above.

[0199] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0200] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0201] The above-described embodiments are merely illustrative of several implementation methods of the embodiments of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the embodiments of this application, and these modifications and improvements all fall within the protection scope of the embodiments of this application.

Claims

1. A method for dynamic control of mine ventilation, characterized in that, Includes the following steps:

101. Acquire multi-source data within the mine and preprocess the multi-source data to obtain a standardized dataset; the multi-source data includes at least one of sensor data, equipment status data, and personnel positioning data; 102. Based on the standardized dataset, calculate the environmental feature vector, production disturbance characteristics, and geological risk parameters of each area in the mine to obtain the risk prediction matrix; 103. Based on the standardized dataset and the risk prediction matrix, calculate the air volume demand for each region to obtain an air volume demand table; 104. Based on the air volume demand table, construct and solve the optimization proposition corresponding to the air volume demand to obtain the solution set of the optimization proposition; map the solution set based on preset rules to obtain a feasible allocation scheme; 105. Based on the feasible allocation scheme, generate equipment coordination instructions to obtain a safety instruction set; the safety instruction set is used to instruct the wind power equipment in the mine to perform operations.

2. The method for dynamic control of mine ventilation as described in claim 1, characterized in that: Step 102 includes: Based on the standardized dataset, the gas concentration slope and the dust concentration change rate are calculated; and the gas concentration slope and the dust concentration change rate are integrated into an environmental feature vector. Based on the standardized dataset, features of the operating status of production equipment are extracted to obtain production disturbance features; Based on the standardized dataset, geological conditions are quantified to obtain geological risk parameters; The environmental feature vector, the production disturbance feature, and the geological risk parameter are input into a pre-trained risk prediction engine to obtain the predicted gas emission rate and dust risk value; the predicted gas emission rate and the dust risk value are integrated to obtain the risk prediction matrix.

3. The method for dynamic control of mine ventilation as described in claim 2, characterized in that: The standardized dataset also includes geological stress distribution; step 102, "quantifying geological conditions and obtaining geological risk parameters," includes: calculating the geological stress coefficient based on the geological stress distribution using the following formula: Where, k stress σ is the geological stress coefficient. max This represents the maximum geological stress. Based on the borehole gas pressure values ​​and pressure monitoring times in the standardized dataset, the geological pressure change rate is calculated using the following formula: Where, r p ΔP is the geological pressure change rate, ΔP is the borehole gas pressure gradient, and Δt is the pressure monitoring time interval; by integrating the geological stress coefficient and the geological pressure change rate, the geological risk parameter is obtained.

4. The method for dynamic control of mine ventilation as described in claim 3, characterized in that: The standardized dataset also includes tunneling machine load rate and ambient temperature; step 103, "calculating the air volume demand for each area to obtain an air volume demand table," includes: Based on the standardized dataset, the gas dilution air volume, dust control air volume, and personnel oxygen supply air volume are calculated; and the maximum value among the gas dilution air volume, dust control air volume, and personnel oxygen supply air volume is taken to determine the basic air volume requirement; based on the risk prediction matrix, the air volume redundancy coefficient is calculated. Based on the tunneling machine load rate and the ambient temperature, the heat compensation air volume is calculated using the following formula: ΔQ thermal =ΔQ L +ΔQ T Where, ΔQ thermal For heat compensation air volume, ΔQ L For load thermal compensation, ΔQ T For high temperature compensation, L is the tunneling machine load rate, Q base The base air volume is T, where T is the ambient temperature. Based on the basic air volume requirement, the air volume redundancy coefficient, and the heat compensation air volume, the air volume requirement is calculated using the following formula, and the air volume requirements of each region are integrated to obtain the air volume requirement table: Q req =Q base ×k redund +ΔQ thermal Among them, Q req For air volume requirements, Q base Basic air volume, k redund ΔQ is the air volume redundancy factor. thermal Air volume for heat compensation.

