Head-mounted respiratory protection system pollution monitoring and air supply control method and storage medium

Abstract

The invention discloses a pollution monitoring and air supply control method for a head-mounted respiratory protection system and a storage medium. The pollution monitoring and air supply control method comprises the following steps: collecting pollutant time sequence data in real time by using an arranged multispectral sensor, and transmitting the pollutant time sequence data to a central processing unit of the head-mounted respiratory protection system; the method comprises the following steps: processing time sequence data of pollutants by using fast Fourier transform, extracting dominant frequency characteristics of the pollutants, and calculating an average particle size of the pollutants; the air volume of each air inlet channel of the head-mounted respiratory protection system is dynamically distributed by combining the average particle size and the diffusion characteristic of pollutants and the fluid constraint of the respiratory protection system and utilizing a particle swarm optimization algorithm, and the air volume distribution is optimized according to the characteristics of the pollutants with different particle sizes and the capacity of the respiratory protection system; the lowest energy consumption is realized; meanwhile, the optimal pollutant removal effect is ensured; and according to the optimized air volume value, a control instruction is generated to adjust the air valve opening degree of each air inlet channel, the air volume distribution of each air inlet channel is accurately controlled, and the health of workers in the current workplace is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of headband-mounted respiratory protection systems, and more particularly to a method and storage medium for pollution monitoring and air supply control of headband-mounted respiratory protection systems. Background Technology

[0002] Mining and chemical production sites accumulate large amounts of dust, posing a threat to workers. Dust consists of solid particles suspended in the air, and dust generated during production processes is called "industrial dust." Its harm to human health is multifaceted, severe, and often irreversible. Major hazards include: triggering respiratory diseases (such as pneumoconiosis, chronic bronchitis, and asthma, which are the most direct and common) and even skin damage. Dust (especially silica dust, coal dust, and asbestos dust) deposits in the lungs, causing pulmonary fibrosis, leading to hardening and loss of elasticity of the lung tissue. Patients may experience coughing, chest pain, and difficulty breathing; in severe cases, it can lead to loss of working ability and even death.

[0003] With the increasing depth of mining and the increasing complexity of chemical production environments, pollutants in production sites exhibit three major characteristics: dynamic multi-source diffusion, time-varying particle size distribution, and enhanced spatial heterogeneity. Traditional head-mounted respiratory protection systems face the following technical bottlenecks: relying on a single sensor, they cannot capture spatial pollutant concentration gradients; forming monitoring blind spots in areas where equipment is obstructed or where airflow vortices form; and having a fixed airflow distribution, they ignore dynamic changes in pollutant particle size, resulting in large particles being retained due to insufficient airflow and small particles escaping filtration due to Brownian motion.

[0004] Furthermore, traditional head-mounted respiratory protection systems typically adjust airflow based on concentration thresholds, making it impossible to predict the migration path of pollutants, thus making workers more susceptible to exposure to sudden contamination plumes. Summary of the Invention

[0005] The purpose of this invention is to provide a method and storage medium for pollution monitoring and air supply control of a head-mounted respiratory protection system, which solves the above-mentioned technical problems pointed out in the prior art.

[0006] This invention provides a method for pollution monitoring and air supply control of a head-mounted respiratory protection system, comprising the following steps:

[0007] By setting up multiple distributed multispectral sensors in the current work site, time-series data of pollutants and air data are collected within the local location area of ​​each multispectral sensor within a time period.

[0008] Fast Fourier Transform was applied to the pollutant time series data to extract the pollutant dominant frequency features, and the average particle size of the pollutants was calculated based on the pollutant dominant frequency features.

[0009] Based on the average particle size of pollutants, combined with the pollutant diffusion dynamics and the fluid constraints of the respiratory protection system, the airflow configuration of each air intake channel of the head-mounted respiratory protection system currently worn by the user is performed through particle swarm optimization, and the airflow value of each air intake channel is obtained.

[0010] Control commands are generated based on the airflow value of each air intake channel to adjust the opening of the air valves in each air intake channel of the head-mounted respirator.

[0011] Preferably, the aforementioned pollutant time series data includes sulfide time series data, carbide time series data, and particulate matter time series data; the aforementioned air data includes pollutant particle density in local areas and wind speed vector in local areas.

[0012] Preferably, based on the average particle size of pollutants, combined with the pollutant diffusion kinetics and the fluid constraints of the respiratory protection system, the airflow configuration of each air intake channel of the currently worn head-mounted respiratory protection system is performed through particle swarm optimization to obtain the airflow value of each air intake channel, including the following steps:

[0013] Initialize the particle swarm parameters, which include the particle position vectors corresponding to the N airflow distribution schemes for each air intake channel.

[0014] The baseline total air volume value is determined based on the average particle size of the pollutants.

[0015] Based on the pollutant density, average particle size, and air viscosity in each local area, the pollutant settling velocity in each local area is calculated using Stokes' law.

[0016] Based on pollutant settling velocity, baseline total air volume, and historical data, we analyze the pollutant diffusion influence coefficient caused by each particle position vector and the comprehensive fitness corresponding to the pollutant diffusion influence coefficient.

[0017] The iteration ends when the overall fitness reaches the fitness threshold or the number of iterations reaches the iteration threshold. The air volume value allocation scheme corresponding to the overall fitness is output to obtain the air volume value of each air intake channel. Otherwise, the multiple particle position vectors with high fitness are subjected to cross mutation processing to generate N' new particle position vectors corresponding to the air volume value allocation scheme of each air intake channel. The above pollutant settling velocity calculation processing operation is returned until the air volume value of each air intake channel is output.

[0018] Preferably, the pollutant diffusion influence coefficient caused by each particle position vector and the comprehensive fitness corresponding to the pollutant diffusion influence coefficient are analyzed based on the pollutant settling velocity, the baseline total air volume value, and historical data. This includes the following operational steps:

[0019] The pollutant movement path is predicted by numerical integration based on the pollutant settling velocity, the baseline total air volume, and the turbulent diffusion effect.

[0020] Based on historical data analysis, the first sensitivity of the baseline total air volume value to the pollutant trajectory is obtained, and the trajectory change rate of the current baseline total air volume value to the pollutant movement path is obtained within a preset time period.

[0021] Based on the trajectory change rate of pollutant movement paths, the diffusion sensitivity integral algorithm is used to generate the pollutant diffusion influence coefficient caused by the particle position vector corresponding to the air volume value allocation scheme of each air intake channel under the current baseline total air volume value.

[0022] The overall fitness is calculated based on the pollutant diffusion influence coefficient and the particle position vector corresponding to the air volume distribution scheme of each air intake channel.

[0023] Preferably, the pollutant diffusion influence coefficient is generated based on the trajectory change rate of the pollutant movement path using a diffusion sensitivity integral algorithm, corresponding to the particle position vector of the airflow allocation scheme for each intake channel under the current baseline total airflow value. This includes the following steps:

[0024] Obtain the center point of each local location region, calculate the regional Euclidean distance between each two adjacent local location regions based on the center point, and generate the spatial weight between regions based on the spatial attenuation coefficient obtained by combining the regional Euclidean distance with the measured airflow turbulence vortex scale analysis of the current work site.

