Roadway dust concentration monitoring system and monitoring method

Abstract

The invention relates to the technical field of dust concentration monitoring, and discloses a roadway dust concentration monitoring system and method, and the system comprises a multispectral light field induction module which is used for constructing a detection field; the synchronous control and data acquisition module is used for obtaining projection data; the chromatography reconstruction and concentration field resolving module is used for obtaining a dust mass concentration distribution diagram and a water mist mass concentration distribution diagram; the time sequence analysis and diffusion prediction module is used for predicting the future spatial position of the high-concentration dust area; the data processing and output module is used for generating a cooperative control instruction; the method comprises the steps of constructing a detection field, and obtaining projection data; based on projection, dust water mist concentration is separated; predicting the dust position; high concentration is recognized, and a control instruction is generated. According to the invention, light intensity attenuation data of two specific wavelengths are obtained, and a mathematical equation set based on the difference of unit mass extinction coefficients of dust and water mist is solved by the chromatography reconstruction and concentration field solving module, so that the interference of water mist on dust concentration measurement is effectively eliminated.

Description

Technical Field

[0001] This invention relates to the field of dust concentration monitoring technology, specifically to a dust concentration monitoring system and method for roadways. Background Technology

[0002] Dust pollution is a long-standing problem affecting production safety and occupational health in underground working environments such as mines and tunnels. High concentrations of dust not only cause serious damage to the respiratory system of workers, inducing irreversible occupational diseases such as pneumoconiosis, but also pose a threat to the safety of facilities and the lives and property of personnel in the tunnels due to their potential explosion risk.

[0003] Currently, in the field of dust concentration monitoring in tunnels, traditional technologies mainly include sensors based on the principles of light scattering and light absorption. These devices are typically fixedly installed at specific monitoring points within the tunnel, or used for localized sampling and detection via handheld devices. Their operating mode is usually to acquire the average concentration value along an optical path at a specific point or cross-section and display this data to on-site management personnel in real time. When dust concentration exceeds the standard, the system triggers an audible and visual alarm to alert workers to safety precautions.

[0004] Existing roadway dust monitoring technologies often involve dust suppression spraying operations within the roadway. The water mist in the air significantly interferes with light signals, making it difficult for sensors to accurately distinguish between dust and water mist. This leads to distorted monitoring data and affects the reliability of decision-making. Most existing monitoring devices can only provide discrete single-point or limited-area concentration information, failing to comprehensively represent the spatial distribution of dust across the entire roadway cross-section. They also have blind spots for suddenly formed high-concentration dust clumps, making it difficult to comprehensively assess the pollution situation. Furthermore, they lack the ability to capture dynamic changes in dust airflow within the roadway, making it difficult to predict the direction, speed, and future location of dust diffusion. Therefore, this invention provides a roadway dust concentration monitoring system and method to address the shortcomings of existing technologies. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a roadway dust concentration monitoring system and method, which solves the problems of insufficient accuracy under complex working conditions, difficulty in effectively distinguishing dust from water mist, lack of dynamic tracking and forward-looking prediction capabilities, and difficulty in achieving coordinated control with intelligent dust suppression systems.

[0006] To achieve the above objectives, the present invention provides a roadway dust concentration monitoring system, comprising: A multispectral optical field sensing module is used to construct a detection field consisting of light beams containing multiple specific spectral components on the cross-section of the roadway to be monitored. The synchronous control and data acquisition module is used to drive and control the multispectral light field sensing module and acquire the light intensity attenuation data of the light beam after passing through the cross section of the alley to obtain projection data to represent the integral attenuation. The tomographic reconstruction and concentration field calculation module is used to reconstruct a spatial distribution map of the attenuation coefficient representing the tunnel cross section based on the projection data, and to separate the spatial distribution map of the attenuation coefficient into a quantitative dust mass concentration distribution map and a water mist mass concentration distribution map by utilizing the spectral absorption characteristics of different substances. The time series analysis and diffusion prediction module is used to calculate a vector field describing the dust motion state by analyzing the dust mass concentration distribution map, and predict the future spatial location of the high-concentration dust area based on the vector field. The data processing and output module is used to generate collaborative control commands based on the current location of the high-concentration dust area, the future spatial location of the high-concentration dust area, and the preset light intensity.

[0007] Preferably, the multispectral light field sensing module includes: The transmitting unit includes multiple multi-wavelength light source modules, each of which is used to transmit light at a wavelength of [wavelength value missing]. and A light beam, the wavelength of which is and The beam has the following characteristics: the target object and potential optical interference objects are at the same wavelength. and The extinction coefficient per unit mass varies; The receiving unit, comprising multiple photodetector arrays, is positioned opposite the transmitting unit and is used to receive the wavelength. and The beam of light.

