Intelligent dust suppression system for metal mine tunneling roadway

Abstract

The present application relates to the technical field of mine dust suppression, and particularly relates to an intelligent dust suppression system for a metal mine tunneling roadway, comprising: a ventilation module, which is used for providing circulating ventilation for the metal mine tunneling roadway, and introducing dust in the metal mine tunneling roadway into a circulating air loop for purification treatment; a spraying module, which is used for spraying water mist to the dust in the metal mine tunneling roadway; a detection module, which is used for detecting dust concentrations in different regions in the metal mine tunneling roadway in real time as first judgment data; and a control module, which is used for analyzing differences in dust concentrations and concentration levels in different regions in the metal mine tunneling roadway based on the first judgment data, generating corresponding differential control instructions for the spraying module, and controlling the spraying module to adjust corresponding spraying amounts in each region. The present application accurately captures global dust concentration differences through the detection module, and outputs differential spraying instructions through the control module, thereby reducing water resource waste and roadway water accumulation caused by excessive spraying in some regions.

Description

Technical Field

[0001] This invention relates to the field of dust suppression technology in mining, specifically to an intelligent dust suppression system for metal mine tunneling roadways. Background Technology

[0002] Metal ores, as important raw materials for industrial production, are primarily mined underground. Excavation tunnels are the core working spaces in underground metal mining, mainly used for personnel passage, equipment transport, and ventilation. They are characterized by their long, narrow shape, single-ended design, limited ventilation, and complex working environment. Dust generation is inevitable during the construction of metal mine tunnels. This dust contains a large number of respirable fine particles (≤5μm), and because metal ore dust has poor hydrophilicity, is hard, and dry, it is more likely to harm the health of workers and cause occupational diseases.

[0003] Currently, the mainstream dust suppression technology for metal mine tunneling roadways mainly employs a circulating ventilation dust suppression system, or a combination of this system and a spray dust suppression device. The core idea is to capture and purify dust through a circulating air loop, while using a spray device to wet and settle the dust. The start-up, shutdown, and parameter adjustments of most actuators are controlled based on data from a dust concentration sensor within the metal mine tunneling roadway, or by relying directly on human experience. However, the dust is unevenly dispersed in the tunnel after it is generated and diffused: on the one hand, the dust generation source is obviously concentrated. During drilling and blasting operations, a high concentration of dust will be generated instantaneously at the face. During mechanical cutting operations, the cutting head will continuously generate dust. Moreover, the particle size of metal ore dust varies greatly. Coarse dust particles settle quickly under the action of gravity, while fine respirable dust particles are suspended and diffused with the airflow, naturally forming concentration differences in different areas. On the other hand, metal mine tunnels are mostly narrow and long single-ended structures. When forced ventilation is used, the airflow pushes the dust in one direction, causing the dust to show a gradient distribution along the tunnel direction, with the concentration at the face being much higher than that in the rear area. As a result, some areas with low dust concentrations are still continuously sprayed, causing excessive spraying, which not only wastes water resources but also makes the tunnel floor muddy. On the other hand, for areas with high initial dust concentrations, such as the face of the tunneling working face and the eddy accumulation zone, it is easy to fail to effectively capture high-concentration dust clouds and fine dust particles due to insufficient spray supply and short spray duration. The dust suppression effect is poor, and the dust is easy to spread along the tunnel to the rear, making it difficult to achieve precise and efficient dust suppression throughout the entire area.

[0004] Therefore, this invention proposes an intelligent dust suppression system for metal mine tunneling roadways to solve the above-mentioned problems. Summary of the Invention

[0005] To address the aforementioned issues, this invention provides an intelligent dust suppression system for metal mine tunneling roadways. This system captures the differences in dust distribution within the tunnels and uses differentiated spraying and targeted ventilation to reduce excessive spraying and dust suppression blind spots, thereby achieving precise control of dust across the entire area and minimizing the health hazards of dust to workers.

