Mining intelligent ventilation system based on data driving and decision control method

Abstract

The invention discloses a mining intelligent ventilation system based on data driving and a decision control method, and belongs to the technical field of ventilation decision management. Mine data multi-interference monitoring; and mine ventilation demand monitoring. According to the method, whether mine blind area data complementation is adopted or not is determined through mine monitoring blind area data analysis, then whether electromagnetic radiation interference suppression and environmental interference correction are adopted or not is determined through mine data multi-interference identification, and finally whether mine ventilation adjustment is adopted or not is determined through mine ventilation demand analysis. The effect of improving the accuracy of the mine ventilation decision data is achieved, and the problem that the accuracy of the mine ventilation decision data is low in the prior art is solved.

Description

Technical Field

[0001] This invention relates to the field of ventilation decision management technology, and in particular to a data-driven intelligent ventilation system and decision control method for mines. Background Technology

[0002] Coal is an important energy resource in my country. With the expansion of mining scale and the extension of mines to deeper areas, disasters such as gas, fire, and heat hazards are becoming increasingly serious, and multiple disasters are superimposed. When mining resources are collected based on mining equipment (such as fully mechanized mining supports and tunneling machines), underground ventilation has become an urgent problem to be solved for safe production in mines. The trend of building smart mines in coal mines is becoming increasingly obvious, but mine ventilation is still in the manual or semi-manual stage, which is difficult to meet the needs of smart mine construction. With the increase of mining depth, the ventilation network becomes more and more complex, and the difficulty of underground ventilation continues to increase. The existing mine ventilation decision-making process mainly involves: First, by using sensors such as wind speed sensors, gas sensors, and temperature and humidity sensors placed at key locations in the mine roadways (such as the working face, return airway, and near the ventilation doors), the air volume and gas concentration of ventilation devices (such as industrial controllers) are collected in real time. Key mine monitoring parameters such as temperature, humidity, and air volume are collected in real time, along with basic mine data such as the location of the mine production face and equipment operating status. This data is then transmitted to the ground monitoring center for preprocessing to remove outliers and fill in missing data, ensuring the effectiveness of ventilation decision-making data. Next, a ventilation network solution model (such as the node method or loop method) is used to analyze the resistance distribution and airflow allocation of the ventilation system, forming the core basis for ventilation decisions. Finally, based on the ventilation analysis results, ventilation adjustment decisions are made based on the ventilation system (such as adjusting the fan speed using the main fan frequency converter control cabinet or adjusting the opening of the electric ventilation window using a PID controller), and the control results are fed back to the ground monitoring center.

[0003] For example, Chinese invention patent CN120046991B discloses a ventilation data analysis method and system for intelligent mine ventilation, which includes: a multi-source environmental parameter sensing module for acquiring mine environmental parameters, including methane concentration, real-time air volume, and temperature; a dynamic risk assessment and fusion module for calculating a dynamic risk index based on the mine environmental parameters; a risk coding and air pressure time-series feature generation module for generating a risk trend status code based on the dynamic risk index; and a multimodal ventilation decision module for acquiring the tunneling status of the tunneling machine; and obtaining a ventilation volume prediction interval for a specified future time range based on the time-series feature vector, risk trend status code, tunneling status of the tunneling machine, and real-time air volume.

[0004] In existing technologies, intelligent ventilation systems for mines have shown a development trend from data analysis to intelligent decision-making. By acquiring environmental parameters to calculate dynamic risk indices, and using adaptive time-series models (such as temporal convolutional networks) to process atmospheric pressure data, and coordinating and scheduling corresponding professional models to execute collaboratively, they achieve quantitative assessment and trend coding of environmental risks. However, although existing technologies have achieved a preliminary connection from data to decision-making, in practical applications, especially when dealing with complex dynamic working conditions in mines, the above-mentioned technical processes still have significant bottlenecks in terms of data reliability, model applicability, and system linkage control, which restricts the full realization of the overall effectiveness of the mine ventilation system.

[0005] The above-mentioned technology has at least the following technical problems: In the process of intelligent ventilation decision-making and control in mines, existing mine ventilation systems have low levels of automation and may lack a linkage mechanism with dynamic operating conditions such as mine production rhythm and face advancement. Furthermore, they often employ simple logic based on single fixed threshold judgments, failing to integrate multi-source mine data (such as gas concentration distribution) for comprehensive decision-making. This may result in insufficient flexibility in control strategies. Simultaneously, existing intelligent ventilation decision-making methods primarily rely on sensor point monitoring, resulting in limited coverage and potential blind spots in local mine monitoring. This leads to insufficient integrity of multi-source mine data. Moreover, sensors are more susceptible to data drift and distortion due to mine humidity and vibration. Existing technologies typically use static filtering (such as moving average filtering) and outlier removal to process data, lacking a multi-source data cross-validation and dynamic calibration mechanism. This makes it difficult to fundamentally solve the data distortion problem. When mining equipment is operating, it may generate strong electromagnetic radiation interference. This interference can intrude into the sensor signal transmission line through electromagnetic superposition, superimposing with the original monitoring signal to form distortion. Moreover, the interference intensity exhibits dynamic fluctuations with the power of the mining equipment, further exacerbating data drift and distortion problems. Consequently, when performing ventilation network calculations, existing technologies are mostly based on fixed roadway topology to construct ventilation network calculations, without considering the dynamic calculation of ventilation networks under nonlinear dynamic interferences such as gas diffusion and airflow disturbances. This leads to large errors in predicting the mine ventilation status under complex working conditions. In addition, the aging and leakage of traditional ventilation facilities and poor resetting further exacerbate the instability of the ventilation status. When adjusting mine ventilation based on the analysis results of unstable ventilation status, excessive or delayed air volume adjustment is likely to occur, ultimately resulting in low accuracy of mine ventilation decision data. Summary of the Invention

[0006] To address the problem of low mine ventilation efficiency in existing technologies, this invention provides a data-driven intelligent mine ventilation system and decision control method. The technical solution is as follows: On the one hand, a data-driven intelligent ventilation decision-making and control method for mines is provided. This method includes: S1, performing mine monitoring blind zone data analysis, and determining whether to supplement mine blind zone data based on the analysis results. If supplementation is performed, then determining whether to identify multiple interferences in the mine data after the blind zone data supplementation is completed; S2, if not, performing mine monitoring data validity verification, and determining whether to identify multiple interferences in the mine data based on the validity verification results. If identification is performed, then determining whether to identify multiple interferences in the mine data based on the identification results. Whether to take electromagnetic radiation interference suppression and environmental interference correction measures; S3, if not, then perform dynamic calculation of the ventilation network. After the dynamic calculation of the ventilation network is completed, perform mine ventilation demand analysis. Based on the results of the mine ventilation demand analysis, determine whether to take mine ventilation regulation measures, including adjustment of the total mine ventilation volume and adjustment of the mine branch roadway ventilation volume. If not, the current ventilation status is deemed qualified. If so, after the mine ventilation regulation is completed, determine whether the mine ventilation demand analysis is qualified. If not, send a ventilation regulation failure prompt. If qualified, the current ventilation status is deemed qualified.

[0007] On the other hand, a data-driven intelligent ventilation system for mines is provided, including: a mine signal sensing module, a mine information transmission module, a mine data platform monitoring module, and a mine ventilation linkage control module. The mine signal sensing module is used to collect mine data with preliminary quality self-diagnosis and screening decision-making capabilities from multiple types of sensors deployed at key nodes underground. The mine information transmission module collects raw mine physical signals and performs preliminary conditioning, analog-to-digital conversion, and local caching to form a mine monitoring data sequence with autonomous transmission path optimization decision-making capabilities. The mine data platform monitoring module encapsulates and executes the core data processing and analysis algorithms covered by the intelligent ventilation decision-making and control methods, executing decision-making methods including mine monitoring blind zone data analysis and decision-making, mine data multi-interference identification and decision-making, dynamic ventilation network calculation and decision-making, and mine ventilation demand analysis and decision-making, generating ventilation control commands with intelligent mine ventilation decision-making capabilities. The mine ventilation linkage control module receives control commands from the mine data platform monitoring module and generates equipment control signals with dynamic adaptive decision-making capabilities by driving the actuators.

[0008] Beneficial effects The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: 1. The data-driven intelligent ventilation decision-making and control method for mines provided by this invention analyzes mine monitoring blind zone data and determines whether to supplement the mine blind zone data based on the analysis results. If supplementation is taken, after the mine blind zone data is supplemented, it determines whether to identify multiple interferences in the mine data. This helps solve the problems of existing technologies that rely mainly on sensor point monitoring, have small coverage areas, and poor data acquisition timeliness. Functional grid division achieves full-area monitoring coverage, and the combination of blind zone supplementation and validity verification forms a basis for cross-validation of multi-source data, improving the completeness and timeliness of data acquisition. If no supplementation is taken, the validity of the mine monitoring data is verified, and the determination of whether to identify multiple interferences in the mine data is based on the validity verification results. If so, the determination of whether to suppress electromagnetic radiation interference and correct environmental interference is based on the results of the multiple interference identification. This helps solve the problem of data drift and distortion caused by sensor dampness and electromagnetic radiation. By replacing static filtering with targeted interference suppression and dynamic calibration mechanisms, the reliability of mine data is improved, providing high-quality mine data support for dynamic ventilation network calculation. If not, dynamic ventilation network calculation is performed. After the dynamic calculation, mine ventilation demand analysis is conducted. Based on the results of the mine ventilation demand analysis, it is determined whether to take mine ventilation regulation measures, including adjusting the total mine ventilation volume and the ventilation volume of mine branch roadways. If not, the current ventilation status is deemed qualified. If so, after the mine ventilation regulation is completed, it is determined whether the mine ventilation demand analysis is qualified. If not, a ventilation regulation failure prompt is sent. If qualified, the current ventilation status is deemed qualified. This helps to solve the problems of large prediction errors, insufficient flexibility of control strategies, and excessive or delayed air volume adjustment in existing technologies. By combining multi-source data such as gas concentration to achieve priority adjustment, the flexibility of control strategies and the accuracy of ventilation regulation are improved, effectively solving the problem of low accuracy of mine ventilation decision data in existing technologies.