[0008] Preferably, the synchronization control and data acquisition module includes: A signal modulation unit is used to assign a unique pseudo-random code sequence to each laser transmitter and to modulate the drive current of the laser transmitter; The synchronous acquisition and demodulation unit is used to receive the mixed signal output by the receiving unit of the multispectral light field sensing module after converting the superimposed light beam into an electrical signal, and to demodulate the light intensity component belonging to a specific optical path by performing correlation operation between the mixed signal and the pseudo-random code sequence. The data processing unit is used to calculate the projection data based on the light intensity components and the pre-calibrated reference light intensity.

[0009] Preferably, the chromatography reconstruction and concentration field calculation module is specifically used for: The cross-section of the roadway to be monitored is discretized into a two-dimensional pixel grid; Algebraic reconstruction technology is used to iteratively calculate the geometric relationship between the projection data calculated by the synchronous control and data acquisition module and the two-dimensional pixel grid to obtain the spatial distribution map of the total linear attenuation coefficient corresponding to the two wavelengths.

[0010] Preferably, the chromatography reconstruction and concentration field calculation module is further used for: Based on a physical model of the linear superposition of attenuation effects from dust and water mist, for each pixel in the two-dimensional pixel grid, a system of two linear equations is solved to separate the dust mass concentration distribution map and the water mist mass concentration distribution map: ; in, Indicates the bus linear attenuation coefficient. Represents the spatial coordinates of a two-dimensional pixel grid on the cross-section of the tunnel. This indicates the time when the measurement was performed. Indicates the specific wavelength of the light beam used. This represents the dust mass concentration to be determined. This represents the water mist mass concentration to be determined. The extinction coefficient per unit mass of dust. This represents the extinction coefficient per unit mass of water mist.

[0011] Preferably, the time series analysis and diffusion prediction module is specifically used for: An optical flow analysis method is used to define a local neighborhood for each pixel in the dust mass concentration distribution map, and the optical flow constraint equation is solved to calculate the vector field based on the constraint assumption that the motion vector within the local neighborhood remains constant. Based on the calculated vector field, the future spatial location of the high-concentration dust region is predicted at a future time step using a linear extrapolation model.

[0012] Preferably, the calculation formula for the linear extrapolation model is: ; in, It is a high concentration of dust clumps at all times center coordinates , This represents the motion vector calculated from optical flow analysis. This indicates the predicted dust agglomeration at a future time. The center coordinates, It indicates a short time step in the future.

[0013] Preferably, the data processing and output module is specifically used for: Threshold segmentation and connected component analysis are performed on the dust mass concentration distribution map to identify the high-concentration dust region, and the centroid coordinates, coverage area and peak concentration of the high-concentration dust region are extracted as feature parameters. Based on the extracted feature parameters, the current overall pollution level is determined.

[0014] Preferably, the data processing and output module is further used for: Generate a collaborative control instruction, which includes the target dustfall area center coordinates obtained by weighted calculation of the current centroid position and the future spatial position of the high-concentration dust area, as well as the spray intensity and duration determined according to the total pollution level.

[0015] The present invention also provides a method for monitoring dust concentration in roadways, comprising the following steps: S1. Construct a detection field consisting of beams containing multiple specific spectral components on the cross-section of the roadway to be monitored, and collect the light intensity attenuation data of the beams after passing through the cross-section of the roadway to obtain projection data to represent the integral attenuation. S2. Based on the projection data, a spatial distribution map of the attenuation coefficient representing the tunnel cross-section is reconstructed, and the spatial distribution map of the attenuation coefficient is separated into a quantitative dust mass concentration distribution map and a water mist mass concentration distribution map using the spectral absorption characteristics of different substances. S3. By analyzing the dust mass concentration distribution map, calculate the vector field describing the dust motion state, and predict the future spatial location of the high-concentration dust area based on the vector field. S4. Process the dust mass concentration distribution map to identify high-concentration dust areas, and generate collaborative control commands based on the current location of the high-concentration dust areas, the future spatial location of the high-concentration dust areas, and the preset light intensity.

[0016] This invention provides a dust concentration monitoring system and method for roadways, which has the following beneficial effects: This invention uses a multispectral optical field sensing module to acquire light intensity attenuation data for two specific wavelengths, and a tomographic reconstruction and concentration field calculation module to solve a set of mathematical equations based on the difference in extinction coefficients per unit mass of dust and water mist. This allows the total attenuation coefficient to be accurately separated into independent dust mass concentration distribution maps and water mist mass concentration distribution maps, effectively eliminating the optical interference of water mist generated by dust suppression spraying in roadways on dust concentration measurement, and improving the accuracy and reliability of monitoring results.

[0017] This invention uses a time-series analysis and diffusion prediction module to calculate a vector field describing the motion state of dust using optical flow analysis. Based on this, a linear extrapolation model is established to predict the future spatial location of high-concentration dust areas. This realizes the transformation from static concentration presentation to dynamic diffusion trend tracking and forward-looking prediction, providing a predictive time window for dust suppression operations and enabling proactive prevention and control.