[0025] For each of the N airflow allocation schemes corresponding to each local location region, analyze the second sensitivity of channel p when the airflow of channel k changes with channel p. Based on the second sensitivity, establish the global channel coupling matrix.

[0026] The initial diffusion influence coefficient vector is obtained by analyzing the continuous trajectory changes based on the global channel coupling matrix and inter-regional spatial weights, combined with the data based on a preset time window.

[0027] The target pollutant diffusion influence coefficient is obtained by combining the regional average wind speed collected by a multispectral sensor within a local area over a preset time period with the initial diffusion influence coefficient vector and the pollutant settling velocity.

[0028] Preferably, the initial diffusion influence coefficient vector is obtained by analyzing the continuous trajectory changes based on the global channel coupling matrix and inter-regional spatial weights, combined with a preset time window. This includes the following steps:

[0029] The second sensitivity fusion coefficient between each intake channel in each of two adjacent local location regions is calculated based on the global channel coupling matrix and the spatial weight between regions.

[0030] Collect the continuous trajectory change rate of each air intake channel within a preset time window, and calculate the average trajectory change rate of each air intake channel within the time window;

[0031] The mutual influence correction between intake channels is calculated using the global channel coupling matrix and the second sensitivity fusion coefficient.

[0032] The initial diffusion influence coefficient vector is calculated by combining the mean trajectory change rate with the mutual influence correction between the intake channels.

[0033] Preferably, the target pollutant diffusion influence coefficient is obtained by combining the regional average wind speed within a local area collected by a multispectral sensor over a preset time period with the initial diffusion influence coefficient vector and the pollutant settling velocity.

[0034] The average wind speed in a local area is collected based on a multispectral sensor over a preset time period.

[0035] The turbulence correction factor is calculated by combining the pollutant settling velocity, the regional average wind speed, and the initial diffusion influence coefficient vector.

[0036] Determine if the pollutant settling velocity exceeds the pollutant settling velocity threshold; if so, trigger the sedimentation compensation correction strategy.

[0037] The excess of pollutant settling velocity is calculated, and a compensation value is generated by exponential decay using the excess of pollutant settling velocity. The target pollutant diffusion influence coefficient is then calculated based on the initial diffusion influence coefficient vector, turbulence correction factor, and compensation value.

[0038] Preferably, the pollutant settling velocity in each local location area is calculated using Stokes' law based on the pollutant density, average particle size, and air viscosity. The pollutant movement path is then predicted through numerical integration based on the settling velocity, baseline total airflow, and turbulent diffusion effects. This includes the following steps:

[0039] After dividing each local area into multiple three-dimensional grid blocks, air data and average particle size of pollutants are collected for each grid block. Based on the equipment location and air data in the current work site, the wind speed vector change rate caused by equipment obstruction is analyzed.

[0040] Extract the temperature and air viscosity data of the current work site; apply Stokes-Cunningham correction to the air data, average particle size of pollutants, and wind speed vector change rate of each grid block, and then perform principal component extraction to obtain the pollutant settling velocity of the current local area.

[0041] Collect historical site data for a preset time period, calculate the wind speed standard deviation based on the historical site data, extract the grid wind speed of each grid block in the current local location area, calculate the average grid wind speed of the current local location area based on the grid wind speed, and calculate the grid turbulence intensity based on the average grid wind speed and the wind speed standard deviation.

[0042] The grid turbulence diffusion coefficient is calculated based on the grid turbulence intensity; principal component extraction is performed on the grid turbulence diffusion coefficients of each grid block within the current local location region to obtain the regional turbulence diffusion coefficient within the current local location region.

[0043] For each grid block, the pollutant concentration time series variation data is analyzed based on the regional turbulent diffusion coefficient; based on the pollutant concentration time variation data, the pollutant diffusion process is analyzed based on the pollutant settling velocity of the current local area to obtain the pollutant concentration prediction data;

[0044] The pollutant movement path is obtained by linear fitting based on the pollutant concentration prediction data.

[0045] Preferably, pollutant concentration prediction data is obtained by analyzing the pollutant diffusion process based on the pollutant deposition velocity in the current local area, according to the pollutant concentration change data over time. This includes the following steps:

[0046] Based on the law of conservation of mass and Fick's diffusion law, a pollutant transport control equation is constructed for each grid block;

[0047] The finite volume method is used to transform the continuous pollutant transport control equations into a discrete grid computing model.

[0048] Based on the time-varying data of pollutant concentration, the grid computing model is subjected to time integration to obtain the predicted pollutant concentration data.

[0049] Compared with the prior art, the embodiments of the present invention have at least the following technical advantages:

[0050] Analysis of the pollution monitoring and air supply control method and storage medium of the head-mounted respiratory protection system provided by this invention reveals that, in practical applications, multiple dispersed multispectral sensors are deployed to collect real-time pollutant time-series data and transmit it to the central processing unit of the head-mounted respiratory protection system. This ensures accurate and real-time pollutant distribution data for each wearer, providing real-time basis for subsequent pollutant control and protection plan formulation. Furthermore, the central processing unit of the head-mounted respiratory protection system further processes the pollutant time-series data using Fast Fourier Transform (FFT) to extract the dominant frequency characteristics of the pollutants and calculate the average particle size. Further, the central processing unit combines the average particle size and diffusion characteristics of the pollutants with the fluid constraints of the respiratory protection system, utilizing particle swarm optimization (PSO). The algorithm dynamically allocates the airflow to each air intake channel of the head-mounted respiratory protection system. Then, based on the characteristics of pollutants of different particle sizes and the purification capacity of the respiratory protection system, it optimizes the airflow allocation and ultimately provides feedback to adjust the operation of the head-mounted respiratory protection system to achieve the lowest energy consumption while ensuring the best pollutant removal effect. Clearly, the aforementioned head-mounted respiratory protection system interacts with various sensors in the surrounding area to analyze the pollutant situation and location, ultimately providing feedback to control the system for intelligent regulation. Based on the optimized airflow value, it generates control commands to adjust the opening of the air valves in each air intake channel, precisely controlling the airflow allocation of each channel to protect the health of workers in the current workplace (most practically, it makes the head-mounted respiratory protection system more effective and maximizes its functionality). Attached Figure Description

[0051] Figure 1 A schematic diagram of the main process for pollution monitoring and air supply control methods for head-mounted respiratory protection systems;

[0052] Figure 2 This is a schematic diagram of data acquisition simulation in the pollution monitoring and air supply control method of a head-mounted respiratory protection system.

[0053] Figure 3 This is a schematic diagram of a three-dimensional mesh model simulating the location of equipment in a head-mounted respiratory protection system for pollution monitoring and air supply control. Detailed Implementation

[0054] 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.

[0055] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings.