[0009] 2. This invention divides the grid according to core functional requirements (such as gas concentration monitoring, air volume monitoring, etc.), achieving a one-to-one correspondence between grid units and preset sensors of the same function. This ensures a high degree of compatibility between the mine data monitoring area coverage and functional requirements. Simultaneously, it specifically selects two key parameters: the grid-sensor straight-line distance and the sensor signal strength deviation, reducing the problem of blind spots caused by misjudging solely based on distance parameters and improving the comprehensiveness and accuracy of mine data monitoring coverage assessment. It determines whether the grid-sensor straight-line distance is greater than a preset monitoring radius threshold and whether the sensor signal strength deviation is less than a preset signal strength deviation threshold. If so, mine blind spot data is supplemented; otherwise, the validity of the mine monitoring data is verified. This helps reduce decision-making biases caused by directly including mine blind spot data in subsequent stages. At the same time, qualified mine data can be selected and output directly, improving data processing efficiency.

[0010] 3. By selectively choosing wind speed signal distortion and ventilation fan harmonic intensity as characteristic parameters of electromagnetic radiation interference, and sensor data drift and mine data fluctuation quantification indicators as characteristic parameters of environmental interference, this approach helps to accurately characterize the distortion features of mine monitoring data under the influence of electromagnetic radiation and environmental factors in mine ventilation scenarios. This overcomes the shortcomings of existing technologies that only use static filtering or single threshold to remove outliers, which cannot effectively distinguish dynamic interference types and cannot fundamentally eliminate mine data distortion. By judging whether the characteristic parameters of electromagnetic radiation interference meet the electromagnetic radiation interference judgment conditions, electromagnetic radiation interference suppression is adopted if so; otherwise, environmental interference correction is adopted if the characteristic parameters of environmental interference meet the environmental interference judgment conditions; otherwise, dynamic calculation of the ventilation network is performed. This enables accurate identification and targeted processing of interference types in mine monitoring data, and automatically triggers corresponding electromagnetic radiation interference suppression or environmental interference correction for the interfered data. This achieves accurate matching and resolution of mine monitoring data distortion problems and interference root causes, providing reliable and qualified mine data support for subsequent dynamic calculation of the ventilation network. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A flowchart of a data-driven intelligent ventilation decision control method for mines, provided for embodiments of this application; Figure 2 A flowchart outlining the data-driven intelligent ventilation decision-making and control method for mines provided in the embodiments of this application; Figure 3 A logic diagram for completing mine blind zone data in the data-driven intelligent ventilation decision control method provided in the embodiments of this application; Figure 4 This is a schematic diagram of a data-driven intelligent ventilation system for mines, provided as an embodiment of this application. Detailed Implementation

[0013] The following provides explanations of some terms used in this application. It should be noted that these explanations are for the convenience of those skilled in the art and do not constitute a limitation on the scope of protection claimed in this application.

[0014] The embodiments of this application involve at least one, including one or more; where "multiple" means two or more. Furthermore, it should be understood that in the description of this specification, terms such as "first," "second," and "third" are used only for descriptive purposes and should not be construed as indicating relative importance or order. For example, "first device" and "second device" do not represent the degree of importance of the two or their order, but are merely for descriptive distinction. In the embodiments of this application, "and / or" merely describes an association relationship, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects are in an "or" relationship.

[0015] The directional terms mentioned in the embodiments of this application, such as "up", "down", "left", "right", "inner", and "outer", are only for reference to the directions in the accompanying drawings. Therefore, the directional terms used are for better and clearer explanation and understanding of the embodiments of this application, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the embodiments of this application.

[0016] References to "one embodiment," "in some examples," or "some embodiments" as described in the embodiments of this application mean that one or more embodiments of this specification include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in some examples," "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0017] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0018] like Figure 1The diagram shows a flowchart of a data-driven intelligent ventilation decision control method for mines provided in this application. The method includes the following steps: S1, mine monitoring blind zone data analysis. In a mine ventilation scenario, mine monitoring blind zone data analysis is performed, and based on the results of the analysis, it is determined whether to supplement the mine blind zone data. If so, after the mine blind zone data supplementation is completed, it is determined whether to identify multiple interferences in the mine data. By analyzing the mine monitoring blind zone data, the limitations of traditional sensor point monitoring can be solved, achieving full coverage of the mine roadway monitoring space. This reduces the vulnerability of mine data blind zones in ventilation decisions caused by local monitoring gaps. At the same time, by supplementing the blind zone data first and then judging interference identification, the logic ensures that the data entering the subsequent process has spatial integrity, providing comprehensive basic support for multi-source data comprehensive analysis and helping to solve the problems of discontinuous monitoring coverage and data fragmentation in existing technologies.

[0019] S2. Mine data multi-interference monitoring: If not implemented, the validity of the mine monitoring data will be verified, and based on the verification results, it will be determined whether to identify multiple interferences in the mine data. If so, based on the identification results, it will be determined whether to implement electromagnetic radiation interference suppression and environmental interference correction. Mine data multi-interference monitoring helps to achieve refined data governance by first verifying validity and then addressing interference in a targeted manner. This reduces invalid or interfered data from entering the dynamic calculation stage of the wind network, improves the reliability of mine data from the source, and helps to solve the shortcomings of existing technologies, such as vague judgment of interference types and single processing methods, which make it difficult to eradicate data distortion.

[0020] S3. Mine ventilation demand monitoring: If not implemented, dynamic calculation of the ventilation network will be performed. After the dynamic calculation of the ventilation network is completed, mine ventilation demand analysis will be conducted. Based on the results of the mine ventilation demand analysis, it will be determined whether to take mine ventilation regulation measures, including adjustment of the total mine ventilation volume and adjustment of the ventilation volume of mine branch roadways. If not taken, the current ventilation status is deemed qualified. If taken, after the mine ventilation regulation is completed, it will be determined whether the mine ventilation demand analysis is qualified. If not qualified, a ventilation regulation failure prompt will be sent. If qualified, the current ventilation status is deemed qualified. By conducting mine ventilation demand monitoring, it is helpful to dynamically match the mine production conditions with the ventilation supply, reduce excessive or delayed air volume adjustment, improve the adaptability of the mine ventilation system to complex working conditions, and help solve the problems of disconnect between existing technology ventilation decisions and dynamic working conditions, and insufficient control flexibility.

[0021] It should be noted that the data-driven intelligent ventilation decision-making and control method for mines provided in this application relies on a pre-built dedicated data support platform for its research and development and implementation. This platform integrates and stores multi-dimensional data resources, including basic configuration parameters set by ventilation experts and technical teams, such as preset reflection signal thresholds, preset monitoring radius thresholds, preset signal strength deviation thresholds, and preset interference source safety distance thresholds. Simultaneously, the database includes standardized historical mine monitoring data (covering time-series records of wind speed, gas concentration, and wind pressure under different mining stages and geological conditions, along with corresponding ventilation control results) and algorithm training and verification data. (e.g., parameter optimization records during multi-round feature extraction and working condition identification model debugging). This database adopts a hybrid data storage scheme. It manages structured parameters (such as wind network solution model coefficients, PID control parameters, adjustment priority rule tables, etc.) through a relational database and stores unstructured data, such as original sensor waveform sequences and fan operation acoustic data, through a non-relational database. Based on newly collected field monitoring data and ventilation effect feedback, technicians can periodically check and optimize the preset parameters stored in the database, thereby ensuring that the data system continuously adapts to the complex and ever-changing production environment underground, such as actual working conditions such as the advancement of mining faces and changes in ventilation routes.

[0022] In this embodiment, a closed-loop ventilation control system is constructed, encompassing mine data coverage, targeted data processing, and final ventilation decision execution, through mine monitoring blind zone data analysis, mine data multi-interference monitoring, and mine ventilation demand monitoring. This ensures that ventilation decisions are based on complete and reliable mine data support, thereby improving the intelligence and safety assurance level of mine ventilation. Specifically, mine monitoring blind zone data analysis provides a comprehensive mine data foundation for mine data multi-interference monitoring, reducing interference monitoring from being limited to only a portion of effective monitoring areas. Furthermore, the mine data purified by multi-interference monitoring provides a high-quality calculation basis for mine ventilation demand monitoring, ensuring the accuracy of ventilation demand analysis results and ventilation parameter adjustments, forming a virtuous cycle of mutual support and dynamic optimization among the three.

[0023] like Figure 2The diagram shown is a general flowchart of the data-driven intelligent ventilation decision control method for mines provided in this application embodiment. It includes: analyzing mine monitoring blind zone data and obtaining the grid-sensor straight-line distance value and sensor signal strength deviation value; determining whether the grid-sensor straight-line distance value is greater than a preset monitoring radius threshold and whether the sensor signal strength deviation value is less than a preset signal strength deviation threshold; if so, completing the mine blind zone data; otherwise, verifying the validity of the mine monitoring data and obtaining the environmental parameters of the mine monitoring unit; determining whether the environmental parameters of the mine monitoring unit meet the mine data qualification conditions; if so, performing dynamic calculation of the ventilation network; otherwise, identifying multiple interferences in the mine data and obtaining electromagnetic radiation interference characteristic parameters and environmental interference characteristics. The system determines whether the electromagnetic radiation interference characteristic parameters meet the electromagnetic radiation interference judgment conditions. If so, electromagnetic radiation interference suppression is implemented; otherwise, it determines whether the environmental interference characteristic parameters meet the environmental interference judgment conditions. If so, environmental interference correction is implemented; otherwise, dynamic calculation of the ventilation network is performed. After the dynamic calculation of the ventilation network is completed, mine ventilation demand analysis is performed, and ventilation status assessment parameters are obtained. The system determines whether the ventilation status assessment parameters meet the mine ventilation qualification judgment conditions. If they do, the current ventilation status is deemed qualified, and mine data is continuously monitored. Otherwise, mine ventilation adjustment is performed. After the mine ventilation adjustment is completed, the system determines whether the mine ventilation demand analysis is qualified. If so, the current ventilation status is deemed qualified, and mine data is continuously monitored. Otherwise, a ventilation adjustment failure prompt is sent.

