Method, system and device for monitoring dust exposure level of personnel in tunnel construction area

Abstract

The application provides a tunnel construction area personnel dust exposure level monitoring method, system and equipment, the method comprises the following steps: acquiring the measured dust concentration data of a plurality of monitoring points, and preprocessing the measured dust concentration data to obtain the preprocessed measured dust concentration data; based on the preprocessed measured dust concentration data, the peak value estimation value of the dust concentration is obtained; based on the peak value estimation value and the preprocessed measured dust concentration data, the dust concentration spatial distribution model is used to obtain the dust concentration distribution data of the construction area; based on the position data of the workers in the construction area, the dust concentration distribution data of the construction area is combined to calculate the cumulative dust exposure of the workers. The scheme realizes the prediction of the dust concentration distribution of the entire construction area from the dust concentration of a few monitoring points, reduces the layout density of the sensor, thereby increasing the feasibility of the scheme, and the accuracy of the scheme for predicting the dust concentration distribution is high, and the method is convenient.

Description

Technical Field

[0001] This invention relates to the fields of neural network learning, environmental monitoring, and dust exposure early warning, and particularly to a method, system, and equipment for monitoring personnel dust exposure levels in tunnel construction areas based on spatiotemporal distribution feature learning. Background Technology

[0002] Tunnel construction generates a large amount of dust, and long-term exposure to high concentrations of dust can lead to occupational diseases such as pneumoconiosis. Therefore, monitoring dust exposure levels among workers in tunnel construction areas is of great importance. Currently, there are two main methods for monitoring dust exposure levels among workers: fixed-point monitoring in key areas and individual monitoring.

[0003] Fixed-point monitoring primarily targets key work areas, using dust sensors or samplers to monitor dust concentrations in the environment. While this method offers some convenience, data from fixed locations or time-averaged data struggles to reflect spatiotemporal distribution and evolution patterns. This can lead to biases when calculating dust exposure levels for workers using fixed-point monitoring data, making it difficult to guarantee the reliability of the results.

[0004] Individual monitoring targets workers by attaching personal dust samplers to their bodies. After each shift, the sampler is removed, and the dust concentration is calculated using a filter weighing method. This method offers strong tracking capabilities and accurately reflects workers' dust exposure levels. However, the measurement process is cumbersome and highly susceptible to human error, making continuous monitoring impossible and hindering its widespread practical application.

[0005] While methods exist for studying the spatial distribution of dust concentration using numerical simulations and combining this with on-site dust sensor monitoring data to determine the regional distribution of dust concentration and thus the dust exposure level of workers, most remain at the theoretical research stage. These methods suffer from two main problems: first, the integration of sensor data with the spatial distribution of dust concentration is problematic; the models used are relatively coarse and fail to accurately reflect the intrinsic correlation between actual monitoring data and the spatial distribution of dust concentration; second, for long construction areas, such as tunnel construction, a certain sensor deployment density is required to ensure the reliability of the obtained data, but the increased cost is difficult to resolve. Summary of the Invention

[0006] In view of this, in order to at least partially solve the above-mentioned problems existing in the prior art, and to realize the real-time reflection of the distribution and changes of dust in the tunnel construction area through a small number of dust monitoring points, to measure the occupational exposure health risks of workers to dust, and to provide early warning for personnel with high cumulative dust exposure, the present invention provides the following technical solution:

[0007] On the one hand, the present invention provides a method for monitoring the dust exposure level of personnel in tunnel construction areas, the method comprising:

[0008] S1. Obtain measured dust concentration data from multiple monitoring points, and preprocess the measured dust concentration data to obtain preprocessed measured dust concentration data.

[0009] S2. Based on the preprocessed measured dust concentration data, obtain the peak value of dust concentration.

[0010] S3. Based on the peak value estimate and the pre-processed measured dust concentration data, the dust concentration distribution data of the construction area is obtained through the dust concentration spatial distribution model.

[0011] S4. Based on the location data of workers in the construction area and the dust concentration distribution data of the construction area, calculate the cumulative dust exposure of workers.

[0012] Preferably, the number of monitoring points is 5.

[0013] Preferably, in step S2, the peak estimate is obtained by calculating a peak prediction model;

[0014] The peak prediction model is established as follows:

[0015] S21. Obtain dust concentration data at different time points from multiple monitoring points, and preprocess the dust concentration data to form a basic database.