[0006] To achieve the above objectives, the technical solution of the present invention is as follows: an intelligent dust suppression system for metal mine tunneling roadways, comprising:

[0007] The ventilation module is used to provide circulating ventilation for metal mine tunnels, introducing dust in the metal mine tunnels into the circulating air circuit for purification.

[0008] The spray module is used to spray water mist onto dust in the tunnels of metal mines;

[0009] The detection module is used to detect the dust concentration in different areas of the metal mine tunnel in real time, as the primary judgment data;

[0010] The control module is used to analyze the differences and concentration levels of dust concentration in different areas of the metal mine tunnel based on the first judgment data, generate corresponding differentiated control commands for the spray module, and control the spray module to adjust the spray volume for each area.

[0011] Furthermore, the detection module acquires hyperspectral images of the metal mine tunnel in real time, preprocesses and extracts feature spectra from the hyperspectral images, and then combines them with a preset model of the correspondence between the spectrum and concentration of metal ore dust to obtain the dust concentration value corresponding to each pixel region in the hyperspectral image, which is defined as the first judgment data.

[0012] Furthermore, in the control module, based on the metal ore dust concentration classification strategy, the first judgment data is divided into regions using an image segmentation algorithm, resulting in several diffusion regions of different levels. The diffusion regions are classified into three levels: high concentration region, medium concentration region, and low concentration region. Then, based on the level of the diffusion region and the coordinate position of the diffusion region in the first judgment data, differentiated control commands are generated.

[0013] Furthermore, the differentiated control instructions include: firstly, setting the basic spraying parameters corresponding to the low-concentration area, including the basic spraying time and basic spraying flow rate, which serve as the benchmark for the spraying parameters of all dust concentration level areas;

[0014] For medium concentration areas: increase the spray flow rate by 30% to 50% based on the basic spray flow rate, and extend the spray time by 20% to 40% based on the basic spray time;

[0015] High-concentration areas: Increase the spray flow rate by 60% to 100% based on the basic spray flow rate, and extend the spray time by 50% to 80% based on the basic spray time.

[0016] Furthermore, the differentiated control instructions also include: when the dust concentration in a certain concentration level area is detected to be lower than the preset stop threshold and the preset duration is met, a corresponding stop spray instruction is generated; and before the stop spray instruction is triggered, the spray parameters of each diffusion area are adjusted step by step according to the dust concentration data fed back by the detection module in real time.

[0017] Furthermore, in the control module, the first judgment data is pre-divided into several sub-positioning units with unique pixel coordinates and position numbers according to the actual spatial size and coordinate rules of the metal mine tunneling roadway, so that each sub-positioning unit corresponds to a fixed actual physical area in the metal mine tunneling roadway; each diffusion area is composed of several adjacent and continuously positioned sub-positioning units.

[0018] In the spray module, the first judgment data is used as the matching benchmark. The entire area of ​​the metal mine tunnel is divided into several independent sub-spraying zones according to each sub-positioning unit. Each sub-spraying zone corresponds to a sub-positioning unit, and each sub-spraying zone is independently configured with a spraying unit. Each spraying unit independently receives control commands issued by the control module.

[0019] Furthermore, the control module calculates the area ratio of low-concentration areas within the detection range of the metal mine tunneling roadway covered by the first judgment data. When the area ratio of low-concentration areas is greater than or equal to the preset area threshold, the dynamic tracking and monitoring process of metal mine dust is initiated.

[0020] The dynamic tracking and monitoring process for metal ore dust is as follows: Identify the metal ore dust accumulation areas that are distributed in clumps in the first judgment data and set them as suspected areas; continuously detect the boundary features and concentration features of the suspected areas within a preset judgment time; if the amplitude of the boundary features of the suspected areas is less than or equal to the preset amplitude value, and the downward trend of the concentration features of the suspected areas does not meet the preset threshold for reducing the concentration of metal ore dust, then the suspected areas are judged to be dead zones.