[0024] Preferably, the analysis of blind spot data in mine monitoring involves the following process: Based on functional requirements (such as gas concentration monitoring, air volume monitoring, temperature and humidity monitoring, etc.), a spatial grid partitioning algorithm is used to divide the mine roadway into several equal grid units, and the three-dimensional coordinates of the center point of each grid unit monitored by the mine mapping instrument are obtained; wherein, the grid unit division corresponds one-to-one with the functional requirements, the functional requirements specify the monitoring type, and the grid unit is divided only for the sensor corresponding to that monitoring type; the grid-sensor straight-line distance value is obtained based on the three-dimensional coordinates of the center point of each grid unit; the grid-sensor straight-line distance value represents the straight-line distance between the three-dimensional coordinates of the center point of each grid unit monitored by the mine mapping instrument, obtained based on the spatial coordinate Euclidean distance algorithm, and the three-dimensional coordinates of the installation center of the preset type sensor (such as gas sensor, wind speed sensor, temperature and humidity sensor, etc.) closest to the center point of the grid unit, wherein the preset type sensor is preset by the preset personnel according to the functional requirements; the sensor signal strength deviation value is obtained, and the sensor signal strength deviation value... The difference between the amplitude of the reflected signal received by the preset type sensor closest to the center point of the grid cell and the preset reflected signal threshold is used. The amplitude of the reflected signal is represented by the peak value of the reflected signal monitored by the preset type sensor during the data integrity analysis period, and the preset reflected signal threshold is represented by the average amplitude of the reflected signal in the grid cell area over a historical period. The system determines whether the grid-sensor straight-line distance is greater than a preset monitoring radius threshold and whether the sensor signal strength deviation is less than a preset signal strength deviation threshold. If so, the corresponding grid cell is marked as a mine blind zone cell, and mine blind zone data is completed based on the mine blind zone cell. Otherwise, the corresponding grid cell is marked as a mine monitoring unit, and the validity of the mine monitoring data is verified based on the mine monitoring unit. The preset monitoring radius threshold is represented by the average value of the grid-sensor straight-line distance over a historical period, and the preset signal strength deviation threshold is represented by the average value of the sensor signal strength deviation over a historical period.

[0025] Specifically, the validity verification of mine monitoring data is used to identify whether sensor data within the mine monitoring unit is distorted due to environmental interference (such as electromagnetic radiation or high humidity), reducing the influx of interference-free data into subsequent mine data multi-interference identification, and improving mine data processing efficiency. The specific process is as follows: Obtain environmental parameters of the mine monitoring unit, including the distance value between the mine monitoring unit and the interference source to quantify the risk of interference to sensor signal transmission based on the spatial relationship between the mine monitoring unit and the interference source, and the humidity of the mine monitoring unit to assess the impact of the humid environment on sensor accuracy and data stability. The distance value between the mine monitoring unit and the interference source represents the three-dimensional coordinates of the center point of the mine monitoring unit obtained based on the Euclidean distance algorithm of spatial coordinates, monitored by the mine surveying instrument, and the three-dimensional coordinates of the center point of the interference source (such as a tunneling machine). The linear distance in 3D coordinates; the humidity of the mine monitoring unit is represented by the arithmetic mean of the humidity data collected by the humidity sensor closest to the center point of the mine monitoring unit within the data integrity analysis period; it is used to determine whether the environmental parameters of the mine monitoring unit meet the qualified conditions of mine data. If so, the corresponding mine data is marked as qualified mine data, and the wind network dynamic solution is performed based on the qualified mine data; otherwise, multiple interference identification of mine data is performed; the qualified conditions of mine data indicate that the distance value of the interference source of the mine monitoring unit is less than the preset safe distance threshold of the interference source, and the humidity of the mine monitoring unit is less than the preset maximum humidity threshold. The preset safe distance threshold of the interference source is represented by the average value of the distance value of the interference source of the mine monitoring unit in the historical time period, and the preset maximum humidity threshold is represented by the average value of the humidity of the mine monitoring unit in the historical time period.

[0026] In this embodiment, by analyzing mine monitoring blind zone data and combining functional requirements, precise matching between grid units and monitoring sensors is achieved. Blind zone identification is completed using two parameters: the grid-sensor straight-line distance value and the sensor signal strength deviation value. This helps fill the coverage gaps in mine monitoring space, solves the problem of data fragmentation, reduces the safety risk of ventilation decision deviations due to incomplete mine data monitoring, improves the integrity and continuity of mine monitoring data, and realizes the transformation from point monitoring to full-area monitoring. At the same time, the validity verification of mine monitoring data helps to screen out distorted data affected by environmental interference, equipment aging, and other factors in advance, improves mine data processing efficiency, and provides comprehensive and reliable basic support for subsequent mine data processing and ventilation decisions.

[0027] like Figure 3 The diagram shown is a logic diagram for completing mine blind zone data in a data-driven intelligent ventilation decision-making and control method provided in this application embodiment. Figure 3It can be seen that: to complete the data of the mine blind area, firstly, the data of the mine blind area units is fused and completed, then the validity of the fused and completed data is verified, and the correlation index of the mine blind area data is obtained. It is then determined whether the correlation index of the mine blind area data is greater than the preset high validity threshold. If so, the mine data is identified as having multiple interferences. Otherwise, it is determined whether the correlation index of the mine blind area data is less than the preset low validity threshold. If so, a warning that there may be ventilation abnormalities in the mine blind area is sent and manual intervention is performed. Otherwise, a warning that the completed data is invalid is sent.

[0028] Preferably, the specific process for completing mine blind zone data is as follows: S101, Obtain mine reference data; Mine reference data represents the mine monitoring data sequence of a preset number of preset type sensors (such as 3 gas sensors, etc.) that are closest to the mine blind zone unit and meet the blind zone data completion conditions; The mine monitoring data sequence is the continuous monitoring data generated by the sensors according to the sampling frequency during the data integrity analysis period, such as the continuous change value of gas concentration monitored by the gas sensor, represented as {(t1,c1), (t2,c2), ..., (tm,cm)}, where t1, t2, ..., tm represent the sampling times generated sequentially according to the preset sampling frequency during the data integrity analysis period, and c1, c2, ..., cm represent the effective value of gas concentration monitored by the gas sensor at the corresponding sampling time. The preset sampling frequency is set in advance by preset personnel; Fusion generation means that for multiple monitoring data of preset sensors at the same sampling point, the average value is taken as the value of that sampling point. The mine reference data; the blind zone data completion condition indicates that the mine data acquisition success rate is greater than the preset data acquisition success threshold, and the mine data fluctuation coefficient is less than the preset data fluctuation threshold. The preset data acquisition success threshold is represented by the average mine data acquisition success rate over a historical time period, and the preset data fluctuation threshold is represented by the average mine data fluctuation coefficient over a historical time period. The mine data acquisition success rate is represented by the ratio of the amount of data collected by the preset type of sensors monitored by the data monitoring instrument to the preset total amount of data collected during the data integrity analysis period. The preset total amount of data collected is set in advance by the preset personnel. The mine data fluctuation coefficient is represented by the ratio of the standard deviation of mine data (such as the gas concentration value collected by the gas sensor) to the mean of mine data during the data integrity analysis period. The mean of mine data is represented by the sum of the values ​​of all mine data sampling points during the data integrity analysis period, and the ratio of the summation result to the number of sampling points.

[0029] Specifically, the formula for the fluctuation coefficient of mine data is as follows:

[0030]

[0031] Where C represents the mine data fluctuation coefficient, i represents the mine data sampling point number within the data integrity analysis period, i=1, 2, 3, ..., n, n represents the total number of mine data sampling points, u represents the mine data mean, and x i This represents the value of the mine data corresponding to the i-th sampling point. Specifically, the mine data fluctuation coefficient is used to quantify the relative fluctuation of the mine data within the data integrity analysis period. The mine data covers multi-dimensional core data related to mine safety monitoring and ventilation decisions, including but not limited to gas concentration (methane concentration) monitored by gas sensors, wind speed of each branch roadway monitored by wind speed sensors, wind pressure value of roadway nodes monitored by wind pressure sensors, and underground air humidity monitored by temperature and humidity sensors. These data are key basic information supporting mine ventilation status perception, blind spot data completion, and ventilation decisions. Their fluctuation directly affects the stability and reliability of the data.

[0032] S102. Data fusion and completion of mine blind zone units based on Kriging interpolation method to obtain fused and completed mine blind zone unit data; Mine blind zone unit data fusion and completion refers to the process of using a mine reference dataset as a basis, fusing roadway condition related data through the Kriging interpolation algorithm to obtain fused and completed mine blind zone unit data. The roadway condition related data includes geometric parameters such as the length and cross-sectional dimensions of the roadway where the blind zone is located. The specific process of mine blind zone unit data fusion and completion is as follows: Based on the Kriging interpolation algorithm, the mine monitoring data in the mine reference dataset is used as sample points. A spatial variogram function is constructed by combining the geometric parameters such as the length and cross-sectional dimensions of the roadway where the blind zone is located. Then, the spatial variogram function is obtained by Kriging interpolation... The method estimates the data value to be completed for the blind zone unit of the mine, and then weights and fuses the data value to be completed with the working condition adaptation value derived from the roadway geometric parameters to finally obtain the fused and completed data of the blind zone unit of the mine. The specific process of constructing the spatial variogram function by combining the geometric parameter data such as the length and cross-sectional dimensions of the roadway where the blind zone is located is as follows: The mine monitoring data in the mine reference dataset is used as sample points. The monitoring parameter values ​​of each sample point, the coordinate values ​​under the linear coordinate system established along the roadway axis with the starting point of the roadway where the blind zone is located as the origin, and the roadway geometric parameters (cross-sectional area, perimeter, slope) of the sample point and the blind zone are collected. The cross-sectional area similarity and slope difference between the blind zone and each sample point are calculated.