[0056] Example 1

[0057] like Figure 1 As shown, Embodiment 1 of the present invention provides a method for pollution monitoring and air supply control of a head-mounted respiratory protection system, including the following operating steps:

[0058] S10, by setting up multiple dispersed multispectral sensors in the current work area, the system collects pollutant time-series data within the local location area of ​​each multispectral sensor within a time period (i.e., to explain that the control part (i.e., the central processing unit) of this head-mounted breathing system can interact with multiple multispectral sensors in the current work area, so the central processing unit of the head-mounted breathing system is a central processing system that can collect pollutant time-series data within multiple local location areas in the work area through multispectral sensors) and air data;

[0059] The aforementioned pollutant time-series data includes sulfide time-series data, carbide time-series data, and particulate matter time-series data; the aforementioned air data includes pollutant particle density and wind speed vector in local areas.

[0060] It should be noted that, as Figure 2 As shown, in the embodiments of this application, a multispectral sensor is deployed at regular intervals (e.g., 10 meters) in highly polluted work environments (including mines, chemical plants, and other complex indoor environments where air pollution is severe and can easily cause respiratory damage to workers), forming a sensor array. Each sensor is responsible for monitoring pollutants in its local area (radius 2-3 meters). By deploying multispectral sensors at different locations in the mine, pollutant data (initially in analog signal form) is collected in real time at a fixed frequency. These sensors can detect sulfides (such as SO2), carbides (such as CO, CO2), and particulate matter (e.g., dust in mines mainly comes from the particulate matter generated by the mechanical crushing of rocks and ores during drilling, blasting, tunneling, loading, and transportation. This generally includes free silica (SiO2), coal dust, diesel particulate matter, etc.; in chemical plants, the sources of particulate matter are extremely complex, depending on the raw materials, intermediates, and products produced; for example, reactant product dust, catalyst dust, etc.). The collected pollutant data in analog signal form is converted into digital signals using an analog-to-digital converter. The digital signals are then transmitted via wired or wireless means to a central processing unit mounted on the user's head-mounted respirator. This provides real-time, accurate pollutant distribution data for each worker's head-mounted respirator, laying the data foundation for subsequent processing. Multi-point data acquisition avoids data bias from single locations and improves the system's ability to perceive changes in the mine environment.

[0061] S20: Apply Fast Fourier Transform to the pollutant time series data to extract the pollutant main frequency characteristics, and calculate the average particle size of the pollutants based on the pollutant main frequency characteristics.

[0062] It should be noted that the embodiments described above utilize a central processing unit mounted on a user-worn respirator to perform a Fast Fourier Transform (FFT) on the time-series data (e.g., data from the past 30 seconds) of each pollutant. The dominant frequency component (the frequency component with the largest amplitude) is then extracted from the obtained spectrum. The average particle size of the pollutant is calculated based on a preset frequency-particle size mapping model. Through frequency domain analysis, the vibration characteristics of the pollutant can be identified, thereby inferring its physical properties (e.g., particle size). The average particle size is a crucial basis for subsequent airflow allocation because pollutants of different sizes require different treatment methods (e.g., large particles require physical filtration, while small particles require chemical adsorption).

[0063] S30, based on the average particle size of pollutants, combined with the pollutant diffusion dynamics and the fluid constraints of the respiratory protection system, uses particle swarm optimization to configure the airflow of each air intake channel of the currently worn head-mounted respiratory protection system, and obtains the airflow value of each air intake channel.

[0064] It should be noted that in the above embodiments of this application, based on the average particle size and diffusion characteristics of pollutants, the air volume of each air intake channel of the head-mounted respiratory protection system is dynamically allocated using a particle swarm optimization algorithm. The air volume of the three channels is dynamically allocated through an intelligent optimization algorithm (in terms of hardware, the head-mounted respiratory protection system described above in this embodiment is equipped with three matching air intake channels, referred to as air intake channels or channels), so as to achieve the best pollutant removal effect with the lowest energy consumption, while ensuring the comfort of the wearer.

[0065] S40 generates control commands based on the airflow value of each air intake channel to adjust the opening of the air valves in each air intake channel of the head-mounted respirator.

[0066] It should be noted that, in the embodiments of this application described above, control commands are generated based on the optimized airflow value to adjust the opening degree of the air valves in each channel. For example, in channel 1 (primarily for physical filtration), the opening degree of the butterfly valve is controlled by a stepper motor to adjust the airflow. The opening degree is calculated based on the airflow ratio to ensure adaptability to the filtration requirements of different pollutant concentrations. In channel 2 (primarily for sulfide adsorption), the displacement of the piezoelectric ceramic valve is proportional to the airflow, adjusting the adsorption efficiency of sulfides and ensuring the stability of the adsorption process. In channel 3 (primarily for carbide decomposition), the deformation rate of the shape memory alloy is used to adjust the airflow distribution, ensuring that the carbide decomposition reaction reaches its optimal state. Through real-time air valve control, combined with a feedback mechanism (such as an air pressure sensor), the working state of each channel is continuously corrected and optimized to ensure the efficiency of pollutant removal and the stability of the system.

[0067] The embodiments described above utilize multiple distributed multispectral sensors to collect real-time pollutant time-series data and transmit it to the central processing unit of the head-mounted respiratory protection system. This ensures accurate and real-time pollutant distribution data for each wearer, providing a real-time basis for subsequent pollutant control and protection plan development. Furthermore, the above technical solution processes the pollutant time-series data using Fast Fourier Transform (FFT) to extract the dominant frequency characteristics of the pollutants and calculate their average particle size. Further, the above technical solution combines the average particle size and diffusion characteristics of the pollutants with the fluid constraints of the respiratory protection system, employing a Particle Swarm Optimization (PSO) algorithm to dynamically allocate the airflow of each air intake channel of the head-mounted respiratory protection system. Based on the characteristics of pollutants with different particle sizes and the capabilities of the respiratory protection system, the airflow allocation is optimized to achieve the lowest energy consumption while ensuring optimal pollutant removal. Based on the optimized airflow value obtained in S30, control commands are generated to adjust the opening of the air valves in each air intake channel, precisely controlling the airflow allocation of each air intake channel and protecting the health of workers in the current workplace.

[0068] In S30, based on the average particle size of pollutants, combined with the pollutant diffusion dynamics and the fluid constraints of the respiratory protection system, the airflow configuration of each air intake channel of the currently worn head-mounted respiratory protection system is performed through particle swarm optimization to obtain the airflow value of each air intake channel. The operation steps include the following:

[0069] S31, Initialize particle swarm parameters, which include particle position vectors corresponding to N airflow distribution schemes for each air intake channel.

[0070] It should be noted that the above embodiments of this application initialize N particles (the particles are the air volume value allocation schemes of each air intake channel) and obtain the particle position vector of each particle accordingly.