[0018] This invention uses a data processing and output module to comprehensively analyze and make decisions on the current location, future spatial location, and pollution level of a high-concentration dust area. It can generate structured collaborative control commands that include the center coordinates of the target dust suppression area, spray intensity, and duration. This enables closed-loop linkage with downstream intelligent dust suppression devices, guiding them to perform predictive, zoned, and graded precise operations, thereby improving dust suppression efficiency and reducing water consumption. Attached Figure Description

[0019] Figure 1 This is a diagram illustrating the architecture of a roadway dust concentration monitoring system according to an embodiment of the present invention. Figure 2 This is a schematic diagram illustrating the principle of tomographic reconstruction in an embodiment of the present invention; Figure 3 This is a flowchart of the information processing of the data processing and output module in an embodiment of the present invention. Figure 4 This is a flowchart of a method according to an embodiment of the present invention. Detailed Implementation

[0020] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] See Figure 1 , Figure 1 This is a schematic diagram of a roadway dust concentration monitoring system according to an embodiment of the present invention. The embodiment of the present invention provides a roadway dust concentration monitoring system, including a multispectral optical field sensing module, a synchronous control and data acquisition module, a tomographic reconstruction and concentration field calculation module, a time series analysis and diffusion prediction module, and a data processing and output module.

[0022] The physical environment information of the tunnel cross-section is acquired through active sensing. This information is then subjected to multi-level in-depth processing and analysis, ultimately outputting multi-dimensional, structured data that can be directly used for precise decision-making and control. Specifically, the multispectral light field sensing module constructs a detection field on the tunnel cross-section to be monitored, consisting of multiple sets of beams intersecting at multiple angles and containing various specific spectral components.

[0023] The synchronous control and data acquisition module is used to precisely drive and control the multispectral light field sensing module, and synchronously acquire the light intensity attenuation data of the light beam after passing through the tunnel cross-section. This module converts the physical effects caused by media such as dust and water mist in the environment into raw digital signals that can be processed later.

[0024] The tomographic reconstruction and concentration field calculation module receives the raw light intensity attenuation data output by the synchronization control and data acquisition module. Based on the principle of computer tomography, this module performs mathematical inversion on linear projection data from multiple angles to reconstruct a two-dimensional spatial distribution map representing the attenuation coefficient of the entire tunnel cross-section. Subsequently, utilizing the prior knowledge that different substances have different absorption characteristics under different spectra, this module establishes and solves a system of mathematical equations to accurately separate the mixed attenuation coefficient distribution map into independent, quantitative dust mass concentration distribution maps and water mist mass concentration distribution maps.

[0025] The time-series analysis and diffusion prediction module receives a series of time-ordered pure dust concentration distribution maps continuously output from the previous module. This module analyzes the morphological and positional changes of dust clouds between consecutive frames to calculate a vector field describing their motion, thereby dynamically tracking the diffusion direction and velocity of the dust mass. Based on this, a prediction model is established to estimate its spatial location within a short future timeframe.

[0026] As the final stage of the system, the data processing and output module comprehensively analyzes and processes all the calculated and predicted information. This module automatically identifies and locates high-concentration dust areas, determines their pollution levels according to preset rules, and integrates multi-dimensional information such as current location, predicted location, and pollution level in a structured manner. Finally, it generates collaborative control commands that can be directly used to guide downstream intelligent dust suppression devices to perform precise operations in advance, by zone, and by grade.

[0027] The roadway dust concentration monitoring system provided in this embodiment of the invention includes a multispectral optical field sensing module. This module is the physical sensing front end of the system, used to construct a detection field composed of a large number of beams of light intersecting at multiple angles on the cross-section of the roadway to be monitored, and to acquire in real time the attenuation information of the detection field caused by the presence of media (such as dust and water mist) in the roadway.

[0028] The multispectral light field sensing module includes a transmitting unit, a receiving unit, and a support and mounting structure for securely installing them.

[0029] The emitting unit is fixed to one side of the tunnel and consists of multiple laser emitters mounted on a rigid support frame. To achieve accurate identification of different phases such as dust and water mist, each laser emitter in this embodiment is a multi-wavelength light source module, internally integrating at least two laser sources capable of independently emitting beams of different specific wavelengths. These two laser sources emit wavelengths of... and The beam of light.

[0030] wavelength and The selection of wavelength is based on the significant difference in the unit mass extinction coefficient between the target monitored object (e.g., coal dust) and common potential optical interferences in the tunnel (e.g., water mist sprayed for dust suppression) at the two wavelengths. Specifically, a wavelength can be selected. Located in the near-infrared band, coal dust particles exhibit strong absorption or scattering characteristics due to their physicochemical properties; while at another wavelength... Then, another near-infrared band with characteristic absorption peaks for water molecules was selected. This differentiated wavelength configuration provides the necessary raw physical input for subsequent algorithm modules to perform phase decoupling and concentration separation. In addition, the beam output from each laser emitter is collimated by an optical system to ensure that it maintains a small divergence angle after passing through the entire tunnel width, thereby ensuring the accuracy of spatial detection.