[0033] Specifically, the formula for cross-sectional area similarity is as follows:

[0034] Among them, S sim This represents the similarity of cross-sectional areas, where S is the cross-sectional area of ​​the roadway where the blind zone is located. iLet be the cross-sectional area of ​​the roadway where the i-th sample point is located. The cross-sectional area of ​​the roadway is obtained by monitoring the underground laser cross-section measuring instrument. The cross-sectional area similarity is used to quantify the degree of matching between the blind zone and the cross-sectional area of ​​the roadway where the sample point is located. Specifically, the formula for slope difference is:

[0035] Where, θ p θ represents the slope of the alleyway where the blind spot is located. i This represents the slope of the roadway where the i-th sample point is located. Specifically, the slope of the roadway where the blind zone is located is obtained by monitoring the underground slope measuring instrument. The slope of the roadway where the blind zone is located is used to quantify the degree of difference between the slope of the blind zone and the roadway where the mine monitoring data sample point is located.

[0036] Based on cross-sectional area similarity and slope difference, a geometric parameter correction factor is constructed to weaken the interference of different roadway geometric parameters on the spatial correlation of sample points. The specific construction process of the geometric parameter correction factor is as follows:

[0037] Where λ represents the geometric parameter correction factor, a represents the correction weight for cross-sectional area similarity, b represents the correction weight for slope difference, and θ0 represents the preset mine roadway slope, which is set in advance by the personnel (e.g., the industry standard value is 15°). The specific process for obtaining the correction weights for cross-sectional area similarity and slope difference involves first collecting historical data related to blind spot completion of similar roadways in the mine, covering completion results corresponding to different cross-sectional area matching situations and different slope difference situations, as well as subsequent actual verification monitoring data. Then, the correction weights for cross-sectional area similarity and slope difference are analyzed. Based on the degree of influence of blind spot data completion accuracy, factors with a greater impact on completion accuracy have a higher corresponding correction weight. This yields the initial weight tendency. Then, multiple different weight combinations are set and substituted into the construction logic of the geometric parameter correction factor to complete multiple sets of blind spot completion simulation calculations. The deviation between the completed data obtained from each calculation and the actual monitoring data is compared. Finally, the weight set that minimizes the deviation between the completed data and the actual data is selected as the final correction weight. Subsequently, the weights will be recalibrated periodically based on newly added monitoring and completion data from the mine to adapt to changes in roadway conditions.

[0038] For single mine monitoring data (such as gas concentration, wind speed, etc.), the corrected semi-variance value is calculated to quantify the spatial variability of the mine monitoring parameters under different spatial distances, corresponding to the distance along the roadway axis between any two sample points. The specific calculation process is as follows:

[0039] Where r(h,λ) represents the corrected semivariogram, h represents the linear distance between sample points along the roadway axis, j_h represents the grouping index of sample point pairs with a linear distance of h along the roadway axis, j_h=1,2,3,...,N(h), N(h) represents the number of sample point pairs with a distance of h, and Z(x j_h () indicates that in the j-th sample point pair with a distance of h, the coordinates in the one-dimensional linear coordinate system are x. j_h The mine monitoring parameter values ​​(such as gas concentration, wind speed, etc.) of the sample points at the location, Z(x j_h+h ) represents the pair of sample points in the j-th group with a distance of h, where x is the distance between them. j_h The location of a sample point is determined by the linear distance h along the roadway axis from another sample point, representing the mine monitoring parameter value (such as gas concentration, wind speed, etc.). Specifically, the linear distance between sample points along the roadway axis is represented by the absolute value of the coordinate difference between the sample points in a one-dimensional linear coordinate system, used to characterize the spatial position difference of the sample points. The number of sample point pairs with a distance of h is represented by the number of sample point pairs with a distance equal to h along the roadway axis from all sample points monitored by the counter, used to determine the effective sample size for calculation, Z(x). j_h ) and Z(x j_h+h It is represented by real-time monitoring data collected by sensors at corresponding sample points.

[0040] Based on the data pairs of distances between sample points along the roadway axis and the corresponding corrected semi-variance values, a pre-defined spherical model is fitted to obtain the spatial variability function, which is as follows:

[0041] η(h,λ) represents the corrected spatial semivariogram, C0 represents the nugget value, C represents the sill value, and p represents the range. The nugget value and sill value are obtained by fitting a preset spherical model, reflecting the measurement error of the mine sensor and the degree of random variation of the mine monitoring parameters at the microscale. The range is represented by the linear distance h between the sample points along the roadway axis when the corrected spatial semivariogram reaches (C0+C), reflecting the effective range of the spatial correlation of the mine monitoring parameters.

[0042] The spatial coordinates of the blind zone, the distance between sample points along the roadway axis, and the spatial variability model are input into the Kriging equations to finally output the data values ​​to be completed for the mine blind zone unit. The specific training process of the spherical model is as follows: First, based on the overall fluctuation characteristics of the mine monitoring data and the spatial range of the sample coverage, the initial benchmark parameters of the model (null value and off-center sill value) are set, corresponding to the initial assumptions of the influence of sensor measurement error, the strength of parameter spatial correlation, and the effective range of spatial correlation, respectively. Then, the matching data of the spatial distance of sample points and the corresponding corrected half-variability values ​​are input into the initialized model to obtain the theoretical spatial variability output by the model. The model's baseline parameters are then adjusted repeatedly using the least squares optimization logic. The theoretical values ​​output by the model are compared with the deviations of the corrected semi-variable values ​​obtained from actual calculations. The overall level of deviation is reduced to the minimum. Then, cross-validation is used to remove each sample point in turn. The model is rebuilt with the remaining sample points, and the monitoring parameter values ​​of the removed sample points are estimated. The deviation between the estimated results and the actual monitoring results is compared. If the average level of deviation meets the industry accuracy requirements for mine monitoring, the current model parameters are determined as the final parameters, and training is completed. If not, the initial baseline parameters are readjusted and the above process is repeated.

[0043] The specific process of weighted fusion is as follows: First, two types of basic data are acquired for fusion. The first type is the data value to be completed after Kriging interpolation using a customized spatial variogram function, reflecting the actual spatial distribution trend of blind zone monitoring parameters. The second type is the working condition adaptation value derived from the geometric parameters of the roadway where the blind zone is located, combined with the parameter transfer law of mine ventilation fluid dynamics. This value is derived from the fluid operation logic of the roadway physical structure and reflects the theoretical reasonable range of blind zone monitoring parameters. Then, the weights of the two types of basic data are determined according to their reliability, and the weight coefficients of the data value to be completed and the working condition adaptation value are obtained. The result of multiplying the data value to be completed with its corresponding weight coefficient is summed with the result of multiplying the working condition adaptation value with its corresponding weight coefficient to obtain the fused and completed data of the mine blind zone unit. The specific formula for weighted fusion is as follows:

[0044]

[0045]

[0046] Where Z1(X) represents the data value to be completed, X represents the target mine monitoring parameter, k represents the sequence number of the target mine monitoring parameter sample point, k=1,2,3,...,f, f represents the total number of target mine monitoring parameter sample points, αk represents the Kriging interpolation weight of the k-th target mine monitoring parameter sample point, wk represents the monitoring parameter value of the k-th target mine monitoring parameter sample point (such as the gas concentration value collected by the gas sensor, the roadway wind speed value collected by the wind speed sensor, etc.), Z2(X) represents the working condition adaptation value, βk represents the geometric matching weight of the k-th target mine monitoring parameter sample point, yk represents the blind zone working condition inference value corresponding to the k-th target mine monitoring parameter sample point, Z(X) represents the mine blind zone unit fusion and completion data, W represents the weight coefficient of the data value to be completed, and Y represents the weight coefficient of the working condition adaptation value. The target mine monitoring parameter is the single type of mine monitoring parameter to be completed, such as gas concentration, branch roadway wind speed, roadway node wind pressure value, etc.

[0047] Specifically, the criteria for determining the weighting coefficients of the data to be supplemented, the weighting coefficients of the working condition adaptation values, the weighting of the Kriging interpolation, and the weighting of the geometric matching are the stability of the monitoring data of the reference sample points and the degree of matching between the geometric parameters of the blind zone and the roadway where the reference sample points are located. The smaller the fluctuation and the higher the stability of the data of the reference sample points, the higher the weighting of the interpolation estimation value. The higher the degree of matching between the geometric parameters of the blind zone and the reference sample points, the higher the weighting of the working condition adaptation value. The sum of the two types of weights is 1 to ensure the rationality of the fusion ratio. If there is a significant anomaly in one type of basic data (such as exceeding the mine safety monitoring threshold), the weighting of that type of data will be dynamically reduced.

[0048] S103. Verify the validity of the fused and completed data. The specific process is as follows: Based on correlation analysis methods (such as Pearson correlation coefficient, Spearman correlation coefficient, etc.), verify the correlation between the fused and completed data of the mine blind zone unit and the mine monitoring data of the mine monitoring unit closest to the mine blind zone unit. Obtain the mine blind zone data correlation index used to evaluate the credibility of the fused and completed data of the mine blind zone unit. The specific acquisition process is as follows: Substitute the fused and completed data sequence of the mine blind zone unit and the mine monitoring data sequence of the mine monitoring unit closest to the blind zone unit in the same time dimension into the calculation formula of the correlation analysis method. By calculating the ratio of the product of the covariance and the standard deviation of the two data sequences, the mine blind zone data correlation index that quantifies the linear correlation between the two is obtained; If the correlation index of the mine blind zone data is greater than the preset high validity threshold, it indicates that the fused and supplemented data is effective, and multiple interferences in the mine data are identified. The preset high validity threshold is represented by the average value of the correlation index of the mine blind zone data over a historical period. Otherwise, it is determined whether the correlation index of the mine blind zone data is less than the preset low validity threshold. If it is, it indicates that the fused and supplemented data of the mine blind zone unit and the mine monitoring data have an inverse relationship. The mine blind zone unit may have abnormal conditions such as ventilation disturbances caused by ventilation circulation and abnormal gas accumulation. A warning of ventilation abnormality in the mine blind zone is sent, and manual intervention is carried out. Otherwise, a warning of invalid supplemented data is sent. The preset low validity threshold is set in advance by preset personnel, and the preset high validity threshold is greater than the preset low validity threshold.