[0016] S22. Filter and standardize the data in the basic database to form standardized data;

[0017] S23. Establish a fitting function based on standardized data:

[0018]

[0019] Indicates a position at a certain moment. Dust concentration at the location, Indicates characteristic parameters;

[0020] S24. Based on Gaussian regression, establish the relationship between standardized data and dust concentration peak value, thereby enabling the prediction of dust concentration peak value based on standardized data to obtain the peak value estimate of dust concentration.

[0021] Preferably, in step S3, the method for obtaining the dust concentration distribution data of the construction area is as follows:

[0022] S31. Based on Gaussian regression, establish the relationship between standardized data, peak estimates, and characteristic parameters p1, p2, p3, q1, and q2 to form a spatial distribution model of dust concentration:

[0023]

[0024] in, They are respectively The estimated value, This represents the peak estimate;

[0025] S32. Based on the spatial distribution model of dust concentration and the fitting function, the characteristic parameters corresponding to the measured dust concentration data are solved in reverse.

[0026] S33. Based on the characteristic parameters corresponding to the reverse solution and the measured dust concentration data, the dust concentration distribution data of the construction area is obtained.

[0027] Preferably, step S33 further includes:

[0028] Substituting the relationship between position x and the feature parameters into the fitting function, the feature parameters corresponding to the measured dust concentration data are solved in reverse. The relationship between position x and the feature parameters is as follows:

[0029] .

[0030] Preferably, the Gaussian regression is achieved by establishing a Gaussian regression model:

[0031] First, establish the Gaussian process: Where the random variable is the function at position a. The value; The defined mean function; Let covariance function be used. For hyperparameters;

[0032] Secondly, training output With predicted output The joint distribution is:

[0033]

[0034] Predicted output The distribution under the observed values ​​is as follows: Where K represents the covariance, For training input, For predicting input;

[0035] Secondly, utilize Solving for the maximum hyperparameter ;

[0036] Finally, the hyperparameters obtained from the solution will be... Substitute into the predicted output The distribution under the observed values ​​yields the predicted output. The mean is The variance is .

[0037] Preferably, the cumulative dust collection amount is calculated as follows:

[0038]

[0039] Where CDE represents the cumulative dust exposure of the workers, Q is the average respiratory rate per minute, and t is time. This is a correction factor for breathing intensity. This is an estimate of the dust concentration at the location of the worker, obtained from dust concentration distribution data of the construction area.

[0040] Preferably, the method further includes:

[0041] S5. Based on the cumulative dust exposure of workers and in conjunction with occupational dust exposure restriction requirements, issue early warning information.

[0042] Preferably, the preprocessing in S1 includes:

[0043] S11. Monitor the dust cloud map based on the measured dust concentration data, and take the plane where the breathing zone is located in the dust cloud map as a cross section, and then grid the cross section.

[0044] S12. Calculate the average dust concentration at each grid point as the grid dust concentration value;

[0045] S13. Apply window filtering to the dust concentration value of the grid;

[0046] S14. Standardize the grid dust concentration values ​​after window filtering to obtain preprocessed measured dust concentration data.

[0047] On the other hand, the present invention also provides a dust exposure level monitoring system for personnel in tunnel construction areas, the system comprising:

[0048] The monitoring point module is deployed within the tunnel construction area to obtain measured dust concentration data.

[0049] The basic database module is used to preprocess historical dust concentration data to form a basic database;

[0050] The peak prediction module establishes a peak prediction model based on the basic database module, and uses measured dust concentration data as input values ​​to obtain peak estimates.

[0051] The dust concentration spatial distribution module, based on the basic database module and peak estimation values, establishes a dust concentration spatial distribution model to obtain dust concentration distribution data in the construction area;

[0052] The worker positioning module is used to acquire worker location data;

[0053] The early warning module is used to calculate the cumulative dust exposure of workers based on the location data of workers and the dust concentration distribution data of the construction area obtained by the dust concentration spatial distribution module, and to issue early warning information based on the cumulative dust exposure.

[0054] Preferably, the preprocessing includes:

[0055] First, monitor the dust cloud map based on the measured dust concentration data, and take the plane where the breathing zone is located in the dust cloud map as a cross section, and then grid the cross section;

[0056] Second, calculate the average dust concentration at each grid point as the grid dust concentration value;

[0057] Third, window filtering is applied to the dust concentration value of the grid.

[0058] Fourth, the dust concentration values ​​of the grid after window filtering are standardized to obtain the pre-processed measured dust concentration data.

[0059] Preferably, after the cross-section is meshed, interpolation is used to fill in the missing dust concentration data.