[0021] Furthermore, the ventilation module includes a movable dust collection unit. When the control module determines the dead corner area in the first judgment data, it outputs the corresponding movement control command based on the coordinate information of the sub-positioning unit corresponding to the dead corner area. The movable dust collection unit moves to the spatial position of the corresponding dead corner area according to the movement control command, performs directional suction and collection of the metal ore dust accumulated in the dead corner area, and sends the collected metal ore dust into the circulating air circuit to complete the purification treatment.

[0022] Furthermore, before extracting features from the hyperspectral image, the detection module performs background interference removal processing: a standard background spectral database of the walls and bottom surfaces of metal mine tunnels is pre-built and stored. The real-time acquired hyperspectral image is compared with the standard background spectral database to remove background pixel areas in the hyperspectral image that are consistent with the spectral features of the walls and bottom surfaces. Then, background masking processing is applied to the hyperspectral image to shield the background area of ​​non-metallic mineral dust, retaining only the effective spectral feature area corresponding to the metallic mineral dust target, and then performing subsequent feature spectral extraction, dust concentration identification, and region division.

[0023] Furthermore, after the detection module completes the background removal processing of the metal mine tunnel in the hyperspectral image, it further separates the water mist spectrum during the dust suppression process, as follows: a standard spectral feature library of water mist with different particle sizes and concentrations is pre-established; the spectral data of each pixel in the real-time hyperspectral image is matched with the water mist spectral feature library to identify and mark the water mist interference area in the hyperspectral image; a band-stop spectral filtering algorithm is used to remove the water mist spectral signal in the water mist interference area, while spectral compensation is used to restore the true spectral characteristics of the metal mine dust in the area blocked by water mist; then, the processed image is corrected for light intensity and concentration to obtain the first judgment data.

[0024] The above approach has the following beneficial effects:

[0025] 1. This solution utilizes hyperspectral image acquisition and analysis by the detection module, combined with a gridded sub-positioning unit, to achieve precise detection and differentiated control of dust concentration. Compared to traditional single-sensor detection and uniform spraying across the entire area, this solution captures the differences in metal ore dust distribution in different regions and matches corresponding spraying parameters based on concentration levels. This reduces water waste caused by excessive spraying in low-concentration areas while ensuring sufficient spraying volume in high-concentration areas. Furthermore, background spectral removal and water mist interference separation further improve the accuracy of metal ore concentration detection and ensure the adaptability of spraying parameters to the actual dust distribution.

[0026] 2. This solution identifies dust accumulation areas without diffusion trends through continuous frame image analysis. These dust accumulation areas are designated as dead zones, and then targeted treatment is carried out by a movable dust collection unit. Each sub-positioning unit corresponds to a spray unit, ensuring precise coverage even if dust only accumulates in dead zones through dedicated control commands. This reduces secondary pollution caused by dust spreading with airflow in dead zones and minimizes instability in dust suppression effects due to dust accumulation in dead zones, further ensuring the safety and stability of the working environment.

[0027] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of an embodiment of the intelligent dust suppression system for metal mine tunneling roadways according to the present invention. Detailed Implementation

[0029] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0030] The following detailed description illustrates the specific implementation method:

[0031] Example 1:

[0032] like Figure 1 As shown, an intelligent dust suppression system for metal mine tunneling includes a ventilation module, a spray module, a detection module, and a control module.