[0049] In this embodiment, by supplementing blind spot data in the mine, the missing data in the blind spot is accurately filled based on the mine reference data and ventilation condition characteristics. This ensures the logical consistency and condition adaptability of the supplemented data with the surrounding effective monitoring data. It helps to fill the spatial gaps in mine monitoring, solves the pain points of discontinuous and fragmented data coverage in the traditional point monitoring mode, reduces the safety risks of ventilation network calculation deviation and misjudgment of ventilation demand caused by missing mine data in blind spots, improves the overall integrity, temporal continuity and overall reliability of mine monitoring data, and realizes the transformation from local effective data to high-quality data across the entire domain. It provides comprehensive and accurate data support for dynamic calculation of ventilation network and ventilation decision-making, further strengthens the adaptability of the ventilation system to complex mine environments, and ensures the scientific and targeted nature of ventilation adjustment strategies.

[0050] Preferably, the identification of multiple interferences in mine data involves the following process: Obtaining electromagnetic radiation interference characteristic parameters, including wind speed signal distortion (used to assess the degree of data distortion caused by electromagnetic interference from wind speed sensors) and fan harmonic intensity (used to quantify the energy intensity of electromagnetic interference from major ventilation equipment); obtaining environmental interference characteristic parameters, including sensor data drift (used to quantify the degree of sensor performance degradation) and mine data fluctuation quantification indicators (used to quantify the energy intensity of electromagnetic interference from major ventilation equipment); and determining the wind speed signal distortion by comparing the amplitude of the mine wind speed signal monitored by the wind speed sensor at the corresponding sampling point within the data integrity analysis period with a preset value. The root mean square error of the wind speed signal amplitude is represented by the average value of the mine wind speed signal amplitude over a historical period. The harmonic intensity of the ventilation fan is represented by the sum of squares of the abnormal increments of the signal amplitude at each sampling point in the current signal spectrum of the main ventilation fan power supply circuit. The abnormal increment of the signal amplitude is represented by the difference between the signal amplitude at each sampling point in the current signal spectrum of the ventilation fan monitored by the current sensor and the preset frequency amplitude, where the preset frequency amplitude is represented by the average value of the signal amplitude at each sampling point in the current signal spectrum of the ventilation fan over a historical period.

[0051] Specifically, sensor data drift is determined by linearly fitting the sensor data (such as wind resistance data, gas data, etc.) sequence using the least squares method within the data integrity analysis period to obtain the slope of the sensor data trend change. The absolute value of the sensor data trend change slope is then multiplied by the duration of the data integrity analysis period. Mine data fluctuation quantification is determined by the ratio of the mine data fluctuation amplitude to a preset stable fluctuation threshold within the data integrity analysis period. The mine data fluctuation amplitude is represented by the standard deviation of the mine data, where the preset stable fluctuation threshold is represented by the average value of the mine data fluctuation amplitude over historical periods. Finally, it is determined whether the electromagnetic radiation interference characteristic parameters meet the electromagnetic radiation interference judgment conditions; if so, electromagnetic radiation... Interference suppression is performed; otherwise, the environmental interference characteristic parameters are checked to see if they meet the environmental interference judgment conditions. If they do, environmental interference correction is applied; otherwise, dynamic calculation of the wind network is performed. The electromagnetic radiation interference judgment condition indicates that the wind speed signal distortion is greater than the preset distortion threshold and the fan harmonic intensity is greater than the preset harmonic threshold. The preset distortion threshold is represented by the average value of the wind speed signal distortion over a historical period, and the preset harmonic threshold is represented by the average value of the fan harmonic intensity over a historical period. The environmental interference judgment condition indicates that the sensor data drift is greater than the preset drift threshold and the mine data fluctuation quantification index is greater than the preset interference fluctuation threshold. The preset drift threshold is represented by the average value of the sensor data drift over a historical period, and the preset stability threshold is represented by the average value of the data fluctuation coefficient over a historical period.

[0052] In this embodiment, by identifying multiple interferences in mine data, electromagnetic radiation interference characteristic parameters such as wind speed signal distortion and ventilation fan harmonic intensity are accurately selected, as well as environmental interference characteristic parameters such as sensor data drift and mine data fluctuation coefficient. Combined with dual threshold judgment logic, interference types are distinguished and targeted processing is triggered. This helps to comprehensively capture data distortion problems caused by electromagnetic radiation and environmental factors in the complex environment of the mine, breaking the limitations of traditional single filtering or static threshold processing. It enables accurate positioning and classification management of dynamic interference, while reducing invalid data entering the subsequent ventilation network calculation and ventilation decision-making stages. This reduces the safety risks of large prediction errors in ventilation network and delayed or excessive adjustment strategies due to data distortion, improves the reliability, stability and accuracy of mine monitoring data, and realizes the entire process from passively eliminating anomalies to actively identifying interference. It strengthens the adaptability of the mine ventilation system to the dynamic working conditions of the mine and ensures the scientific and timely nature of ventilation decisions.

[0053] Preferably, the specific process for suppressing electromagnetic radiation interference is as follows: S201, obtain the mine signal spectrum distribution based on Fast Fourier Transform; S202, obtain the discrete spectrum interference components of the mine data based on the mine signal spectrum distribution, and generate a mine signal interference spectrum feature set. The specific generation process is as follows: screen the amplitude of the mine signal spectrum distribution at each frequency point, select discrete mine signal components with amplitudes greater than the preset mine signal amplitude threshold as interference components, extract the interference component center frequency and interference component bandwidth of each interference component, and integrate these feature parameters according to the preset format to generate a mine signal interference spectrum feature set containing the feature information of all discrete spectrum interference components. The preset format is set in advance by the preset personnel. The discrete spectrum interference component of the mine signal represents the discrete mine signal component in the mine signal spectrum distribution whose amplitude is greater than a preset mine signal amplitude threshold. The preset mine signal amplitude threshold is represented by the average amplitude of the mine signal over a historical time period. The discrete spectrum interference component of the mine signal includes at least the center frequency and bandwidth of each interference component. The center frequency of the interference component represents the frequency value corresponding to the point of maximum amplitude of the discrete spectrum interference component on the frequency axis. The bandwidth of the interference component represents the frequency range covered by the discrete spectrum interference component of the mine signal. S203. Dynamically construct the corresponding digital band-stop filter bank based on the mine signal interference spectrum feature set. The specific construction process is as follows: using interference... The center frequency of the interference component serves as the center stopband frequency of the corresponding digital bandstop filter. Simultaneously, the stopband bandwidth of the corresponding digital bandstop filter is calculated by multiplying the bandwidth of the interference component by a preset bandwidth expansion factor. The preset bandwidth expansion factor is pre-set by designated personnel to compensate for frequency dispersion that may occur during transmission of the interference signal, ensuring that the stopband range fully covers the actual interference frequency fluctuation range. The mine data sequence is input into the digital bandstop filter bank for electromagnetic interference filtering. The mine data sequence represents a continuous numerical sequence obtained by sampling the original mine monitoring signal at fixed time intervals, and is a time-domain data sequence based on the same data source as the original signal used for obtaining the spectral distribution via fast Fourier transform. S204, Mine Data... After magnetic interference filtering is completed, the signal-to-noise ratio (SNR) of the mine data is obtained. Based on this SNR, the band-stop filtering effect is verified. The specific process is as follows: It is determined whether the mine data SNR is greater than a preset SNR threshold. If so, the corresponding mine monitoring data is marked as qualified mine data, and dynamic wind network calculation is performed based on the qualified mine data. Otherwise, environmental interference correction is applied. The preset SNR threshold is represented by the average SNR of the mine data over a historical time period. The mine data SNR is represented by the ratio of the average power of the signal in the fundamental frequency band monitored after re-performing a fast Fourier transform of the mine data to the noise power. The average power in the fundamental frequency band is represented by the sum of squares of the signal amplitudes within the fundamental frequency band monitored by the spectrum analyzer.Noise power is represented by calculating the sum of squares of the signal amplitude across the entire frequency band, excluding the fundamental and harmonic bands.

[0054] Specifically, the environmental interference correction process is as follows: The environmental correction coefficients for the mine sensors are obtained by inputting environmental interference characteristic parameters and the humidity of the mine monitoring unit into a preset environmental correction mapping set, and then querying the data to obtain the corresponding mine sensor environmental correction coefficients. The amplitude corresponding to the mine sensor environmental correction coefficient is set as the adjustment step size. Since the mine sensor environmental correction coefficient is positively correlated with the degree of environmental interference, the adjustment step size increases as the mine sensor environmental correction coefficient increases to optimize correction efficiency. Based on the adjustment step size, the sensor data correction amount is gradually adjusted in the opposite direction of the mine data drift, and this sensor data correction amount is superimposed on the original... In the initial mine monitoring data (after each sensor data correction adjustment, the data correction validity value is reacquired; if the data correction validity value is not greater than the preset validity threshold, the adjusted sensor data correction amount is used as the initial value for the next adjustment, and the sensor data correction amount is gradually adjusted in the opposite direction of the mine data drift), this helps to avoid overcorrection caused by excessive single correction amplitude, improves correction efficiency in severely interfered scenarios, and ensures correction accuracy in slightly interfered scenarios, achieving a balance between sensor environmental correction effect and efficiency. The sensor data correction amount represents the reduction of the impact of mine humidity on sensor measurement results. The deviation is a compensation value calculated based on the environmental correction coefficient and adjustment step size of the mine sensor, which needs to be superimposed on the original mine monitoring data. The magnitude of this value is adapted to the environmental interference characteristic parameters and the humidity of the mine monitoring unit. The opposite direction of data drift is determined based on the analysis of environmental interference characteristic parameters. For example, if dust causes positive drift in the sensor data, the adjustment direction is negative, ensuring that the correction amount accurately offsets the interference. The validity value of the data correction is continuously monitored. When the validity value exceeds a preset validity threshold, the corresponding mine monitoring data is marked as qualified mine data, and dynamic calculation of the ventilation network is performed based on the qualified mine data. The preset validity threshold... The data correction validity value is represented by the average value of historical data over a specific time period. If the data correction validity value is not greater than the preset validity threshold, environmental interference correction continues. If the sensor data correction value is greater than the preset maximum data correction threshold, but the data correction validity value is still not greater than the preset validity threshold, an environmental correction processing failure prompt is sent. The preset maximum data correction threshold is set in advance by preset personnel. The data correction validity value is represented by the Pearson correlation coefficient between the mine data obtained after environmental interference correction and the preset mine data, which is used to quantify the correction effect of environmental interference correction processing. The preset mine data is represented by the average value of mine data over a specific time period.