[0071] S32, determine the baseline total air volume value based on the average particle size of pollutants;

[0072] It should be noted that the above-mentioned baseline total air volume value refers to the total air volume value of the three channels of the above-mentioned head-mounted respirator. The total air volume value of the three channels of the head-mounted respirator is determined by the average particle size of pollutants in the current workplace, which is collected by a multispectral sensor and processed by the central processing unit of the head-mounted respirator in the above-mentioned embodiment of the application.

[0073] Generally, when the average particle size of pollutants is less than or equal to a first particle size threshold, the maximum total airflow value is used; when the average particle size of pollutants is greater than the first particle size threshold but less than or equal to a second particle size threshold, a medium total airflow value is used; and when the average particle size of pollutants is greater than the second particle size threshold, a small total airflow value is used. Based on the above processing operation of determining the baseline total airflow value in the embodiments of this application, small-diameter pollutants require higher airflow velocities to overcome Brownian motion. Brownian motion refers to the random motion of pollutants in a fluid (such as air) due to continuous molecular collisions. Brownian motion is a phenomenon at the microscopic scale, usually occurring in very small particles, such as dust, smoke particles, and fine pollutants. Brownian motion is more pronounced in tiny pollutant particles. Because these pollutant particles have very small masses, they are more affected by collisions with surrounding gas molecules, thus exhibiting more obvious random motion. Analysis of the above technical solutions shows that a larger airflow can provide sufficient airflow velocity to help these tiny particles overcome the effects of Brownian motion, thereby effectively guiding them into the filter or collection device. Filters in head-mounted respirators typically have different collection efficiencies for particles of different sizes. For very small particles (such as dust and even smaller particles), their inertia is low, making them difficult to capture by traditional filter media. Therefore, the embodiments of this application are designed based on the above-mentioned principles, namely, a larger air volume can increase the contact opportunity between particles and the filter media, thereby improving the capture efficiency; that is, smaller particles have a weaker settling velocity due to gravity, and in still air, these particles may remain suspended in the air for a long time. Therefore, increasing the air volume can enhance airflow, allowing tiny particles to be drawn into the filtration system more quickly and reducing the residence time of particles in the air.

[0074] Smaller particles in the air are more affected by Brownian motion and tend to disperse easily. A larger airflow can enhance airflow, counteracting the random diffusion effect of Brownian motion and thus more effectively removing pollutants from the breathing zone. Therefore, due to the influence of Brownian motion, smaller pollutant particles require a larger airflow to ensure they are effectively captured and filtered, thereby protecting the respiratory health of the mask user.

[0075] By determining the baseline total air volume value based on the average particle size of pollutants, the problem of pollutants not being effectively filtered due to excessive or insufficient air volume (i.e., the fluid constraint of the above-mentioned respiratory protection system) is prevented.

[0076] S33: Based on the pollutant density, average particle size, and air viscosity in each local area, the pollutant settling velocity in each local area is calculated using Stokes' law; the pollutant movement path is predicted by numerical integration based on the pollutant settling velocity, the baseline total airflow value, and the turbulent diffusion effect; the first sensitivity of the baseline total airflow value to the pollutant trajectory is obtained by analyzing historical data, and the trajectory change rate of the current baseline total airflow value with respect to the pollutant movement path is obtained within a preset time period; based on the trajectory change rate of the pollutant movement path, the pollutant diffusion influence coefficient is generated by the diffusion sensitivity integral algorithm using the particle position vector corresponding to the airflow value allocation scheme of each intake channel under the current baseline total airflow value (i.e., the airflow value of each channel under multiple airflow allocation schemes).

[0077] S34, calculate the overall fitness based on the pollutant diffusion influence coefficient and the particle position vector corresponding to the air volume distribution scheme of each air intake channel;

[0078] S35: When the overall fitness reaches the fitness threshold or the number of iterations reaches the iteration threshold, the iteration ends and the airflow value allocation scheme corresponding to the overall fitness is output to obtain the airflow value of each air intake channel. Otherwise, the multiple particle position vectors with high fitness are subjected to cross-mutation processing to generate N' new particle position vectors corresponding to the airflow value allocation scheme of each air intake channel, and the processing operation of S33 above is returned until the airflow value of each air intake channel is output.

[0079] Analysis of the above scheme shows that step S32 involves determining the baseline total air volume value based on the average particle size of the pollutants. Therefore, it only calculates a total air volume value, i.e., the baseline total air volume value. However, how to reasonably allocate the branch air volume of the above baseline total air volume value to the three air intake channels is specifically achieved through steps S33-S35, especially S35. In this process, multiple allocation logics or allocation schemes are generated by cross-mutation of particle position vectors, and then the optimal allocation scheme is finally selected to coordinately control the air volume value of each air intake channel.

[0080] In S33, the pollutant diffusion influence coefficient caused by the particle position vector corresponding to the airflow value allocation scheme of each intake channel under the current baseline total airflow value is generated by the diffusion sensitivity integral algorithm based on the trajectory change rate of the pollutant movement path. The steps include the following:

[0081] S331: Obtain the center point of each local location region; calculate the regional Euclidean distance between each two adjacent local location regions based on the center point; generate the spatial weight between regions based on the spatial attenuation coefficient obtained by analyzing the airflow turbulence vortex scale obtained from the actual measurement of the current work site, according to the regional Euclidean distance; analyze the second sensitivity of channel p when the airflow of channel k changes in each of the N airflow value allocation schemes corresponding to each local location region; establish the global channel coupling matrix based on the second sensitivity.

[0082] The second sensitivity is calculated as follows:

[0083] ;

[0084] In the formula, This represents the second sensitivity of channel k to channel p in the local location region m. The number of sampling points within the time period. This refers to the time sequence number within the time period. Let be the normalized Euclidean distance between local location region a and local location region b. Let be the change in airflow in intake channel k at time sequence number t within local location region a; The spatial attenuation coefficient, Let be the rate of change of the pollutant trajectory at time t in the local location region b. The motion direction vectors of local location region a and local location region b are consistent at the i-th motion direction vector at time sequence number t; The change in air volume is grouped into one item;

[0085] It should be noted that the calculation method of the second sensitivity in the above embodiments of this application first uses the spatial attenuation term. Quantifying the propagation attenuation of airflow changes in the surrounding area, among which... The larger the value, the stronger the attenuation. The larger the value, the wider the impact range. During the calculation process, airflow variation and spatial attenuation are the two most critical parameters, dominating their influence on the second sensitivity, along with directional synergy. This reduces the complexity of multidimensional space and motion direction.

[0086] In the embodiments of this application described above, there is a strong correlation between the diffusion of pollutants in adjacent areas. By analyzing the spatial weights between each pair of adjacent local locations, the isolated analysis of each region is avoided, and the mutual influence of the diffusion of pollutants in adjacent areas is considered. On the other hand, for each local location, the cross-second sensitivity between channels (i.e., the aforementioned second sensitivity) is calculated. This is achieved by fixing other conditions (i.e., fixing the air intake volume and environmental parameters of one of the three air intake channels, including airflow turbulence vortex, average particle size of pollutants, etc.), and then changing the air intake volume of channel k in the remaining two air intake channels. The change in the trajectory change rate of channel p relative to the pollutant movement path is analyzed to obtain the second sensitivity (for example, if the air volume of channel k is increased by 1, the trajectory change rate of channel p relative to the pollutant movement path changes from 0.3 to 0.35, then the second sensitivity of air intake channel k relative to air intake channel p is 0.05). Then, the second sensitivity of each channel in all local locations is averaged to obtain a global channel coupling matrix, which is used to reveal the mutual influence between the three air intake channels. Specifically, each element in the global channel coupling matrix represents the second sensitivity of air intake channel k relative to air intake channel p.