[0033] The ventilation module provides circulating ventilation for the metal mine tunneling roadway, introducing dust from the tunneling roadway into the circulating air loop for purification. Specifically, the ventilation module includes a main ventilation unit and a circulating air purification unit. The main ventilation unit preferably uses an FBDNo.6.3 / 2×15 type mine explosion-proof ventilator, located 10m outside the tunnel entrance. Fresh air is forced into the tunneling roadway through a flame-retardant ventilation duct (e.g., Φ800mm), with a ventilation volume set at 200-220m³ / min, pushing the dust in the tunneling roadway towards the return air end. The return air end is equipped with a return air duct (Φ600mm) laid along the side wall of the tunneling roadway, and a grid-type dust interception net is installed at the inlet of the return air duct to initially intercept large particles of metal mine dust (particle size ≥10μm). The circulating air purification unit is connected in series in the middle of the return air duct. It adopts a dual purification structure of bag filter and activated carbon adsorption. The bag filter is made of PPS high-temperature resistant dust collector bag (which can effectively filter respirable fine particulate dust with a particle size ≤5μm), while the activated carbon adsorption layer is used to adsorb harmful impurities in the dust. The purified clean air is re-pressurized into the roadway by the main ventilation fan, forming a closed-loop ventilation circuit of pressurization-return air-purification-recirculation. At the same time, the spray module sprays water mist into the dust in the metal mine tunnel, which, together with the ventilation module, further improves the dust suppression effect.

[0034] Traditional single-point dust sensors can only detect local areas and are difficult to identify differences in dust concentration distribution throughout the tunnel, which can easily lead to poor dust suppression effects (for example, some areas with low dust concentration are still continuously sprayed, resulting in excessive spraying, which not only wastes water resources but also causes the tunnel floor to become muddy; at the same time, for areas with high initial dust concentration, such as the face of the tunneling working face and the eddy accumulation zone, it is easy to fail to effectively capture high-concentration dust clouds and fine particulate dust due to insufficient spray supply and short spray duration). In this scheme, a detection module acquires hyperspectral images of the metal mine tunnel in real time (metal dust is composed of various mineral particles and has characteristic spectral absorption and reflection responses in the visible to near-infrared bands. The spectral signals of dust with different concentrations and particle sizes are significantly different. Hyperspectral imaging technology is used to achieve full-area non-contact signal acquisition, which is not limited by single-point detection location, so as to reflect the true distribution of metal dust in various areas of the metal mine tunnel). The hyperspectral images are preprocessed and feature spectra are extracted. Then, combined with a preset metal dust spectrum and concentration correspondence model, the dust concentration value corresponding to each pixel area in the hyperspectral image is obtained and defined as the first judgment data. The preset model for the correspondence between the spectrum and concentration of metal ore dust was established using existing laboratory calibration and data fitting methods: metal ore dust samples consistent with those on-site were selected, and multiple dust environments with gradient concentrations (covering the actual low, medium, and high concentration ranges on-site) were configured in a standard detection dark box. Standard spectral data of dust at each concentration were collected using the same model of hyperspectral imaging equipment, and key spectral features such as reflectance and absorption peak intensity of characteristic bands were extracted. Using the measured values ​​of the dust concentration detection instrument as the calibration benchmark, a mapping relationship model between spectral feature parameters and dust concentration values ​​was established through multiple linear regression and neural network fitting algorithms. After verification and error correction by multiple sets of samples, the model was pre-stored in the detection module for rapid inversion of dust concentration from real-time hyperspectral images.