[0055] It should be noted that in the data-driven intelligent ventilation decision-making and control method for mines provided in this application embodiment, the construction process and data foundation of the preset environmental correction mapping set on which the environmental interference correction step relies need further explanation. This mapping set, as the core basis for determining the environmental correction coefficient, has its internal parameter relationships calibrated by technicians based on actual mine field measurement data and stored in a dedicated system database, providing data support for accurately obtaining sensor correction coefficients under different environmental conditions.

[0056] Specifically, the construction of this environmental correction mapping set integrates the quantitative analysis results of multi-source mine environmental parameter combinations and corresponding sensor error characteristics. Its data sources cover typical environmental interference samples under different mining conditions, including environmental interference characteristic parameters such as electromagnetic radiation intensity, temperature gradient change, and mechanical vibration spectrum characteristics, as well as multi-dimensional combination data with mine air humidity conditions. By analyzing the systematic deviation impact of each parameter combination on sensor monitoring data, a quantitative mapping relationship between parameter combinations and corresponding correction coefficients is established.

[0057] During the construction of the mapping set, technicians configure differentiated weights based on the degree of influence of parameter combinations on the reliability of sensor data. For example, when the electromagnetic radiation intensity exceeds the threshold and the humidity reaches the critical saturation state, a higher correction coefficient is matched to offset the compound interference effect. At the same time, through statistical analysis of long-term monitoring data, abnormal calibration data caused by non-environmental factors such as temporary equipment failures and power supply fluctuations are eliminated to ensure that the correlation between each parameter combination and the correction coefficient in the mapping relationship has statistical significance and engineering applicability, thereby ensuring the adaptability and reliability of the mapping set in complex mining environments.

[0058] In this embodiment, by suppressing electromagnetic radiation interference and correcting environmental interference, targeted processing is triggered based on accurately identified interference characteristic parameters to address different types of data distortion caused by electromagnetic radiation and environmental factors in the complex environment of mines. This helps to fundamentally solve the technical pain points of traditional static filtering, which can only passively remove anomalies, cannot distinguish interference types, and is difficult to adapt to dynamic interference. It reduces data drift and distortion caused by electromagnetic radiation superposition and environmental aging, reduces ventilation network calculation deviations and ventilation decision errors caused by unreliable mine data, and improves the accuracy, stability, and reliability of mine monitoring data. It provides a real and reliable quantitative basis for mine ventilation demand analysis and improves the operating efficiency and safety assurance level of the mine ventilation system.

[0059] Preferably, the specific process of dynamic ventilation network calculation is as follows: Qualified mine data is input into a preset ventilation network calculation model (such as the node method or loop method), and ventilation quantification parameters are output. The specific construction process of the ventilation network calculation model is as follows: First, based on the actual mine roadway topology (including roadway connection relationships, length, cross-sectional dimensions, and other physical parameters), a ventilation network topology diagram is constructed using mine 3D modeling technology (such as CAD drawing software). The correspondence between nodes (roadway intersections) and branches (each roadway segment) is clarified. Then, based on fluid mechanics and ventilation engineering principles, characteristic parameters such as wind resistance and local resistance are assigned to each branch. The model incorporates characteristic curve equations for ventilation power equipment (such as fans), and then establishes a set of mathematical equations based on the principles of mass conservation (nodal airflow balance) and energy conservation (loop air pressure balance). These equations are solved using iterative algorithms (such as the Scoud-Hensley method), ultimately forming a ventilation network solution model that can dynamically receive monitoring data and output quantitative results. The ventilation quantification parameters include the actual airflow in each mine roadway and the operating status parameters of the ventilation facilities (such as the main fan operating power, real-time door opening, and window ventilation resistance). Based on the output ventilation quantification parameters, a mine ventilation demand analysis is performed to quantify the mine's ventilation status.

[0060] Specifically, the process of mine ventilation demand analysis is as follows: Obtain ventilation status assessment parameters, including mine roadway ventilation deviation values ​​reflecting the deviation between air supply and demand, and methane concentration ratio values ​​used to assess the ventilation system's ability to control methane dilution. The mine roadway ventilation deviation value is the difference between the actual air volume of each roadway and the preset safe air volume for that roadway, where the preset safe air volume is set in advance by designated personnel. The methane concentration ratio value is the ratio of the methane concentration in each roadway monitored by mine methane sensors to the preset upper limit of methane concentration, where the preset upper limit of methane concentration is set in advance by designated personnel. Based on the ventilation status assessment parameters... The process for determining mine ventilation demand is as follows: First, it is determined whether the ventilation status assessment parameters meet the mine ventilation qualification criteria. If they do, the current ventilation status is deemed qualified, and mine data is continuously monitored. Otherwise, mine ventilation is adjusted. The mine ventilation qualification criteria indicate that the absolute value of the mine roadway ventilation deviation is less than the preset air volume adaptability threshold, and the methane concentration percentage is greater than the preset methane concentration threshold. The preset air volume adaptability threshold is represented by the average absolute value of the mine roadway ventilation deviation over a historical time period, and the preset methane concentration threshold is represented by the average methane concentration percentage over a historical time period.

[0061] In this embodiment, dynamic calculation of the ventilation network and analysis of mine ventilation demand help to break through the limitations of traditional static ventilation analysis being disconnected from dynamic operating conditions. It accurately captures the real-time operating characteristics and potential risk points of the mine ventilation system, providing a comprehensive and quantitative basis for judging mine ventilation demand. This reduces safety hazards such as mine air volume supply and demand imbalance and gas accumulation caused by ambiguous ventilation status assessment, improves the timeliness and pertinence of ventilation demand analysis, and realizes the transformation from experience-based decision-making to data-driven precise decision-making. At the same time, it provides a clear direction and quantitative support for the formulation of subsequent mine ventilation adjustment strategies, ensuring that mine ventilation adjustment can accurately match changes in mine production conditions and safety ventilation needs.

[0062] Preferably, the specific process for mine ventilation regulation is as follows: The total mine ventilation volume is regulated, and the main fan frequency regulation is obtained. The specific process is as follows: Using the mine ventilation deviation value as the core input, the methane concentration ratio is used as a calibration factor and introduced into the PID controller. Weighted calculations are performed using the PID controller's preset proportional, integral, and derivative parameters, as well as ventilation facility operating status parameters, to finally obtain the main fan frequency regulation that adapts to the mine's total air volume regulation requirements. The main fan frequency regulation represents the adjustment value of the main fan power supply frequency calculated based on the PID algorithm and multiple dimensions such as the mine ventilation deviation value and methane concentration ratio, to bring the total mine air volume back to the preset safe air volume range. If the mine ventilation deviation value is greater than 0 (indicating that the mine air volume is too high), the main fan frequency regulation is adjusted accordingly. If the mine ventilation deviation value is too high and the mine ventilation volume needs to be reduced, a negative frequency conversion adjustment command is obtained based on the PID algorithm and sent to the main fan frequency conversion control cabinet. The main fan power supply frequency is dynamically reduced based on the main fan frequency adjustment amount. If the mine ventilation deviation value is less than 0 (indicating insufficient mine ventilation volume and the need to increase mine ventilation volume), a positive frequency conversion adjustment command is obtained based on the PID algorithm and sent to the main fan frequency conversion control cabinet. The main fan power supply frequency is dynamically increased based on the main fan frequency adjustment amount, thereby changing the main fan speed and achieving coarse adjustment of the total mine ventilation volume. The mine ventilation deviation value is represented by the sum of the ventilation deviation values ​​of all mine roadways. If the mine ventilation deviation value is equal to 0, the ventilation volume of the mine branch roadways is adjusted.

[0063] Specifically, the ventilation volume adjustment for mine branch roadways is carried out as follows: First, determine if the ventilation deviation value of the mine roadway is greater than 0. If so, mark the corresponding mine branch roadway as having excessive ventilation; otherwise, mark it as having insufficient ventilation. For both excessive and insufficient ventilation branch roadways, prioritize their adjustment based on an intelligent algorithm (such as particle swarm optimization) to obtain the adjustment sequence. The adjustment sequence includes both the excessive and insufficient ventilation branch roadway adjustment sequences. The branch roadway adjustment sequence represents a sequence in which each mine branch is arranged in descending order of mine roadway ventilation deviation values. Based on the adjustment sequence of branch roadways with excessive mine air volume, branch PID regulation is initiated sequentially for each mine branch roadway. The air volume adjustment amount of the roadway branch is obtained based on the PID algorithm. The specific acquisition process is as follows: the ventilation deviation value of the corresponding mine roadway is used as the input deviation of the PID controller. Combined with the preset safe air volume of the roadway, the real-time opening degree of the air door, and the proportion of gas concentration as calibration parameters, the air volume adjustment amount that can offset the ventilation deviation of the roadway is calculated through the P (proportional), I (integral), and D (derivative) parameters of the PID controller. The compensation value is used to ultimately obtain the branch roadway airflow adjustment amount. This branch roadway airflow adjustment amount represents the quantitative value calculated based on the PID algorithm to bring the actual airflow of the mine branch roadway back to the preset safe airflow range. It indicates the need to adjust the ventilation facilities (such as the opening of air windows and dampers) of the branch roadway. A window opening reduction adjustment command is generated and sent to the electric window control module. This reduces the excess airflow of the branch by increasing the roadway's air resistance. Based on the insufficient mine airflow branch roadway adjustment sequence, branch PID adjustment is sequentially initiated for each mine branch roadway. The branch roadway airflow adjustment amount is obtained based on the PID algorithm, generating... The ventilation window opening is increased by sending an adjustment command to the electric ventilation window control module. This increases the excess air volume of the branch by reducing the roadway resistance. After the mine ventilation adjustment is completed, the ventilation status assessment parameters are reacquired. If the ventilation status assessment parameters do not meet the mine ventilation qualification criteria, the mine ventilation adjustment continues. If the number of mine ventilation adjustments exceeds the preset maximum number of adjustments and the ventilation status assessment parameters still do not meet the mine ventilation qualification criteria, a ventilation adjustment failure prompt is sent. Otherwise, the current ventilation status is deemed qualified, and mine data is continuously monitored. The preset maximum number of adjustments is set in advance by the preset personnel.