[0031] The receiving unit is located on the other side of the tunnel, deployed opposite the transmitting unit. This receiving unit consists of one or more linearly arranged photodetector arrays. The total length and spatial resolution of the arrays are matched to those of the transmitting unit to ensure that beam signals from all laser emitters in the transmitting unit can be received. Each photodetector possesses high sensitivity and high-speed response characteristics. High sensitivity ensures that even after the beam has undergone severe attenuation due to a high-concentration dust field, a valid light intensity signal can still be detected; the high-speed response characteristic is to cooperate with the signal modulation and demodulation techniques used in subsequent synchronization control and data acquisition modules to accurately distinguish light signals from different emitters.

[0032] The support and mounting structure is used to precisely and securely fix the transmitting and receiving units to the tunnel wall. This structure possesses sufficient mechanical strength and vibration resistance to ensure the long-term stability of the spatial coordinates of all laser emitters and photodetectors in the complex working conditions of a mine. This geometric stability is a necessary prerequisite for subsequent tomographic reconstruction calculations, as it guarantees that the spatial path of each optical path is known and fixed.

[0033] The roadway dust concentration monitoring system provided in this embodiment of the invention further includes a synchronization control and data acquisition module. This synchronization control and data acquisition module is used to precisely drive, coordinate and manage the working timing of the sensing module, while simultaneously acquiring, processing and formatting the raw physical signals output by the sensing module without distortion, providing the necessary high-precision projection dataset for subsequent tomographic reconstruction.

[0034] The synchronization control and data acquisition module includes a timing control unit, a signal modulation unit, a synchronization acquisition and demodulation unit, and a data processing unit.

[0035] The timing control unit is the command center of the entire data acquisition process. To ensure that the receiving unit can clearly distinguish which part of the energy in the composite optical signal received at the same time comes from which specific laser emitter, the timing control unit employs a specific signal multiplexing technique. In a specific embodiment, this multiplexing technique is Code Division Multiplexing (CDM). According to the instructions of the timing control unit, the signal modulation unit assigns a unique pseudo-random code sequence with good autocorrelation and low cross-correlation to each laser emitter. When emitting a beam, the driving current of each laser emitter is high-frequency modulated by this unique code sequence, so that its emitted light intensity carries a digital fingerprint of its own identity.

[0036] Because all laser emitters can be modulated by different code sequences and operate simultaneously, the system can achieve near-instantaneous scanning of the tunnel cross-section. The synchronous acquisition and demodulation unit works closely with the receiving unit. Each photodetector in the receiving unit converts the total light intensity signal received—composed of multiple modulated beams superimposed—into an electrical signal and transmits it to the synchronous acquisition and demodulation unit. This synchronous acquisition and demodulation unit internally contains a correlator array corresponding to the code sequence from the transmitting end. For the signal from the first... The unit combines the mixed signal from the photodetector with the signal from the transmitter. By performing correlation operations on the pseudo-random code sequences used by each laser emitter, the optical path can be accurately demodulated from the mixed signal. The light intensity component. This technical solution based on code division multiplexing and correlation demodulation has extremely strong resistance to inter-channel crosstalk and environmental noise interference.

[0037] The data processing unit converts the demodulated analog electrical signals representing the light intensity of each optical path into digital quantities via a high-speed analog-to-digital converter (ADC), which are recorded as real-time light intensity. ,in These represent different wavelengths. This data processing unit relies on the reference light intensity stored during the system initialization and calibration phase. The integral attenuation of each optical path at both wavelengths, i.e., the projected value, is calculated. Its calculation follows the formula below: ; in, At any moment , wavelength is At that time, along the first The transmitter to the first The optical path of each receiver The projection value; It is the reference light intensity calibrated for this optical path in clean air; At any moment Real-time measured light intensity It represents the natural logarithm.

[0038] After completing the scanning and calculation of a complete cross-section, the data processing unit will process all... The projection values ​​are organized to ultimately output two complete and independent sets of wavelengths. and These correspond to the projected data matrices. These two sets of matrices are then transmitted to the tomographic reconstruction and concentration field calculation module as direct input for all subsequent calculations.

[0039] See Figure 2 , Figure 2 This is a schematic diagram illustrating the principle of tomographic reconstruction according to an embodiment of the present invention. The tunnel dust concentration monitoring system provided in this embodiment further includes a tomographic reconstruction and concentration field calculation module. This module receives two sets of multi-angle projection data matrices output by the synchronization control and data acquisition module, and through performing a series of precise mathematical operations, finally outputs a spatial distribution map of the independent quantitative mass concentrations of dust and water mist within the tunnel cross-section.