[0060] In another aspect, the present invention also provides a device for monitoring the dust exposure level of personnel in a tunnel construction area. The device includes at least a memory and a processor. The processor calls instructions stored in the memory to execute the above-mentioned method for monitoring the dust exposure level of personnel in a tunnel construction area.

[0061] Compared with the prior art, the technical solution of the present invention has at least the following beneficial effects:

[0062] This paper extracts the spatial distribution characteristics of dust concentration during tunnel construction from validated numerical simulation results, establishes a predictive model for the spatial distribution characteristics of dust concentration, and enables the prediction of the dust concentration distribution of the entire construction area from a relatively small number of monitoring points (e.g., 5 monitoring points). This reduces the sensor deployment density and increases the feasibility of the scheme. Furthermore, this scheme improves the predictive model for the spatial distribution characteristics of dust concentration by utilizing the constraint relationships between characteristic parameters, thereby increasing the accuracy of the predicted dust concentration distribution. Simultaneously, the obtained cumulative dust exposure can provide a decision-making basis for the occupational classification management of pneumoconiosis and a data foundation for the research of pneumoconiosis evolution models. Attached Figure Description

[0063] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0064] Figure 1 This is a schematic diagram of the process for obtaining the spatiotemporal distribution model of dust concentration in a tunnel construction area according to an embodiment of the present invention;

[0065] Figure 2 This is a schematic diagram illustrating the process of calculating the cumulative dust exposure of workers according to an embodiment of the present invention.

[0066] Figure 3 This is a schematic diagram illustrating the spatial variation of four sets of axial dust concentrations according to an embodiment of the present invention;

[0067] Figure 4 This is a partial dust concentration distribution curve after dimensionality reduction and feature extraction of the simulation results in an embodiment of the present invention;

[0068] Figure 5 This is a predicted effect diagram of dust concentration distribution in the construction area 12 minutes after tunnel blasting, according to an embodiment of the present invention. Detailed Implementation

[0069] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be understood that the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0070] Those skilled in the art should understand that the following specific embodiments or implementation methods are a series of optimized configurations listed to further explain the specific content of the invention. These configuration methods can be combined or used in conjunction with each other, unless the invention explicitly states that some or a specific embodiment or implementation method cannot be associated with or used in conjunction with other embodiments or implementation methods. Furthermore, the following specific embodiments or implementation methods are merely optimized configurations and are not intended to limit the scope of protection of the invention.

[0071] In one specific embodiment, combined with Figure 1 , Figure 2 The method for monitoring personnel dust exposure levels in tunnel construction areas provided by this invention can be achieved in the following ways:

[0072] Step 1: Using on-site monitoring data and CFD simulation technology, obtain spatiotemporal distribution cloud maps of dust concentration under single-peak dust source, double-peak dust source, and discrete low-intensity dust source scenarios during tunnel construction. Dimensionality reduction of the original data is performed using high-dimensional array mean squared method to establish a basic database.

[0073] Step 2: Use a rational function with 5 parameters to extract the spatial distribution characteristic parameters of dust concentration in the simulation results, and extract the spatiotemporal distribution characteristic parameters of dust concentration through peak finding algorithm and rational function fitting model;

[0074] Step 3: Determine the location of monitoring points in the tunnel based on the accuracy of model prediction. Combine monitoring point data, dust peak concentration, and characteristic parameters to construct a machine learning sample set. Use Gaussian process regression based on exponential kernel function to establish and train a regression model of dust concentration and dust peak concentration at the monitoring point to form a dust peak concentration prediction model. At the same time, establish a regression model of dust concentration at the monitoring point and five characteristic parameters through Gaussian process regression to form a dust concentration spatial distribution characteristic parameter prediction model.

[0075] Step 4: Based on rational functions, determine the intrinsic constraint relationship of the five characteristic parameters, optimize the prediction model of the spatial distribution characteristic parameters of dust concentration, and obtain the spatial distribution characteristic parameters of dust concentration, as well as the peak dust concentration corresponding to these characteristic parameters.

[0076] Step 5: Using the actual dust concentration at the monitoring point as input, predict the spatial distribution of dust concentration. Combined with the personnel positioning system, obtain the dust concentration at the location of the workers to obtain the dust exposure level of the workers. Issue an early warning based on the occupational exposure limits for hazardous factors in the workplace.