[0064] In this embodiment, mine ventilation regulation first determines whether total mine ventilation volume needs adjustment to match the overall ventilation supply and demand balance based on dynamic calculations of the ventilation network and ventilation demand analysis. Then, it implements precise ventilation volume adjustment for specific ventilation deviations in branch roadways, forming a hierarchical regulation logic of first overall control and then local refinement. Simultaneously, it dynamically selects appropriate regulation technologies (such as frequency conversion adjustment of main fan power to achieve efficient control of total air volume, and remote adjustment of branch roadway air volume through intelligent ventilation windows) by combining ventilation quantification parameters and real-time mine monitoring data. This helps break away from the traditional one-size-fits-all, extensive approach to ventilation regulation. It ensures the stable operation of the overall mine ventilation system while addressing the issues of excessive or insufficient airflow in local roadways. This reduces energy waste or local safety risks caused by improper regulation, improves the accuracy, energy efficiency ratio, and response speed of ventilation regulation, and ensures that the ventilation system can meet the basic needs of overall mine safety production while adapting to the differentiated ventilation requirements of different areas (such as mining faces and return airways). This provides key support for mines to optimize ventilation energy consumption and improve production efficiency while ensuring safety.

[0065] like Figure 4 The diagram shown is a schematic of the data-driven intelligent ventilation decision control system for mines provided in this application embodiment. The data-driven intelligent ventilation system for mines includes: a mine signal sensing module, a mine information transmission module, a mine data platform monitoring module, and a mine ventilation linkage control module. The mine signal sensing module is used to collect mine data with preliminary quality self-diagnosis and screening decision-making capabilities by deploying various types of sensors (such as wind speed sensors, wind pressure sensors, gas concentration sensors, humidity sensors, etc.) at key nodes underground. Specifically, through optimized underground sensor layout, wind speed and wind pressure sensors, gas composition and differential pressure sensors are used to capture mine ventilation parameters and gas concentration information in real time. At the same time, relying on the ventilation facility status feedback system, fan parameter feedback system, and explosion-proof door status parameter feedback system, a comprehensive perception of the operating status of ventilation facilities, fans, and explosion-proof doors is achieved, providing multi-dimensional and high-quality raw mine monitoring data for subsequent data processing.

[0066] The mine information transmission module is used to collect raw mine physical signals and perform preliminary conditioning, analog-to-digital conversion, and local caching of the signals to form a mine monitoring data sequence with autonomous optimization decision-making capability for transmission paths. Specifically, a multi-link transmission architecture is constructed through explosion-proof switches, ventilation network parameter substations, and disaster information substations. Combined with transmission algorithms and transmission cables, the raw signals collected by the sensing module are transmitted stably, while also having the ability to autonomously optimize the transmission path, ensuring that the data is reliably and efficiently transmitted to the data platform in complex underground environments.

[0067] The mine data platform monitoring module encapsulates and executes the core data processing and analysis algorithms encompassed by the intelligent ventilation decision-making and control methods for mines. It executes decision-making methods including mine monitoring blind zone data analysis and decision-making, mine data multi-interference identification and decision-making, dynamic ventilation network calculation and decision-making, and mine ventilation demand analysis and decision-making, generating ventilation control commands with intelligent mine ventilation decision-making capabilities. Specifically, through the ventilation network online calculation module, the ventilation network parameter anomaly diagnosis and early warning module, and the disaster information anomaly diagnosis and location module, it achieves real-time calculation and anomaly diagnosis of the ventilation network. Simultaneously, with the help of the fan operation fault diagnosis and adjustment module and the disaster early warning and process evolution simulation module, it provides fault analysis and scenario simulation support for ventilation decision-making. Finally, through the intelligent decision-making and control command module, it outputs precise ventilation control commands.

[0068] The mine ventilation linkage control module receives control commands from the mine data platform monitoring module and generates equipment control signals with dynamic adaptive decision-making capabilities by driving actuators (such as the main fan frequency converter cabinet and electric air window controller). Specifically, through main fan frequency conversion control, remote control of airtight and local ventilation fans, and other actuators, combined with remote adjustment of air doors and windows and linkage control fault-tolerant algorithms, it precisely controls ventilation equipment such as main fans, air doors and windows. It also has dynamic adaptive decision-making capabilities, which can adjust the control strategy in real time according to changes in the actual working conditions underground, ensuring the stable and efficient operation of the ventilation system. In addition, this mine ventilation system also has a quick reset function for explosion-proof doors to realize ventilation decisions.

[0069] In this embodiment, the mine signal sensing module, mine information transmission module, mine data platform monitoring module, and mine ventilation linkage control module help to construct a fully intelligent mine ventilation management system that encompasses the entire process from mine data sensing and transmission to mine ventilation analysis and finally to mine ventilation decision execution. This system upgrades mine ventilation from traditional manual experience-based decision-making to data-driven intelligent decision-making, comprehensively improving the safety, efficiency, and adaptability of the mine ventilation system. Specifically, the mine signal sensing module provides multi-dimensional, high-quality raw mine monitoring data for mine ventilation decision-making, serving as the foundation for subsequent data processing and decision-making. The mine information transmission module ensures reliable and efficient data transmission in complex underground environments, providing data support for the algorithm execution of the data platform module. The mine data platform monitoring module performs in-depth data analysis and intelligent decision-making, generating precise ventilation control commands to guide the actions of the ventilation linkage control module. These modules influence each other and dynamically optimize, jointly promoting the development of the mine ventilation system towards a more intelligent and reliable direction, improving mine ventilation efficiency and safety assurance levels.

[0070] In summary, the data-driven intelligent ventilation decision-making and control method for mines provided by this invention analyzes mine monitoring blind zone data and determines whether to perform blind zone data completion based on the analysis results. If completion is performed, the method determines whether to perform multi-interference identification of mine data after completion. This helps solve the problems of existing technologies that rely primarily on sensor point monitoring, have small coverage areas, and poor data acquisition timeliness. Functional grid division achieves full-area monitoring coverage, and the combination of blind zone completion and validity verification forms a basis for multi-source data cross-validation, improving the completeness and timeliness of data acquisition. If completion is not performed, the method performs mine monitoring data validity verification and determines whether to perform multi-interference identification of mine data based on the verification results. If multi-interference identification is performed, the method determines whether to implement electromagnetic radiation interference suppression and environmental interference correction based on the multi-interference identification results. This helps solve the problem of data drift and distortion caused by sensor dampness and electromagnetic radiation. By replacing static filtering with targeted interference suppression and dynamic calibration mechanisms, the reliability of mine data is improved, providing high-quality mine data support for dynamic ventilation network calculation. If not, dynamic ventilation network calculation is performed. After the dynamic calculation, mine ventilation demand analysis is conducted. Based on the results of the mine ventilation demand analysis, it is determined whether to take mine ventilation regulation measures, including adjusting the total mine ventilation volume and the ventilation volume of mine branch roadways. If not, the current ventilation status is deemed qualified. If so, after the mine ventilation regulation is completed, it is determined whether the mine ventilation demand analysis is qualified. If not, a ventilation regulation failure prompt is sent. If qualified, the current ventilation status is deemed qualified. This helps to solve the problems of large prediction errors, insufficient flexibility of control strategies, and excessive or delayed air volume adjustment in existing technologies. By combining multi-source data such as gas concentration to achieve priority adjustment, the flexibility of control strategies and the accuracy of ventilation regulation are improved, effectively solving the problem of low accuracy of mine ventilation decision data in existing technologies.

[0071] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs) containing computer-usable program code. The form of a computer program product implemented on ROM, optical memory, etc.

[0072] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0073] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0074] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0075] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the scope and intent of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application is also intended to include such modifications and variations.

Claims

1. A data-driven intelligent ventilation decision-making and control method for mines, characterized in that, Includes the following steps: S1. Conduct mine monitoring blind zone data analysis, and determine whether to supplement mine blind zone data based on the results of the mine monitoring blind zone data analysis. If so, determine whether to identify multiple interferences in mine data after the mine blind zone data supplementation is completed. S2. If not, conduct a validity test of the mine monitoring data, and determine whether to conduct mine data multi-interference identification based on the results of the mine monitoring data validity test. If so, determine whether to take electromagnetic radiation interference suppression and environmental interference correction measures based on the results of the mine data multi-interference identification. S3. If not performed, dynamic calculation of the ventilation network will be performed. After the dynamic calculation of the ventilation network is completed, mine ventilation demand analysis will be performed. Based on the results of the mine ventilation demand analysis, it will be determined whether to take mine ventilation regulation measures, including the adjustment of the total mine ventilation volume and the adjustment of the mine branch roadway ventilation volume. If not taken, the current ventilation status will be deemed qualified. If taken, after the mine ventilation regulation is completed, it will be determined whether the mine ventilation demand analysis is qualified. If not qualified, a ventilation regulation failure prompt will be sent. If qualified, the current ventilation status will be deemed qualified.

2. The data-driven intelligent ventilation decision-making and control method for mines as described in claim 1, characterized in that, The specific process for analyzing the data from the mine monitoring blind spots is as follows: Based on functional requirements, the mine roadway is divided into several equal grid units, and the three-dimensional coordinates of the center point of each grid unit are obtained. The straight-line distance between the grid and the sensor is obtained based on the three-dimensional coordinates of the center point of each grid cell. The sensor signal strength deviation value is obtained, and it is determined whether the grid-sensor straight-line distance value is greater than the preset monitoring radius threshold and whether the sensor signal strength deviation value is less than the preset signal strength deviation threshold. If so, the corresponding grid unit is marked as a mine blind zone unit, and mine blind zone data is completed based on the mine blind zone unit. Otherwise, the corresponding grid unit is marked as a mine monitoring unit, and the validity of the mine monitoring data is verified based on the mine monitoring unit. The specific process for verifying the validity of the mine monitoring data is as follows: Obtain environmental parameters of the mine monitoring unit, including the distance value of the interference source to the mine monitoring unit and the humidity of the mine monitoring unit; Determine whether the environmental parameters of the mine monitoring unit meet the qualified conditions for mine data. If they do, mark the corresponding mine data as qualified mine data and perform dynamic calculation of the ventilation network based on the qualified mine data. Otherwise, identify multiple interferences in the mine data. The qualified conditions for mine data indicate that the distance value of the interference source of the mine monitoring unit is less than the preset safe distance threshold of the interference source, and the humidity of the mine monitoring unit is less than the preset maximum humidity threshold.