[0040] The tomographic reconstruction and concentration field calculation module reconstructs the attenuation field by inverting and reconstructing an attenuation coefficient distribution map that reflects the physical characteristics of the entire two-dimensional cross-section through a series of one-dimensional linear projection data. The module first mathematically discretizes the tunnel cross-section to be monitored into a graph composed of... A two-dimensional grid composed of pixel units. For each pixel... Its internal linear attenuation coefficient Unknown quantities assumed to be uniform Indicates the total number of pixels. Indicates the pixel index.

[0041] Based on this discretization model, along any optical path The integral attenuation can be approximated as a weighted sum of the attenuation contributions of each pixel traversing the optical path. Therefore, the integral form of the Beer-Lambert law is transformed into the following large system of linear equations: ; in, It is known, along the optical path At wavelength The measured projection value; It is the problem to be solved, the first Each pixel at time ,wavelength The average linear decay coefficient under these conditions; It is the geometric weighting factor, representing the optical path. With the The length of the intersection of pixels. All these factors together form a sparse matrix uniquely determined by the system geometry.

[0042] To solve this large-scale system of equations, this embodiment employs the Algebraic Reconstruction Technique (ART) for iterative computation. This technique starts from an initial guess field (e.g., all...). Starting with zero values, the system sequentially corrects the current image field using the projection values ​​of each optical path in a preset order until the calculation results converge. The correction process for each iteration follows the formula below: ; in, This is the index of the projection data currently used for correction; This represents the number of iterations. It is a relaxation factor that takes values ​​in the interval (0,2) to control the convergence speed and stability of the iteration process; It is the first The pixel in the first The linear decay coefficient at the next iteration; It is the first The pixel in the first The linear decay coefficient at the next iteration; It is the first The projection value of the optical path; It is a geometric weighting factor, representing the first... The light path and the first The length of the intersection of pixels; It represents the total number of pixels.

[0043] The tomographic reconstruction and concentration field calculation module processes the two sets of wavelengths acquired. The projection data are reconstructed independently using the above process, resulting in two pixel-aligned spatial distribution maps of the bus linear attenuation coefficient at two different wavelengths. and .

[0044] Subsequently, the module calculates the phase concentration. In this step, the total attenuation coefficient at any point on the cross-section of the tunnel is the linear superposition of the attenuation effects produced by the dust and water mist present at that point. For any pixel in the cross-sectional grid... Its physical model can be expressed as: ; Using the total attenuation coefficients obtained from the previous reconstruction at two different wavelengths, the module establishes and solves a problem for each pixel containing two unknowns. A system of two linear equations in two variables: ; in, Indicates the bus linear attenuation coefficient; and These are the mass concentrations of dust and water mist to be determined for that pixel; and Dust and water mist at wavelengths The extinction coefficient per unit mass. These two coefficients are physical constants representing the optical properties of a material and can be pre-calibrated experimentally.

[0045] The solution to this system of equations is accomplished through matrix operations. This is achieved by applying matrix operations to equations composed of unit mass extinction coefficients. By inverting the coefficient matrix, the pure dust mass concentration can be accurately resolved at each pixel. and water mist mass concentration The validity of this solution process depends on the wavelength. and The choice of wavelength ensures that dust and water mist have linearly independent absorption characteristics at these two wavelengths, thus ensuring that the coefficient matrix is ​​non-singular.

[0046] Ultimately, the tomographic reconstruction and concentration field calculation module outputs two data images: one is a high-resolution, quantitative spatial distribution map of dust mass concentration without water mist interference, and the other is a spatial distribution map of water mist mass concentration. These two images are then transmitted to subsequent modules for further time-series analysis and decision-making.

[0047] The roadway dust concentration monitoring system provided in this embodiment of the invention further includes a time series analysis and diffusion prediction module. This time series analysis and diffusion prediction module dynamically analyzes a series of time series dust concentration maps output by the tomographic reconstruction and concentration field calculation module, thereby improving the monitoring capability from a static presentation of the dust distribution in the roadway cross section to dynamic tracking of its movement patterns and forward-looking prediction of its short-term future state.

[0048] The time-series analysis and diffusion prediction module receives a series of time-series, purified dust mass concentration distribution maps after removing water mist interference. And based on the optical flow analysis method. The basic physical assumption of this optical flow analysis method is that, within a sufficiently short time interval... Within the tunnel cross-section, the dust concentration in a small area is constant, and its apparent movement on the image is caused by the actual physical movement of the dust cloud. This relationship can be described by the following formula: ; in, This indicates the mass concentration of dust to be determined at this pixel. Represents the spatial coordinates of a two-dimensional pixel grid on the cross-section of the tunnel; Indicates time; , Indicates the displacement increment; Indicates a time interval.