[0035] In actual tunneling environments, hyperspectral image acquisition inevitably includes non-dust background areas such as tunnel walls and floors. The materials of these background areas (e.g., concrete walls, gravel floors) also possess their own spectral characteristics in the visible to near-infrared bands, easily overlapping and interfering with the spectral characteristics of metal ore dust. Directly extracting features from the original hyperspectral image can lead to distortion in dust feature recognition, increased concentration inversion errors, and even misclassifying background areas as dust areas, affecting the accuracy of subsequent diffusion area delineation. This, in turn, results in a mismatch between the commands generated by the control module and the actual dust distribution, making it difficult to achieve the desired effect. To achieve precise dust suppression, the detection module performs background interference removal before feature extraction from the hyperspectral image: a standard background spectral database of the walls and bottom of the metal mine tunnel is pre-built and stored; the real-time acquired hyperspectral image is compared with the standard background spectral database to remove background pixel areas in the hyperspectral image that are consistent with the spectral features of the walls and bottom; then, background masking is applied to the hyperspectral image to shield the background area of ​​non-metallic mineral dust, retaining only the effective spectral feature area corresponding to the metallic mineral dust target, before proceeding with subsequent feature spectral extraction, dust concentration identification, and region division. Furthermore, considering the water mist generated during the spraying process, the spectral characteristics of the water mist partially overlap with those of the metal ore dust. Without separation, this can lead to deviations in the metal ore dust concentration detection, thus affecting the accuracy of subsequent control commands. Therefore, after the detection module completes background removal processing of the metal mine tunneling roadway in the hyperspectral image, it further separates the water mist spectrum during the dust suppression process (the hyperspectral image acquired after spraying). Specifically: a standard spectral feature library of water mist with different particle sizes and concentrations is pre-established; the spectral data of each pixel in the real-time hyperspectral image is matched with the water mist spectral feature library to identify and mark the water mist interference areas in the hyperspectral image; a band-stop spectral filtering algorithm is used to remove the water mist spectral signal within the interference areas, while spectral compensation restores the true spectral characteristics of the metal ore dust in the water mist-covered areas; finally, the processed image is corrected for light intensity and concentration to obtain the first judgment data.

[0036] In the control module, based on the metal ore dust concentration classification strategy, an image segmentation algorithm is used to divide the first judgment data into regions, resulting in several diffusion regions of different levels. These diffusion regions are categorized into three levels: high concentration, medium concentration, and low concentration. Then, based on the diffusion region level and its coordinate position in the first judgment data, differentiated control commands are generated. These differentiated control commands include the following:

[0037] First, set the basic spraying parameters for the low-concentration area. The basic spraying parameters include the basic spraying time and the basic spraying flow rate, which serve as the benchmark for the spraying parameters of all dust concentration levels.

[0038] For medium concentration areas: increase the spray flow rate by 30% to 50% based on the basic spray flow rate, and extend the spray time by 20% to 40% based on the basic spray time;

[0039] High-concentration areas: Increase the spray flow rate by 60% to 100% based on the basic spray flow rate, and extend the spray time by 50% to 80% based on the basic spray time;

[0040] Meanwhile, when the dust concentration in a certain concentration level area is detected to be lower than the preset stop threshold and the preset duration is met, a corresponding stop spray command is generated; and before the stop spray command is triggered, the spray parameters of each diffusion area are adjusted step by step according to the dust concentration data fed back by the detection module in real time.

[0041] In the control module, the first judgment data is pre-divided into several sub-positioning units with unique pixel coordinates and position numbers according to the actual spatial size and coordinate rules of the metal mine tunnel. Each sub-positioning unit corresponds to a fixed actual physical area in the metal mine tunnel. Each diffusion area is composed of several adjacent and continuously positioned sub-positioning units.

[0042] In the spray module, the first judgment data is used as the matching benchmark. The entire area of ​​the metal mine tunnel is divided into several independent sub-spraying zones according to each sub-positioning unit. Each sub-spraying zone corresponds to a sub-positioning unit, and each sub-spraying zone is independently configured with a spraying unit. Each spraying unit independently receives control commands issued by the control module.