3. The data-driven based intelligent mine ventilation decision control method of claim 2, wherein, The specific process for completing the mine blind zone data is as follows: S101. Obtain mine reference data; The mine reference data represents a sequence of mine monitoring data from a preset number of preset type sensors that are closest to the mine blind zone unit and meet the blind zone data completion conditions. The blind zone data completion condition indicates that the mine data acquisition success rate is greater than the preset data acquisition success threshold, and the mine data fluctuation coefficient is less than the preset data fluctuation threshold. S102. Perform data fusion and completion of mine blind zone units to obtain mine blind zone unit fusion and completion data; The data fusion and completion of the mine blind zone unit refers to the process of obtaining the mine blind zone unit fusion and completion data by fusing roadway working condition related data based on the mine reference dataset. S103. Perform validity verification of the merged and completed data. The specific process is as follows: Based on the correlation analysis method, the correlation between the fused and completed data of the mine blind zone unit and the mine monitoring data of the mine monitoring unit that is closest to the mine blind zone unit is verified, and the correlation index of the mine blind zone data is obtained. Determine whether the correlation index of the mine blind area data is greater than the preset high validity threshold. If so, it means that the fused and completed data is effective, and multiple interference identification of the mine data is performed. Conversely, if the correlation index of the mine blind area data is less than the preset low validity threshold, a warning message indicating abnormal ventilation in the mine blind area will be sent and manual intervention will be carried out; otherwise, a message indicating invalid data completion will be sent.

4. The data-driven based intelligent mine ventilation decision control method of claim 3, wherein, The process for identifying multiple interferences in the mine data is as follows: Obtain electromagnetic radiation interference characteristic parameters, including wind speed signal distortion and fan harmonic intensity; Acquire environmental interference characteristic parameters, including sensor data drift and quantitative indicators of mine data fluctuation; Determine whether the characteristic parameters of electromagnetic radiation interference meet the electromagnetic radiation interference judgment conditions. If so, electromagnetic radiation interference suppression is adopted; otherwise, determine whether the characteristic parameters of environmental interference meet the environmental interference judgment conditions. If so, environmental interference correction is adopted; otherwise, wind network dynamic calculation is performed. The electromagnetic radiation interference determination condition indicates that the wind speed signal distortion is greater than the preset distortion threshold and the fan harmonic intensity is greater than the preset harmonic threshold. The environmental interference determination condition indicates that the sensor data drift is greater than the preset drift threshold, and the mine data fluctuation quantification index is greater than the preset interference fluctuation threshold.

5. The data-driven based intelligent mine ventilation decision control method of claim 4, wherein, The specific process for suppressing electromagnetic radiation interference is as follows: S201. Obtain the signal spectrum distribution in the mine; S202. Based on the mine signal spectrum distribution, obtain the discrete spectrum interference components of the mine data and generate a mine signal interference spectrum feature set. S203. Dynamically construct the corresponding digital band-stop filter bank based on the interference spectrum feature set of mine signals. The specific construction process is as follows: The center frequency of the interference component is used as the center stopband frequency of the corresponding digital bandstop filter, and the bandwidth of the interference component is multiplied by the preset bandwidth expansion factor as the stopband bandwidth of the corresponding digital bandstop filter. The mine data sequence is input into a digital band-stop filter bank for electromagnetic interference filtering of the mine data. S204. After the electromagnetic interference filtering of the mine data is completed, the signal-to-noise ratio of the mine data is obtained. Based on the signal-to-noise ratio of the mine data, the band-stop filtering effect is verified. The specific process is as follows: If the signal-to-noise ratio of the mine data is greater than the preset signal-to-noise ratio threshold, the corresponding mine monitoring data is marked as qualified mine data, and the ventilation network is dynamically calculated based on the qualified mine data; otherwise, environmental interference correction is adopted.

6. The data-driven based intelligent mine ventilation decision control method of claim 5, wherein, The specific process for correcting environmental interference is as follows: The environmental correction coefficients for mine sensors are obtained through the following process: environmental interference characteristic parameters and humidity of the mine monitoring unit are input into a preset environmental correction mapping set, and the corresponding environmental correction coefficients for the mine sensors are obtained by querying. Set the amplitude corresponding to the environmental correction coefficient of the mine sensor as the adjustment step size; Based on the adjustment step size, the sensor data correction amount is gradually adjusted in the opposite direction of the mine data drift, and this sensor data correction amount is superimposed on the original mine monitoring data; The validity value of the data correction is continuously monitored. When the validity value of the data correction is greater than the preset validity threshold, the corresponding mine monitoring data is marked as qualified mine data, and the ventilation network is dynamically calculated based on the qualified mine data. If the data correction validity value is not greater than the preset validity threshold, then environmental interference correction will continue. If the sensor data correction amount is greater than the preset maximum data correction threshold, and the data correction validity value is still not greater than the preset validity threshold, then an environmental correction processing failure prompt will be sent.

7. The data-driven based intelligent mine ventilation decision control method of claim 6, wherein, The specific process of dynamic calculation of the wind network is as follows: Input qualified mine data into the preset ventilation network solution model and output ventilation quantification parameters; The ventilation quantification parameters include the actual air volume, air pressure distribution data of each roadway in the mine, and the operating status parameters of the ventilation facilities. Mine ventilation demand analysis is performed based on the output ventilation quantification parameters.

8. The data-driven based intelligent mine ventilation decision control method of claim 7, wherein, The specific process of the mine ventilation demand analysis is as follows: Obtain ventilation status assessment parameters, including mine roadway ventilation deviation values ​​and methane concentration ratio values; The specific process for determining mine ventilation demand based on ventilation status assessment parameters is as follows: Determine whether the ventilation status assessment parameters meet the conditions for qualified mine ventilation. If they do, the current ventilation status is deemed qualified, and mine data is continuously monitored. Otherwise, mine ventilation is adjusted. The criteria for determining whether mine ventilation is qualified indicate that the absolute value of the ventilation deviation value in the mine roadway is less than the preset air volume adaptability threshold, and the proportion of methane concentration is greater than the preset methane concentration threshold.

9. The data-driven based intelligent mine ventilation decision control method of claim 8, wherein, The specific process of mine ventilation regulation is as follows: Adjust the total ventilation volume of the mine, input the mine ventilation deviation value into the PID controller, and use the gas concentration ratio value as a calibration factor. If the mine ventilation deviation value is greater than 0, a frequency converter negative adjustment command is obtained and sent to the main fan frequency converter control cabinet. If the mine ventilation deviation value is less than 0, a frequency converter positive adjustment command is obtained and sent to the main fan frequency converter control cabinet. The specific process for adjusting the ventilation volume in the mine's branch roadways is as follows: Determine whether the ventilation deviation value of the mine roadway is greater than 0. If it is, mark the corresponding mine branch roadway as a branch roadway with excessive mine air volume; otherwise, mark the corresponding mine branch roadway as a branch roadway with insufficient mine air volume. For branch roadways with excessive or insufficient ventilation in the mine, the corresponding mine branch adjustment priority is sorted to obtain the mine branch roadway adjustment sequence. The mine branch roadway adjustment sequence includes the mine ventilation volume exceeding the limit branch roadway adjustment sequence and the mine ventilation volume insufficient branch roadway adjustment sequence; Based on the adjustment sequence of the branch roadway with excessive mine air volume, the branch PID adjustment is started in each mine branch roadway in sequence to obtain the roadway branch air volume adjustment amount, generate the window opening reduction adjustment command, and send the window opening reduction adjustment command to the electric window control module. Based on the adjustment sequence of the branch roadway with insufficient mine air volume, the branch PID adjustment is started in each mine branch roadway in sequence to obtain the roadway branch air volume adjustment amount, generate the window opening increase adjustment command, and send the window opening increase adjustment command to the electric window control module. After the mine ventilation adjustment is completed, the ventilation status assessment parameters are reacquired. If the ventilation status assessment parameters do not meet the conditions for qualified mine ventilation, the mine ventilation adjustment will continue. If the number of times the mine ventilation adjustment is executed is greater than the preset maximum number of times and the ventilation status assessment parameters still do not meet the conditions for qualified mine ventilation, a ventilation adjustment failure prompt will be sent. Otherwise, the current ventilation status will be determined to be qualified, and the mine data will be continuously monitored.

10. A data-driven based intelligent ventilation system for mines applying the data-driven based intelligent ventilation decision control method of any one of claims 1-9, characterized in that, A data-driven intelligent ventilation system for mines includes: a mine signal sensing module, a mine information transmission module, a mine data platform monitoring module, and a mine ventilation linkage control module. The mine signal sensing module is used to collect mine data with preliminary quality self-diagnosis and screening decision-making capabilities by deploying multiple types of sensors at key nodes underground. The mine information transmission module is used to collect raw mine physical signals, and to perform preliminary conditioning, analog-to-digital conversion and local caching of the signals to form a mine monitoring data sequence with autonomous optimization decision-making capability for transmission paths; The mine data platform monitoring module is used to encapsulate and execute the core data processing and analysis algorithms covered by the intelligent ventilation decision control method for mines. It executes decision-making methods including mine monitoring blind zone data analysis decision, mine data multi-interference identification decision, ventilation network dynamic calculation decision, and mine ventilation demand analysis decision, and generates ventilation control instructions with intelligent mine ventilation decision-making capabilities. The mine ventilation linkage control module is used to receive control instructions from the mine data platform monitoring module and generate equipment control signals with dynamic adaptive decision-making capabilities by driving the actuator.

Images