[0087] S332: Based on the global channel coupling matrix and inter-regional spatial weights, the second sensitivity fusion coefficient between each intake channel in each of two adjacent local location regions is obtained (specifically, the value in the m-th row and n-th column of the global channel coupling matrix of local location region a and the value in the m-th row and n-th column of the global channel coupling matrix of local location region b, along with the inter-regional spatial weights, are multiplied element-wise to obtain the first intermediate value; the average of the first intermediate values ​​of all adjacent local location regions is then calculated to obtain the second sensitivity fusion coefficient, which represents the overall impact of each channel's airflow adjustment on pollutant diffusion after considering spatial correlation and channel interaction); data are collected for each intake channel within a preset time window. The continuous trajectory change rate is calculated by taking the average trajectory change rate of each intake channel within the time window (eliminating instantaneous fluctuation interference and obtaining a stable trajectory change assessment); the global channel coupling matrix and the second sensitivity fusion coefficient are used to calculate the mutual influence correction between intake channels (quantifying the cooperative / competitive effects between channels, for example, when the trajectory change rates of channel 1 and channel 2 differ significantly, the correction amount increases); the initial diffusion influence coefficient vector is calculated by combining the average trajectory change rate and the mutual influence correction between intake channels (integrating spatiotemporal characteristics, time integration provides stability, avoids instantaneous noise interference, and cross correction introduces the interaction between channels to ensure that the overall system behavior is considered during optimization).

[0088] S333: Based on the regional average wind speed collected by a multispectral sensor within a local area over a preset time period, a turbulence correction factor is calculated using the pollutant settling velocity, regional average wind speed, and initial diffusion influence coefficient vector. This factor quantifies the disturbance effect of turbulence on the pollutant diffusion path; the ratio of settling velocity to wind speed reflects the intensity of turbulence's influence on pollutant diffusion. The larger the ratio (settling-dominated), the smaller the turbulence effect; the smaller the ratio (turbulence-dominated), the larger the correction amount. The system determines whether the pollutant settling velocity exceeds a pollutant settling velocity threshold. If so (if not, no compensation is performed, and the target pollutant diffusion influence coefficient is calculated directly), a sedimentation compensation correction strategy is triggered: the pollutant settling velocity excess is calculated, and a compensation value is generated through exponential decay of the excess (the larger the excess, the greater the compensation; for the gravity settling characteristics of large particles, the required airflow is compensated). The target pollutant diffusion influence coefficient is calculated based on the initial diffusion influence coefficient vector, the turbulence correction factor, and the compensation value.

[0089] In S33, the pollutant settling velocity in each local location area is calculated using Stokes' law based on the pollutant density, average particle size, and air viscosity. The pollutant movement path is then predicted through numerical integration based on the settling velocity, baseline total airflow, and turbulent diffusion effects. This includes the following steps:

[0090] S3301: Traverse each local location region, divide the current local location region into multiple three-dimensional grid blocks of the same size, collect air data and average particle size of pollutants for each grid block, analyze the rate of change of wind speed vector caused by equipment obstruction (wind speed vector is the air flow direction in the current work area; in a closed environment, the air flow direction changes due to equipment obstruction) based on the equipment location and air data in the current work area, and extract the temperature and air viscosity data of the current work area; apply Stokes-Cunningham correction to the air data, average particle size of pollutants, and rate of change of wind speed vector of each grid block to calculate the pollutant settling velocity in each grid block, and perform principal component extraction based on the pollutant settling velocity in each grid block in the current local location region to obtain the pollutant settling velocity in the current local location region;

[0091] S3302: Collect historical site data (including wind speed data within historical work sites) for a preset time period, calculate the wind speed standard deviation based on the historical site data, extract the grid wind speed of each grid block in the current local location area, calculate the average grid wind speed of the current local location area based on the grid wind speed, and calculate the grid turbulence intensity based on the average grid wind speed and the wind speed standard deviation.

[0092] S3303, the grid turbulence diffusion coefficient is calculated based on the grid turbulence intensity; principal component extraction is performed based on the grid turbulence diffusion coefficient of each grid block in the current local location region to obtain the regional turbulence diffusion coefficient in the current local location region;

[0093] S3304 analyzes the time series variation data of pollutant concentration for each grid block based on the regional turbulent diffusion coefficient; based on the time variation data of pollutant concentration, the pollutant diffusion process is analyzed based on the pollutant deposition velocity of the current local area (the pollutant diffusion process includes pollutant diffusion and pollutant deposition) to obtain pollutant concentration prediction data.

[0094] S3305, the pollutant movement path is obtained by linear fitting based on the pollutant concentration prediction data.

[0095] It should be noted that the above-described embodiments of this application divide the work area into three-dimensional grid blocks of equal size, then collect air data and average particle size of pollutants from each grid block, and calculate the wind speed vector change rate (the change in airflow direction caused by equipment obstruction) based on the equipment location, quantify the impact of the equipment on the local airflow field and calculate the pollutant settling velocity. Further, in S3302, the grid turbulence intensity is determined by combining historical data analysis to characterize the degree of airflow instability and provide input data for subsequent diffusion analysis. Further, through diffusion analysis in S3303, the turbulence intensity is converted into a pollutant diffusion capacity index to reflect the overall diffusion intensity of the area. In S3304, the spatiotemporal distribution of pollutant concentration is predicted by combining time (pollutant concentration time change data) and space (current local location area). Then, in S3305, the spatiotemporal distribution of pollutant concentration prediction is linearly fitted to extract the pollutant movement path. Through multiphysics coupling modeling, the prediction of pollutant migration in the current work area is realized, providing a scientific basis for the airflow distribution scheme for workers' respiratory protection.

[0096] Specifically, in S3304, pollutant concentration prediction data is obtained based on the pollutant concentration time variation data through pollutant diffusion process analysis (the pollutant diffusion process includes pollutant diffusion and pollutant deposition) based on the pollutant deposition velocity of the current local location area. The process includes the following steps:

[0097] S33041: Based on the law of conservation of mass and Fick's diffusion law, a pollutant transport control equation is constructed for each grid block;

[0098] It should be noted that the pollutant transport control equation quantitatively describes the physical process of pollutant concentration change over time. This equation is expressed as: Pollutant concentration change rate = Pollutant diffusion (diffusion flux) - Pollutant deposition (deposition flux); where the pollutant concentration change rate represents the amount of change in pollutant concentration per unit time; pollutant diffusion is the diffusion of pollutants caused by turbulent motion, calculated as: regional turbulent diffusion coefficient × concentration difference between adjacent grids / grid spacing; pollutant deposition is the deposition of pollutants caused by gravity, calculated as: pollutant deposition velocity × concentration difference between upper and lower grids / vertical spacing.