[0043] Example 2:

[0044] The difference from Example 1 is that, in actual tunneling operations, airflow vortices easily form in areas such as tunnel corners, behind equipment, around ventilation duct outlets, and gaps in supports. Dust in these areas tends to accumulate into fixed clumps, making it difficult to be effectively captured by the main ventilation circuit. Furthermore, conventional differentiated spraying cannot accurately cover these areas, easily leading to long-term accumulation of metal ore dust, which may even be re-entrained by slight airflow disturbances, affecting the overall dust suppression effect. Therefore, such as Figure 1As shown, the control module calculates the area ratio of low-concentration areas within the detection range of the metal mine tunnel covered by the first judgment data. When the area ratio of low-concentration areas is greater than or equal to a preset area threshold, the dynamic tracking and monitoring process for metal ore dust is initiated. The dynamic tracking and monitoring process for metal ore dust is as follows: Metal ore dust accumulation areas distributed in clumps in the first judgment data are identified and designated as suspected areas; the boundary features and concentration features of the suspected areas are continuously detected within a preset judgment time; if the amplitude of the boundary features of the suspected areas is less than or equal to a preset amplitude value (i.e., there is no obvious diffusion or displacement at the boundary), and the decreasing trend of the concentration features of the suspected areas does not meet the preset metal ore dust concentration reduction threshold, then the suspected areas are determined to be dead zones, ensuring the accuracy of identification.

[0045] Meanwhile, in this solution, the ventilation module also includes a movable dust collection unit to address the issue of dust accumulation in dead-end areas. Once the control module identifies a dead-end area in the initial judgment data, it outputs corresponding movement control commands based on the coordinates of the sub-positioning unit corresponding to the dead-end area. The movable dust collection unit then moves to the corresponding spatial position in the dead-end area according to the movement control commands, performing directional suction and collection of the accumulated metal ore dust. The collected metal ore dust is then sent into the circulating air loop for purification, clearing the dust accumulation in the dead-end area, further improving the dust suppression effect, and ensuring effective dust control throughout the entire tunnel.

[0046] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. An intelligent dust suppression system for metal mine tunneling roadways, characterized in that, include: The ventilation module is used to provide circulating ventilation for metal mine tunnels, introducing dust in the metal mine tunnels into the circulating air circuit for purification. The spray module is used to spray water mist onto dust in the tunnels of metal mines; The detection module is used to detect the dust concentration in different areas of the metal mine tunnel in real time, as the primary judgment data; The control module is used to analyze the differences and concentration levels of dust concentration in different areas of the metal mine tunnel based on the first judgment data, generate corresponding differentiated control commands for the spray module, and control the spray module to adjust the spray volume for each area.

2. The intelligent dust suppression system for metal mine tunneling roadways according to claim 1, characterized in that, The detection module acquires hyperspectral images of the tunneling roadway in the metal mine in real time, performs preprocessing and feature spectrum extraction on the hyperspectral images, and then combines them with a preset model of the correspondence between the spectrum and concentration of metal mine dust to obtain the dust concentration value corresponding to each pixel region in the hyperspectral image, which is defined as the first judgment data.

3. The intelligent dust suppression system for metal mine tunneling roadways according to claim 2, characterized in that, In the control module, based on the metal ore dust concentration classification strategy, the first judgment data is divided into regions using an image segmentation algorithm, and several diffusion regions of different levels are obtained from the first judgment data. The diffusion area is classified into three levels: high concentration area, medium concentration area, and low concentration area. Based on the level of the diffusion area and the coordinate position of the diffusion area in the first judgment data, differentiated control instructions are generated.

4. The intelligent dust suppression system for metal mine tunneling roadways according to claim 3, characterized in that, The differentiated control instructions include: first, setting the basic spraying parameters corresponding to the low concentration area, including the basic spraying time and basic spraying flow rate, which serve as the benchmark for the spraying parameters of all dust concentration level areas; For medium concentration areas: increase the spray flow rate by 30% to 50% based on the basic spray flow rate, and extend the spray time by 20% to 40% based on the basic spray time; High-concentration areas: Increase the spray flow rate by 60% to 100% based on the basic spray flow rate, and extend the spray time by 50% to 80% based on the basic spray time.