[0049] Expanding the right side of the above equation using a first-order Taylor series and neglecting higher-order terms, we can derive the fundamental constraint equations of the optical flow method: ; in, and These are dust clouds in and The velocity component in the direction of motion, i.e., the motion vector to be solved. ; and The gradient of the dust concentration map in space can be obtained by performing a difference operation on the image; The gradient of dust concentration over time can be obtained by comparing the concentration changes between the current frame and the previous frame.

[0050] Since the above optical flow constraint equation contains two unknowns To obtain a unique solution to a single equation, this embodiment employs the Lucas-Kanade algorithm. This algorithm introduces an additional local constraint assumption, namely, within a pixel-based... small neighborhood centered on Within, the motion vectors of all pixels All are the same. Based on this assumption, all pixels within this neighborhood together constitute a region about... and The system of overdetermined linear equations is given. Solving this system of equations using the least squares method allows us to calculate the motion vector of the neighborhood center. By performing this calculation on all pixels or key feature points of the entire image, the temporal analysis and diffusion prediction module ultimately generates a two-dimensional motion vector field that corresponds to the dust concentration map and represents the internal motion state of the dust cloud.

[0051] Obtain the current moment motion vector field Subsequently, the time series analysis and diffusion prediction module further performs the diffusion prediction function. For the current concentration map... The centroid of any high-concentration dust agglomerate identified in the sample is denoted as . This time-series analysis and diffusion prediction module can predict the motion vector at a given location over a short time step using a linear extrapolation model. The spatial position after : ; in, It is a high concentration of dust clumps at all times center coordinates ; It is the motion vector calculated by optical flow analysis at the central coordinate position; It is a prediction that the dust agglomeration will occur in the future. The center coordinates; It indicates a short time step in the future.

[0052] The time-series analysis and diffusion prediction module outputs two key dynamic pieces of information: a motion vector field map representing the overall movement of the current dust field, and predicted data on the future location of high-concentration hazardous areas. These two pieces of information are transmitted to the data processing and output module, serving as the core decision-making basis for achieving proactive and intelligent dust control.

[0053] See Figure 3 , Figure 3 This is an information processing flowchart of a data processing and output module according to an embodiment of the present invention. The roadway dust concentration monitoring system and method provided in this embodiment further includes a data processing and output module. This module is used to comprehensively process and analyze the multi-dimensional, high-density information output by upstream modules, ultimately generating structured, visualized information and collaborative control instructions that can be directly applied to production practice.

[0054] The data processing and output module includes a high-concentration area identification unit, a pollution level assessment unit, and a collaborative control command generation unit.

[0055] The high-concentration area identification unit receives a quantitative spatial distribution map of dust mass concentration output by the tomographic reconstruction and concentration field calculation module. To automatically identify and locate dust clumps that pose a threat to safe production, this unit first performs threshold segmentation processing on the input concentration map. This is done by comparing the concentration value of each pixel in the map with a preset dust concentration safety threshold. By comparison, all pixels with concentration values ​​exceeding the threshold are identified as components of high-concentration pollution areas. Subsequently, the unit employs a connected component analysis algorithm to aggregate spatially adjacent high-concentration pixels into independent, parameterizable dust cloud objects. For each identified independent cloud, the unit calculates and extracts its key geometric and physical features, including the cloud's centroid coordinates, coverage area, and peak concentration within the region.

[0056] The pollution level assessment unit receives parameterized cloud information output by the high-concentration area identification unit. Based on preset multi-level assessment standards corresponding to the mine safety production regulations, this unit quantifies and rates the overall pollution status of the current roadway cross-section.

[0057] In one specific embodiment, the assessment criterion is a decision matrix that comprehensively considers the coverage area of ​​the largest dust cloud and its peak concentration. By comparing the extracted feature parameters with the classification thresholds in the matrix, the unit finally outputs a pollution level representing the current overall hazard level, for example, from Level I (clean) to Level III (severely polluted).

[0058] The collaborative control command generation unit is the decision-making center of this module. It integrates real-time and predictive information from multiple upstream modules to generate forward-looking and intelligent dust suppression control commands. The key input information received by this collaborative control command generation unit includes: Current centroid position of each dust cloud determined by the high-concentration area identification unit ; The predicted future locations of these cloud clusters, output by the time series analysis and diffusion prediction module. ; The current overall pollution level output by the pollution level assessment unit.

[0059] The command generation logic of this collaborative control command generation unit lies in transforming the control objective of dust suppression equipment (such as a zoned spray system) from a passive response to existing high-concentration areas to a predictive, comprehensive attack targeting of the current location and future trajectory of the dust cloud. Specifically, the structured control command data packet generated by this collaborative control command generation unit includes the center coordinates of the target dust suppression area, which are based on... and The aiming point is obtained through weighted calculation; at the same time, the instruction also includes the spray intensity and duration determined according to the pollution level.