[0099] The pollutant transport control equations in the above-described embodiments of this application transform complex physical processes into mathematical equations, providing a theoretical basis for numerical calculations and accurately describing the movement patterns of pollutants in the current workplace environment, thereby ensuring that the prediction model conforms to physical laws.

[0100] S33042: The finite volume method is used to transform the continuous pollutant transport control equations into a discrete grid computing model;

[0101] It should be noted that the above embodiments of this application are based on the volume of each grid block, transforming the pollutant transport control equations of continuous grid blocks in the current local location region into a discrete grid computing model. Specifically, each grid block is treated as an independent control entity, and the pollutant concentration value is stored at the center point of the control entity. Then, pollutant diffusion and pollutant deposition are transformed into grid boundary flux calculations; the complex continuous physical problem is transformed into discrete algebraic equations.

[0102] S33043: Based on the time-varying data of pollutant concentration, the grid computing model is advanced by time integration to obtain the predicted data of pollutant concentration.

[0103] It should be noted that in the above embodiments of this application, based on the pollutant concentration time change data, the explicit Euler method is applied to the grid computing model for time-progression calculation. First, the time step Δt = 0.1 seconds is set (to meet the CFL stability condition), and then the concentration is updated at each time step. The new concentration = old concentration + Δt × (the sum of all boundary fluxes); this realizes the dynamic prediction of the concentration evolution over time and captures the transient process of pollutant diffusion. The applied explicit Euler method is computationally efficient and meets the real-time requirements.

[0104] The purpose of the technical solution adopted in the above-described embodiments of this application is to simulate and predict the movement trajectory of pollutants in the complex environment of the workplace, providing a scientific basis for airflow optimization. The basic idea is to "divide and conquer, and simulate step by step," that is, to decompose the entire large space into countless small grids, calculate the behavior of pollutants in each small grid, and finally summarize the overall trend.

[0105] Specifically, based on the physical laws of pollutant diffusion, pollutants diffuse from areas of high concentration to areas of low concentration. The speed and intensity of this diffusion are determined by the "regional turbulent diffusion coefficient (step S3303)," which quantifies the ability of air turbulence to aid in pollutant diffusion. Airflow carries pollutants with it, and the direction and speed of the pollutants' movement are consistent with the direction and speed of the wind. This data comes from the actual wind speed and direction measured by sensors. Due to gravity, heavier pollutant particles will continue to settle downwards. The faster the settling speed (calculated from Stokes' Law in step S3301), the faster the pollutants will settle to the ground.

[0106] Based on the physical laws of pollutant diffusion mentioned above, it can be seen that the change in pollutant concentration in any place is equal to (the amount diffused in) minus (the amount diffused out + the amount lost due to sedimentation), which is the core physical equation controlling pollutant transport.

[0107] Furthermore, in step S33042, real air is continuous, but computers cannot handle an infinite number of points. Therefore, the entire workspace must be digitized. Specifically, each local area divided in step S3301 is further subdivided into smaller, fixed-volume cubic grids. Then, each cell is assigned current known attributes: pollutant concentration value, wind speed and direction, turbulent diffusion coefficient, and settling velocity. These known attributes are stored at the center of the grid. For each small grid, the "mass exchange," or flux, between it and its six adjacent grids (up, down, left, right, front, and back) is calculated. This includes diffusion flux (calculating how much pollutant is "transferred" from the high-concentration grid to the low-concentration grid) and settling flux (calculating how much pollutant "falls" from the upper grid to the lower grid).

[0108] Based on this, for each grid, the new concentration at the next moment = the current old concentration + (the sum of the fluxes of all neighboring inputs - the sum of the fluxes of all outputs), which is the "discrete algebraic equation" for each grid mentioned above; based on this, the entire complex physical problem is transformed into countless simple arithmetic problems.

[0109] Furthermore, in step S33043, the real pollutant concentration data measured by all sensors at the current moment are filled into the corresponding grid as the starting point of the simulation (Time=0) and the time step is set.

[0110] Then, a cyclic simulation process is performed. Starting from the first frame (T=0s), each grid is traversed. According to the rules established in step S33042, the new concentration value of each grid in the next time step is calculated. After all grids have been calculated, the old concentration value is replaced with the new concentration value. Then, the above process is repeated with the new concentration value as the new "current state" to calculate the state of the next time step. This process is repeated until the preset future time (e.g., 10 seconds later). Finally, the predicted data is obtained, which means that the predicted pollutant concentration at any location point at any future time point can be queried.

[0111] In a further step S3305, the predicted data is analyzed to generate a movement path, which is then fed into a particle swarm optimization algorithm. By simulating how different airflow distribution schemes would affect this path, the optimal configuration that can most effectively divert pollutants from the breathing zone is found.

[0112] Specifically, in S3301, the wind speed vector change rate caused by equipment obstruction is analyzed based on the equipment location and air data in the current work area, including the following operation steps:

[0113] S33011: Construct a 3D mesh model of the equipment locations in the current work area based on the equipment locations within the current work area;

[0114] It should be noted that, as Figure 3 As shown in the above embodiments of this application, before the initial implementation, the staff measures the position and volume of the equipment in the current work site in advance. Then, in embodiment S33011 of this application, an initial three-dimensional mesh model with the same scale as the current work site is constructed. The equipment in the current work site is placed in the initial three-dimensional mesh model with the same scale position and volume to form a three-dimensional mesh model of the equipment location in the site. The mesh area affected by the equipment is identified, providing a spatial basis for subsequent airflow analysis.

[0115] S33012: Extract baseline wind speed data; the baseline wind speed data includes the airflow velocity of the current work site, and then the baseline wind speed vector is calculated from the airflow data of the current work site.

[0116] It should be noted that the above benchmark wind speed data is the wind speed under conditions of no equipment interference (obstruction). Establishing an undisturbed wind speed benchmark isolates the effects of equipment and ensures the accuracy of the rate of change calculation.

[0117] S33013: The airflow variation is calculated and analyzed by applying a fluid dynamics flow model combined with reference wind speed data and a three-dimensional mesh model of the location of equipment. The airflow variation includes the windward stagnation coefficient, the lateral flow splitting coefficient, and the wake attenuation coefficient. The airflow wind speed vector is calculated based on the airflow variation. The airflow wind speed vector includes the windward wind speed vector, the lateral wind speed vector, and the wake wind speed vector.