[0077] In a more preferred embodiment, the construction of the basic database in step one is carried out in the following manner:

[0078] Combining tunnel construction procedures with on-site monitoring data from blasting, muck removal, and shotcreting, and considering actual working conditions, continuous three-dimensional high-resolution dust cloud maps are obtained through CFD simulation. The specific calculation method for the dust cloud maps can utilize existing technologies and will not be elaborated here. For the generated dust cloud maps, we typically extract a plane at a height of 1.5m (i.e., the breathing zone), then mesh the cross-section and export the dust concentration data at each grid point. When meshing the cross-section, we can use, for example, a rectangular grid with a length of 1m and a width of 0.5m; the grid size can be set according to actual needs. For the meshed cross-sectional area, we use algorithms such as Kriging interpolation to fill in missing data values, calculating the average dust concentration at each grid point in sequence. The average dust concentration at each grid point is stored as the grid dust concentration value at that moment, forming a basic database. This basic database mainly stores the dust concentration data of each grid at different times. We can further set this basic database as a two-dimensional matrix, with the first row being the horizontal coordinate of the grid points, and the other rows storing the dust concentration values ​​at the corresponding grid points at different times.

[0079] After establishing the basic database, we perform dimensionality reduction on the dust concentration data in the database and extract the spatial distribution features of dust concentration, which is step two. Here, preferably, we first filter the average data in the database using a windowed moving average method. The filtering operation can be repeated several times; for example, we choose to repeat the filtering operation twice. The window value can be set to a length of 3 (i.e., the window length is 3 grid units). After that, we standardize the filtered data, and the standardization process can use algorithms such as Z-Score standardization.

[0080] For the standardized data, we perform data fitting. When judging the fitting effect, we compare the root mean square error and correlation coefficient obtained from different fitting models and select the model with the better fitting effect. In a preferred embodiment, we use the function model in equation (1) to fit the data and extract feature parameters. The peak-finding algorithm is used to find and record the maximum dust concentration in the fitting results. and its location For peak finding algorithms, we can use existing techniques such as the derivative method, which will not be elaborated here.

[0081] (1)

[0082] In equation (1), Indicates a position at a certain moment. The dust concentration at a given location is expressed in mg / m³. 3 .

[0083] Subsequently, in step three, based on the calculation results from step two, we determine the placement of monitoring points within the tunnel based on the accuracy of the model predictions. In a preferred embodiment, a total of five monitoring points are typically deployed: one on each of the excavation trolley and the secondary lining trolley, and three evenly distributed on the tunnel wall between the excavation and secondary lining trolleys. Of course, the placement of these monitoring points can be adjusted as needed, or based on the tunnel excavation progress. The monitoring points on the excavation trolley are positioned closer to the tunnel exit, while those on the secondary lining trolley are positioned closer to the tunnel face. The air inlets of the dust monitors all face the windward side. The height of the monitoring points above the ground is the breathing zone height, for example, set to 1.5m.

[0084] In step three, the process of establishing the spatial distribution model of dust concentration based on monitoring data from limited locations is as follows:

[0085] First, establish a Gaussian process regression model:

[0086] Establishing a Gaussian process In this Gaussian process, the random variable is the function at position a. The value; The defined mean function; The covariance function is chosen here, specifically the exponential kernel function, i.e. , Let be the hyperparameter to be solved. To simplify the calculation, generally let Set the mean function to 0.

[0087] Suppose there exists a set of observations , For training input, To predict the input, the training output is then... With predicted output The joint distribution is .

[0088] Then we can obtain the predicted output. The distribution under the observed values ​​is as follows: , where K represents the covariance.

[0089] We can solve for the hyperparameters using the marginal log-likelihood estimate, i.e., find the parameter that makes the hyperparameters equal to the marginal log-likelihood estimate. Maximum hyperparameter , here .

[0090] The hyperparameters obtained by the solution Substitute into the predicted output The distribution under the observed values ​​yields the predicted output. The mean is The variance is .

[0091] Create dataset 1: This is used to study the internal relationship between dust concentration and peak dust concentration at monitoring points. The dust concentration at five monitoring points was used as the input variable. The peak dust concentration is used as the response variable. t represents the sample size; the dust concentration at other detection points is expressed in the same way as... The same applies, so I won't repeat myself.

[0092] Using the Gaussian process regression model established above as the basis for the dust concentration peak prediction model, with dataset 1 as the sampling space, and selecting the exponential kernel as the covariance function (as given above), the optimal hyperparameters are found by maximizing the marginal log-likelihood, thus obtaining the regression model between dust concentration data and peak concentration at the 5 monitoring points. ,in This represents an estimate of the peak dust concentration.

[0093] Secondly, a dataset 2 was established to learn the relationship between dust concentration and spatial distribution characteristics at monitoring points: ,in As input variables, The parameter representing the spatial distribution characteristics of dust concentration is used as a response variable.