[0060] The data processing and output module outputs two types of information to the external system: First, it displays the dust concentration distribution, water mist concentration distribution, and motion vector field representing the diffusion trend in real time and dynamically on the human-machine interface in the form of a pseudo-color cloud map, providing intuitive visual decision support for on-site managers; Second, it sends the generated structured collaborative control commands, which include forward-looking target locations and execution intensity, to downstream intelligent dust suppression and treatment devices through the industrial bus, realizing full-process, closed-loop, and intelligent dust control.

[0061] See Figure 4 , Figure 4 This is a flowchart of a method for monitoring dust concentration in roadways according to an embodiment of the present invention. The present invention provides a method for monitoring dust concentration in roadways, comprising the following steps: S100, System Initialization and Environmental Calibration. Under clean air conditions free of dust and water mist in the tunnel, the tunnel dust concentration monitoring system is activated. The synchronous control and data acquisition module drives all laser emitters in the multispectral light field sensing module to emit two different wavelengths of light beams in sequence. Simultaneously, the reference light intensity formed on each photodetector in the receiving unit after passing through the clean air is recorded. This set of data is stored as a reference for subsequent calculation of light attenuation.

[0062] Real-time monitoring and projection data acquisition. During dust-generating operations such as tunneling, the system enters continuous monitoring mode. The synchronous control and data acquisition module continuously drives the sensing module to scan the cross-section at a preset time resolution, acquiring the real-time attenuated light intensity after passing through a tunnel cross-section containing a mixture of dust and water mist. Based on the stored reference light intensity and the real-time acquired attenuated light intensity, this module calculates the integral attenuation projection value of each optical path at two wavelengths, forming two sets of projection data matrices corresponding to the wavelengths.

[0063] S200, Reconstruction and Concentration Field Separation. The tomographic reconstruction and concentration field calculation module receives two sets of projection data matrices and, based on algebraic reconstruction technology, independently inverts and reconstructs the two sets of one-dimensional projection data into two two-dimensional spatial distribution maps of the total linear attenuation coefficient. Subsequently, the module uses pre-calibrated unit mass extinction coefficients of dust and water mist for two wavelengths to establish and solve a system of two linear equations in two variables for each pixel in the distribution map, thereby accurately separating the mixed total attenuation coefficient into a quantitative and pure spatial distribution map of dust mass concentration and a spatial distribution map of water mist mass concentration.

[0064] S300, Time Series Analysis and Diffusion Trend Prediction. This module continuously receives and processes the output pure dust mass concentration distribution map, using time as a sequence. By applying an optical flow estimation algorithm, it compares and analyzes the morphology and positional changes of dust clouds between adjacent time frames, calculating a two-dimensional motion vector field representing the velocity and direction of dust movement within the current cross-section. Based on this motion vector field, the module extrapolates the centroid position of identified high-concentration dust clumps to predict their spatial position a short time step in the future.

[0065] S400, Integrated Decision and Command Information Output. The data processing and output module comprehensively analyzes the current dust concentration map, water mist concentration map, dust motion vector field, and the predicted future location of high-concentration areas. This module performs threshold segmentation on the dust concentration map, automatically identifies and parameterizes all exceeding-standard areas, and assesses the current overall pollution level based on parameters such as the area and peak concentration of the exceeding-standard areas. This module integrates the current location, predicted location, and pollution level of high-concentration areas to generate structured collaborative control commands. These commands contain all the information needed to guide downstream dust suppression devices in proactive, precise positioning, and tiered response operations, and are simultaneously displayed in multi-dimensional information on a visualization terminal.

[0066] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A dust concentration monitoring system for roadways, characterized in that, include: A multispectral optical field sensing module is used to construct a detection field consisting of light beams containing multiple specific spectral components on the cross-section of the roadway to be monitored. The synchronous control and data acquisition module is used to drive and control the multispectral light field sensing module and acquire the light intensity attenuation data of the light beam after passing through the cross section of the alley to obtain projection data to represent the integral attenuation. The tomographic reconstruction and concentration field calculation module is used to reconstruct a spatial distribution map of the attenuation coefficient representing the tunnel cross section based on the projection data, and to separate the spatial distribution map of the attenuation coefficient into a quantitative dust mass concentration distribution map and a water mist mass concentration distribution map by utilizing the spectral absorption characteristics of different substances. The time series analysis and diffusion prediction module is used to calculate a vector field describing the dust motion state by analyzing the dust mass concentration distribution map, and predict the future spatial location of the high-concentration dust area based on the vector field. The data processing and output module is used to generate collaborative control commands based on the current location of the high-concentration dust area, the future spatial location of the high-concentration dust area, and the preset light intensity.