5. The intelligent dust suppression system for metal mine tunneling roadways according to claim 4, characterized in that, The differentiated control instructions also include: when the dust concentration in a certain concentration level area is detected to be lower than the preset stop threshold and the preset duration is met, a corresponding stop spray instruction is generated; and before the stop spray instruction is triggered, the spray parameters of each diffusion area are adjusted step by step according to the dust concentration data fed back by the detection module in real time.

6. The intelligent dust suppression system for metal mine tunneling roadways according to claim 5, characterized in that, In the control module, the first judgment data is pre-divided into several sub-positioning units with unique pixel coordinates and position numbers according to the actual spatial size and coordinate rules of the metal mine tunnel. Each sub-positioning unit corresponds to a fixed actual physical area in the metal mine tunnel. Each diffusion area is composed of several adjacent and continuously positioned sub-positioning units. In the spray module, the first judgment data is used as the matching benchmark. The entire area of ​​the metal mine tunnel is divided into several independent sub-spraying zones according to each sub-positioning unit. Each sub-spraying zone corresponds to a sub-positioning unit, and each sub-spraying zone is independently configured with a spraying unit. Each spraying unit independently receives control commands issued by the control module.

7. The intelligent dust suppression system for metal mine tunneling roadways according to claim 6, characterized in that, The control module calculates the area ratio of low-concentration areas within the detection range of the metal mine tunnel covered by the first judgment data. When the area ratio of low-concentration areas is greater than or equal to the preset area threshold, the dynamic tracking and monitoring process of metal mine dust is initiated. The dynamic tracking and monitoring process for metal ore dust is as follows: Identify the metal ore dust accumulation areas that are distributed in clumps in the first judgment data and set them as suspected areas; continuously detect the boundary features and concentration features of the suspected areas within a preset judgment time. If the boundary feature amplitude of the suspected area is less than or equal to the preset amplitude value, and the decreasing trend of the concentration feature of the suspected area does not meet the preset threshold for reducing the concentration of metal ore dust, then the suspected area is judged to be a dead zone area.

8. The intelligent dust suppression system for metal mine tunneling roadways according to claim 7, characterized in that, The ventilation module includes a movable dust collection unit. When the control module determines the dead corner area in the first judgment data, it outputs the corresponding movement control command according to the coordinate information of the sub-positioning unit corresponding to the dead corner area. The movable dust collection unit moves to the spatial position of the corresponding dead corner area according to the movement control command, performs directional suction and collection of the metal ore dust accumulated in the dead corner area, and sends the collected metal ore dust into the circulating air circuit to complete the purification treatment.

9. The intelligent dust suppression system for metal mine tunneling roadways according to claim 8, characterized in that, Before performing feature extraction on the hyperspectral image, the detection module performs background interference removal processing: a standard background spectral database of the walls and bottom surfaces of the metal mine tunnel is pre-built and stored, and the hyperspectral images acquired in real time are compared with the standard background spectral database to remove background pixel areas in the hyperspectral images that are consistent with the spectral features of the walls and bottom surfaces. Then, a background masking process is applied to the hyperspectral image to shield the background area of ​​non-metallic mineral dust, retaining only the effective spectral feature area corresponding to the metallic mineral dust target, and then performing subsequent feature spectrum extraction, dust concentration identification, and region segmentation.

10. The intelligent dust suppression system for metal mine tunneling roadways according to claim 9, characterized in that, After the detection module performs background removal processing on the hyperspectral image of the metal mine tunnel, it further separates the water mist spectrum during the dust suppression process. The process is as follows: A standard spectral feature library of water mist with different particle sizes and concentrations is pre-established. The spectral data of each pixel in the real-time hyperspectral image is matched with the water mist spectral feature library to identify and mark the water mist interference area in the hyperspectral image. The band-stop spectral filtering algorithm is used to remove the water mist spectral signal in the water mist interference area. At the same time, the true spectral features of the metal ore dust in the water mist-covered area are restored through spectral compensation. Then, the light intensity and concentration are corrected for the processed image to obtain the first judgment data.

Images