[0118] It should be noted that the above-described embodiments of this application analyze and calculate airflow changes by considering the flow obstructed by equipment in the current work area within a three-dimensional mesh model of the equipment location based on reference wind speed data. Specifically, this embodiment applies reference wind speed data to a three-dimensional mesh model of the permanent equipment to calculate airflow changes. The windward stagnation coefficient is calculated as 1 - (equipment width / airflow direction projection length), the lateral splitting coefficient is calculated as arctan(equipment length / characteristic distance), and the wake attenuation coefficient is calculated as free velocity × e^(-downstream distance / turbulence scale). In the calculation of airflow changes, the equipment width, equipment length, and characteristic distance are all length distances within the three-dimensional mesh model of the equipment location. The characteristic distance is the distance from the equipment edge to the mesh; the turbulence scale is the product of the equipment width and the turbulence coefficient (usually set to 0.7).

[0119] Furthermore, the wind speed in each area of ​​the device is calculated using the airflow change, which includes the windward wind speed vector, the lateral wind speed vector, and the wake wind speed vector. The windward wind speed vector = reference wind speed × windward stagnation coefficient; the lateral wind speed vector = reference wind speed × lateral flow splitting coefficient; and the wake wind speed vector = reference wind speed × wake attenuation coefficient.

[0120] Through the implementation of the above embodiments of this application, the obstruction, diversion and attenuation effects of the device on the airflow are quantified, and a flow physics model is established.

[0121] S33014: Calculate the spatial gradient of wind speed around equipment in the current work area based on airflow wind speed vector;

[0122] It should be noted that in the above embodiments of this application, the wind speed spatial gradient reflects the degree of abrupt change in wind speed around the equipment, including the wind direction gradient, which quantifies the rate of kinetic energy loss caused by equipment obstruction (wake deceleration effect), the lateral gradient, which characterizes the lateral diffusion intensity and vertical gradient of the airflow around it, and reflects the lifting / pressurizing effect of the equipment on the vertical airflow; it provides a spatial differential basis for the vector change rate, reflecting the structured influence of the equipment on the local flow field.

[0123] S33015: The rate of change of wind speed vector is calculated by acceleration based on the wind speed spatial gradient and airflow wind speed vector;

[0124] It should be noted that in the above embodiments of this application, in fluid mechanics, the rate of change of wind speed vector (i.e., acceleration) is described by the mass derivative. Considering the quasi-steady-state characteristics of the current work site (the flow field structure is stable in a short time), the local time variation term can be ignored. Therefore, the rate of change of wind speed vector is simplified to migration acceleration. Based on this, the embodiments of this application combine the wind speed vector, including the windward wind speed vector, the lateral wind speed vector, and the wake wind speed vector, into a three-dimensional vector. Then, the wind speed vector rate of change is obtained by matrix operation using the wind speed spatial gradient and the three-dimensional vector, which accurately describes the instantaneous acceleration / deceleration behavior of the airflow when passing through the equipment. Furthermore, the change in the rate of change of airflow wind speed vector caused by the equipment helps to predict the trajectory of particulate matter.

[0125] In summary, the pollution monitoring and air supply control method and storage medium for the head-mounted respiratory protection system proposed in this invention collects real-time pollutant time-series data by deploying multiple distributed multispectral sensors and transmits it to the central processing unit of the head-mounted respiratory protection system. This ensures accurate and real-time pollutant distribution data for each wearer, providing real-time basis for the formulation of subsequent pollutant control and protection plans. Furthermore, by using Fast Fourier Transform (FFT) to process the pollutant time-series data, the dominant frequency characteristics of the pollutants are extracted, and the average particle size of the pollutants is calculated. Further, by combining the average particle size and diffusion characteristics of the pollutants with the fluid constraints of the respiratory protection system, the airflow of each air intake channel of the head-mounted respiratory protection system is dynamically allocated using the Particle Swarm Optimization (PSO) algorithm. Based on the characteristics of pollutants with different particle sizes and the capabilities of the respiratory protection system, the airflow allocation is optimized to achieve the lowest energy consumption while ensuring the best pollutant removal effect. Based on the optimized airflow value, control commands are generated to adjust the opening of the air valves in each air intake channel, precisely controlling the airflow allocation of each air intake channel and protecting the health of workers in the current workplace.

[0126] In the specific implementation process, this embodiment uses a particle swarm optimization algorithm combined with the average particle size and sedimentation information of pollutants to allocate the air volume value of each air intake channel; and, in particular, by analyzing the spatial weight between each two adjacent local location areas, it avoids analyzing each area in isolation and considers the mutual influence of pollutant diffusion in neighboring areas; on the other hand, for each local location area, the cross-second sensitivity between channels is calculated to determine the impact of each channel on neighboring channels when the air volume value changes, thereby accurately analyzing the pollutant diffusion coefficient;

[0127] Furthermore, this embodiment converts turbulence intensity into a pollutant diffusion capacity index and combines it with spatiotemporal analysis to output a spatiotemporal distribution of pollutant concentration prediction, thereby obtaining the pollutant movement path. On the other hand, this embodiment also analyzes the wind speed vector change rate by considering the obstruction effect of equipment in the workplace on airflow, which helps to predict the movement trajectory of particulate matter.

[0128] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; those skilled in the art can modify the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for pollution monitoring and air supply control of a head-mounted respiratory protection system, characterized in that, The following steps are included: By setting up multiple distributed multispectral sensors in the current work site, time-series data of pollutants and air data are collected within the local location area of ​​each multispectral sensor within a time period. Fast Fourier Transform was applied to the pollutant time series data to extract the pollutant dominant frequency features, and the average particle size of the pollutants was calculated based on the pollutant dominant frequency features. Based on the average particle size of pollutants, combined with the pollutant diffusion dynamics and the fluid constraints of the respiratory protection system, the airflow configuration of each air intake channel of the head-mounted respiratory protection system currently worn by the user is performed through particle swarm optimization, and the airflow value of each air intake channel is obtained. Control commands are generated based on the airflow value of each air intake channel to adjust the opening of the air valves in each air intake channel of the head-mounted respirator.

2. The method for pollution monitoring and air supply control of a head-mounted respiratory protection system according to claim 1, characterized in that, The aforementioned pollutant time series data includes sulfide time series data, carbide time series data, and particulate matter time series data; the aforementioned air data includes pollutant particle density and wind speed vector in local areas.

3. The method for pollution monitoring and air supply control of a head-mounted respiratory protection system according to claim 2, characterized in that, Based on the average particle size of pollutants, combined with the pollutant diffusion kinetics and fluid constraints of the respiratory protection system, particle swarm optimization is used to configure the airflow of each air intake channel of the currently worn head-mounted respiratory protection system, obtaining the airflow value of each air intake channel. The process includes the following steps: Initialize the particle swarm parameters, which include the particle position vectors corresponding to the N airflow distribution schemes for each air intake channel. The baseline total air volume value is determined based on the average particle size of the pollutants. Based on the pollutant density, average particle size, and air viscosity in each local area, the pollutant settling velocity in each local area is calculated using Stokes' law. Based on pollutant settling velocity, baseline total air volume, and historical data, we analyze the pollutant diffusion influence coefficient caused by each particle position vector and the comprehensive fitness corresponding to the pollutant diffusion influence coefficient. The iteration ends when the overall fitness reaches the fitness threshold or the number of iterations reaches the iteration threshold. The air volume value allocation scheme corresponding to the overall fitness is output to obtain the air volume value of each air intake channel. Otherwise, the multiple particle position vectors with high fitness are subjected to cross mutation processing to generate N' new particle position vectors corresponding to the air volume value allocation scheme of each air intake channel. The above pollutant settling velocity calculation processing operation is returned until the air volume value of each air intake channel is output.