2. The tunnel dust concentration monitoring system according to claim 1, characterized in that, The multispectral light field sensing module includes: The transmitting unit includes multiple multi-wavelength light source modules, each of which is used to transmit light at a wavelength of [wavelength value missing]. and A light beam, the wavelength of which is and The beam has the following characteristics: the target object and potential optical interference objects are at the same wavelength. and The extinction coefficient per unit mass varies; The receiving unit, comprising multiple photodetector arrays, is positioned opposite the transmitting unit and is used to receive the wavelength. and The beam of light.

3. The tunnel dust concentration monitoring system according to claim 1, characterized in that, The synchronization control and data acquisition module includes: A signal modulation unit is used to assign a unique pseudo-random code sequence to each laser transmitter and to modulate the drive current of the laser transmitter; The synchronous acquisition and demodulation unit is used to receive the mixed signal output by the receiving unit of the multispectral light field sensing module after converting the superimposed light beam into an electrical signal, and to demodulate the light intensity component belonging to a specific optical path by performing correlation operation between the mixed signal and the pseudo-random code sequence. The data processing unit is used to calculate the projection data based on the light intensity components and the pre-calibrated reference light intensity.

4. The tunnel dust concentration monitoring system according to claim 1, characterized in that, The chromatography reconstruction and concentration field calculation module is specifically used for: The cross-section of the roadway to be monitored is discretized into a two-dimensional pixel grid; Algebraic reconstruction technology is used to iteratively calculate the geometric relationship between the projection data calculated by the synchronous control and data acquisition module and the two-dimensional pixel grid to obtain the spatial distribution map of the total linear attenuation coefficient corresponding to the two wavelengths.

5. The tunnel dust concentration monitoring system according to claim 4, characterized in that, The chromatography reconstruction and concentration field calculation module is also used for: Based on a physical model of the linear superposition of attenuation effects from dust and water mist, for each pixel in the two-dimensional pixel grid, a system of two linear equations is solved to separate the dust mass concentration distribution map and the water mist mass concentration distribution map: ； in, Indicates the bus linear attenuation coefficient. Represents the spatial coordinates of a two-dimensional pixel grid on the cross-section of the tunnel. This indicates the time when the measurement was performed. Indicates the specific wavelength of the light beam used. This represents the dust mass concentration to be determined. This represents the water mist mass concentration to be determined. The extinction coefficient per unit mass of dust. This represents the extinction coefficient per unit mass of water mist.

6. The tunnel dust concentration monitoring system according to claim 1, characterized in that, The time series analysis and diffusion prediction module is specifically used for: An optical flow analysis method is used to define a local neighborhood for each pixel in the dust mass concentration distribution map, and the optical flow constraint equation is solved to calculate the vector field based on the constraint assumption that the motion vector within the local neighborhood remains constant. Based on the calculated vector field, the future spatial location of the high-concentration dust region is predicted at a future time step using a linear extrapolation model.

7. The tunnel dust concentration monitoring system according to claim 6, characterized in that, The calculation formula for the linear extrapolation model is as follows: ； in, It is a high concentration of dust clumps at all times center coordinates , This represents the motion vector calculated from optical flow analysis. This indicates the predicted dust agglomeration at a future time. The center coordinates, It indicates a short time step in the future.

8. The tunnel dust concentration monitoring system according to claim 1, characterized in that, The data processing and output module is specifically used for: Threshold segmentation and connected component analysis are performed on the dust mass concentration distribution map to identify the high-concentration dust region, and the centroid coordinates, coverage area and peak concentration of the high-concentration dust region are extracted as feature parameters. Based on the extracted feature parameters, the current overall pollution level is determined.

9. A roadway dust concentration monitoring system according to claim 8, characterized in that, The data processing and output module is also used for: Generate a collaborative control instruction, which includes the target dustfall area center coordinates obtained by weighted calculation of the current centroid position and the future spatial position of the high-concentration dust area, as well as the spray intensity and duration determined according to the total pollution level.

10. A method for monitoring dust concentration in roadways, characterized in that, The method is applied to a roadway dust concentration monitoring system according to any one of claims 1-9, and the method includes the following steps: S1. Construct a detection field consisting of beams containing multiple specific spectral components on the cross-section of the roadway to be monitored, and collect the light intensity attenuation data of the beams after passing through the cross-section of the roadway to obtain projection data to represent the integral attenuation. S2. Based on the projection data, a spatial distribution map of the attenuation coefficient representing the tunnel cross-section is reconstructed, and the spatial distribution map of the attenuation coefficient is separated into a quantitative dust mass concentration distribution map and a water mist mass concentration distribution map using the spectral absorption characteristics of different substances. S3. By analyzing the dust mass concentration distribution map, calculate the vector field describing the dust motion state, and predict the future spatial location of the high-concentration dust area based on the vector field. S4. Process the dust mass concentration distribution map to identify high-concentration dust areas, and generate collaborative control commands based on the current location of the high-concentration dust areas, the future spatial location of the high-concentration dust areas, and the preset light intensity.

Images