5. The method for dynamic control of mine ventilation as described in claim 3, characterized in that: "Calculating the air volume redundancy coefficient based on the risk prediction matrix" includes: extracting the predicted gas outflow from the risk prediction matrix and generating the gas risk redundancy coefficient using the following formula: Where, k gas δ is the gas risk redundancy coefficient. q q represents the outflow deviation rate. pred For the predicted gas outflow, q CH4 This represents the actual gas emission rate. The dust risk value is mapped to the corresponding dust risk redundancy coefficient using a preset mapping rule. The maximum value of the gas risk redundancy coefficient and the dust risk redundancy coefficient in the region is determined as the basic redundancy coefficient. Based on the mining task in the area, the values ​​of the basic redundancy coefficient that are not within the preset threshold are adjusted to be within the preset threshold to obtain the adjusted redundancy coefficient. The adjusted redundancy coefficient and the basic redundancy coefficient within the preset threshold are integrated to obtain the air volume redundancy coefficient.

6. The method for dynamic control of mine ventilation as described in claim 1, characterized in that: Step 104, "Constructing an optimization proposition corresponding to the stated airflow requirement," includes: Based on the tunnel topology database, a dynamic ventilation network diagram is constructed with tunnel intersections as nodes and tunnel branches as edges; the edges of the dynamic ventilation network diagram are attached with wind resistance and initial air volume attributes. Based on the dynamic ventilation network diagram, the objective function is constructed using the following formula: Where k is the ventilation zone number, K is the total number of ventilation zones, and w k As the risk level weight for the partition, Q k Q represents the actual air volume of zone k. req,k Let λ be the required air volume for zone k, λ be the energy consumption penalty coefficient, E be the set of ventilation network branches, (i,j) be the branch from node i to node j, and R be the value of R. ij Let Q be the wind resistance from node i to node j. ij Let β be the airflow from node i to node j, β be the node balance penalty coefficient, V be the set of nodes in the ventilation network, and Q be the airflow from node i to node j. in Q represents the total airflow into node v. out The total air volume flowing out of node v; Based on the air volume demand table and the dynamic ventilation network diagram, safety hard constraints and demand soft constraints are constructed to obtain a constraint set; the safety hard constraints include the upper limit of wind speed and the limit of fan power, and the demand soft constraints include the zone air volume deviation range; the objective function and the constraint set are integrated to obtain the optimization proposition.

7. The method for dynamic control of mine ventilation as described in claim 6, characterized in that: Step 105, "Generate device coordination instructions to obtain a security instruction set," includes: Query the preset fan characteristic curve library, solve for the fan efficiency frequency that satisfies the feasible allocation scheme, and obtain the main fan frequency command; Based on the feasible allocation scheme, the local fan speed of each region is calculated to obtain a speed command set; the speed is calculated based on the region's air volume demand, fan characteristic constant, and impeller diameter. The target air volume between different nodes is extracted from the feasible allocation scheme, and the damper opening angle is calculated based on the target air volume to obtain the damper opening instruction set; the opening angle is calculated based on the damper drag coefficient and the pressure difference between nodes. The conflicting instructions in the main fan frequency command, the speed command set, and the damper opening command set are removed, and the remaining instructions are integrated into the safety command set; the conflicting instructions are combinations of equipment parameters that cause system instability.

8. A dynamic control system for mine ventilation, characterized in that: The system is used to execute a dynamic control method for mine ventilation as described in any one of claims 1-7, comprising: The preprocessing module is used to perform step 101, which involves acquiring multi-source data from the mine and preprocessing it to obtain a standardized dataset. The prediction module is used to perform step 102, which calculates environmental feature vectors, production disturbance features, and geological risk parameters based on a standardized dataset to obtain a risk prediction matrix. The demand module is used to execute step 103, which calculates the air volume demand for each region based on the standardized dataset and risk prediction matrix, and obtains the air volume demand table. The solution module is used to execute step 104, which constructs and solves optimization problems based on the air volume demand table, and maps the solution set to obtain feasible allocation schemes. The instruction module is used to execute step 105, which generates device coordination instructions based on feasible allocation schemes to obtain a security instruction set.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the dynamic control method for mine ventilation according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the dynamic control method for mine ventilation according to any one of claims 1 to 7.

Images