4. The method for pollution monitoring and air supply control of a head-mounted respiratory protection system according to claim 3, characterized in that, The analysis of pollutant diffusion impact coefficients caused by each particle position vector and the corresponding comprehensive fitness coefficients based on pollutant settling velocity, baseline total air volume, and historical data includes the following operational steps: The pollutant movement path is predicted by numerical integration based on the pollutant settling velocity, the baseline total air volume, and the turbulent diffusion effect. Based on historical data analysis, the first sensitivity of the baseline total air volume value to the pollutant trajectory is obtained, and the trajectory change rate of the current baseline total air volume value to the pollutant movement path is obtained within a preset time period. Based on the trajectory change rate of pollutant movement paths, the diffusion sensitivity integral algorithm is used to generate the pollutant diffusion influence coefficient caused by the particle position vector corresponding to the air volume value allocation scheme of each air intake channel under the current baseline total air volume value. The overall fitness is calculated based on the pollutant diffusion influence coefficient and the particle position vector corresponding to the air volume distribution scheme of each air intake channel.

5. The method for pollution monitoring and air supply control of a head-mounted respiratory protection system according to claim 4, characterized in that, Based on the rate of change of the pollutant movement path trajectory, a diffusion sensitivity integral algorithm is used to generate the pollutant diffusion influence coefficient caused by the particle position vector corresponding to the airflow value allocation scheme of each intake channel under the current baseline total airflow value. The steps include the following: Obtain the center point of each local location region, calculate the regional Euclidean distance between each two adjacent local location regions based on the center point, and generate the spatial weight between regions based on the spatial attenuation coefficient obtained by combining the regional Euclidean distance with the measured airflow turbulence vortex scale analysis of the current work site. For each of the N airflow allocation schemes corresponding to each local location region, analyze the second sensitivity of channel p when the airflow of channel k changes with channel p. Based on the second sensitivity, establish the global channel coupling matrix. The initial diffusion influence coefficient vector is obtained by analyzing the continuous trajectory changes based on the global channel coupling matrix and inter-regional spatial weights, combined with the data based on a preset time window. The target pollutant diffusion influence coefficient is obtained by combining the regional average wind speed collected by a multispectral sensor within a local area over a preset time period with the initial diffusion influence coefficient vector and the pollutant settling velocity.

6. The method for pollution monitoring and air supply control of a head-mounted respiratory protection system according to claim 5, characterized in that, The initial diffusion influence coefficient vector is obtained by analyzing the global channel coupling matrix and inter-regional spatial weights in conjunction with the continuous trajectory changes based on a preset time window. The steps include the following: The second sensitivity fusion coefficient between each intake channel in each of two adjacent local location regions is calculated based on the global channel coupling matrix and the spatial weight between regions. Collect the continuous trajectory change rate of each air intake channel within a preset time window, and calculate the average trajectory change rate of each air intake channel within the time window; The mutual influence correction between intake channels is calculated using the global channel coupling matrix and the second sensitivity fusion coefficient. The initial diffusion influence coefficient vector is calculated by combining the mean trajectory change rate with the mutual influence correction between the intake channels.

7. A method for pollution monitoring and air supply control of a head-mounted respiratory protection system according to claim 6, characterized in that, The target pollutant diffusion influence coefficient is obtained by combining the regional average wind speed collected by multispectral sensors within a local area over a preset time period with the initial diffusion influence coefficient vector and the pollutant deposition velocity. The average wind speed in a local area is collected based on a multispectral sensor over a preset time period. The turbulence correction factor is calculated by combining the pollutant settling velocity, the regional average wind speed, and the initial diffusion influence coefficient vector. Determine if the pollutant settling velocity exceeds the pollutant settling velocity threshold; if so, trigger the sedimentation compensation correction strategy. The excess of pollutant settling velocity is calculated, and a compensation value is generated by exponential decay using the excess of pollutant settling velocity. The target pollutant diffusion influence coefficient is calculated based on the initial diffusion influence coefficient vector, turbulence correction factor, and compensation value.

8. The method for pollution monitoring and air supply control of a head-mounted respiratory protection system according to claim 2, characterized in that, Based on the pollutant density, average particle size, and air viscosity in each local area, the pollutant settling velocity in each local area is calculated using Stokes' law. The pollutant movement path is then predicted through numerical integration based on the settling velocity, baseline total airflow, and turbulent diffusion effects, including the following steps: After dividing each local area into multiple three-dimensional grid blocks, air data and average particle size of pollutants are collected for each grid block. Based on the equipment location and air data in the current work site, the wind speed vector change rate caused by equipment obstruction is analyzed. Extract the temperature and air viscosity data of the current work site; apply Stokes-Cunningham correction to the air data, average particle size of pollutants, and wind speed vector change rate of each grid block, and then perform principal component extraction to obtain the pollutant settling velocity of the current local area. Collect historical site data for a preset time period, calculate the wind speed standard deviation based on the historical site data, extract the grid wind speed of each grid block in the current local location area, and calculate the average grid wind speed of the current local location area based on the grid wind speed. The grid turbulence intensity is calculated based on the average grid wind speed and the standard deviation of the wind speed. The grid turbulence diffusion coefficient is calculated based on the grid turbulence intensity; principal component extraction is performed based on the grid turbulence diffusion coefficients of each grid block within the current local location region to obtain the regional turbulence diffusion coefficients within the current local location region. For each grid block, the pollutant concentration time series variation data is analyzed based on the regional turbulent diffusion coefficient; based on the pollutant concentration time variation data, the pollutant diffusion process is analyzed based on the pollutant settling velocity of the current local area to obtain the pollutant concentration prediction data; The pollutant movement path is obtained by linear fitting based on the pollutant concentration prediction data.

9. A method for pollution monitoring and air supply control of a head-mounted respiratory protection system according to claim 8, characterized in that, Pollutant concentration prediction data is obtained by analyzing the pollutant diffusion process based on the pollutant deposition velocity in the current local area, using pollutant concentration change data over time. The process includes the following steps: Based on the law of conservation of mass and Fick's diffusion law, a pollutant transport control equation is constructed for each grid block; The finite volume method is used to transform the continuous pollutant transport control equations into a discrete grid computing model. Based on the time-varying data of pollutant concentration, the grid computing model is subjected to time integration to obtain the predicted pollutant concentration data.

10. A storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the steps of the pollution monitoring and air supply control method for a head-mounted respiratory protection system as described in any one of claims 1-9.

Images