[0094] A prediction model for the spatial distribution characteristic parameters of dust concentration is established using the previously established Gaussian process regression model. The sampling space is dataset 2, and the exponential kernel is selected as the covariance function (this function has been given above). The optimal hyperparameter is found by maximizing the marginal log-likelihood, and the relationship model between dust concentration at the monitoring point and the spatial distribution characteristic parameters of dust concentration is obtained:

[0095] (2)

[0096] In equation (2), They are respectively The estimated value.

[0097] After obtaining the regression model of dust concentration data and peak concentration at the monitoring points, and the relationship model between dust concentration at the monitoring points and the spatial distribution characteristics of dust concentration, in step four, we optimize the prediction model for the spatial distribution characteristic parameters of dust concentration:

[0098] Based on the rational function form used to extract the spatial distribution characteristics of dust concentration We can obtain the following relational equation:

[0099] (3)

[0100] In formula (3): This represents the peak dust concentration. This indicates the location of the peak concentration.

[0101] Mathematical form of rational functions based on the extraction of spatial distribution characteristics of dust concentration Combined with peak dust concentration It can be known that The maximum value is consistent with the peak value of dust concentration, that is... .and The maximum value has a theoretical solution. The theoretical derivation of the peak value is as follows:

[0102] Differentiate g(x):

[0103]

[0104] make ,but

[0105]

[0106] but for Two local extrema.

[0107] Assuming the peak dust concentration is The location of the peak is Then the parameter Satisfy the following equations:

[0108] (4) or (5)

[0109] The specific expressions for x1 and x2 in the above equation are as follows:

[0110]

[0111] Based on the actual dust distribution, we discard the left root x2 and retain the right root x1, that is, we choose equation (4) as the constraint equation, which is the final equation. The solution method is as follows: .

[0112] we will Substituting the expression into relational equation (3), we can obtain The constraint equations between them are obtained using a dust concentration spatial distribution characteristic parameter prediction model. Then use the constraint equations to calculate Therefore, we can obtain the characteristic parameters, and then we can obtain the corresponding peak dust concentration.

[0113] Based on the prediction model established above, in step five, we can use the measured values ​​at the monitoring points to predict the spatial distribution of dust concentration, and combine this with the location of personnel to predict their dust exposure, and then conduct further assessment and warning measures. In a preferred embodiment, this can be achieved in the following way:

[0114] First, obtain the dust concentration at the actual monitoring point, input it into the relationship model between the dust concentration at the monitoring point and the spatial distribution characteristics of dust concentration, calculate the characteristic parameters, and then obtain the location x of the worker according to the personnel positioning system, and use formula (1) to calculate the estimated value of the dust concentration at location x.

[0115] Considering the varying labor intensity of different jobs, formula (6) is used to calculate the dust exposure of workers:

[0116] (6)

[0117] In formula (6), CDE represents the cumulative dust exposure of the worker, with the unit set to mg; Q is the average respiratory volume of the human body per minute, with the unit set to L / min; t is time, min; This is a correction factor for breathing intensity. This is an estimate of the dust concentration at the location of the personnel.

[0118] The present invention extracts the spatial distribution characteristics of dust concentration during tunnel construction from validated numerical simulation results. A Gaussian process regression model is then used to predict the spatial distribution characteristics of dust concentration, enabling the prediction of dust concentration distribution across the entire construction area from five monitoring points. This reduces the density of sensor deployment and increases the feasibility of the solution. Furthermore, the model is improved by utilizing the constraint relationships between characteristic parameters, enhancing the accuracy of dust concentration prediction. The obtained cumulative dust exposure data can provide a basis for decision-making regarding occupational classification management of pneumoconiosis and a data foundation for research on pneumoconiosis evolution models.

[0119] In yet another embodiment, the solution of the present invention can be implemented through a dust exposure level monitoring system for personnel in tunnel construction areas. Preferably, the system includes:

[0120] The monitoring point module is deployed within the tunnel construction area to obtain measured dust concentration data.

[0121] The basic database module is used to preprocess historical dust concentration data to form a basic database;

[0122] The peak prediction module establishes a peak prediction model based on the basic database module, and uses measured dust concentration data as input values ​​to obtain peak estimates.

[0123] The dust concentration spatial distribution module, based on the basic database module and peak estimation values, establishes a dust concentration spatial distribution model to obtain dust concentration distribution data in the construction area;

[0124] The worker positioning module is used to acquire worker location data;

[0125] The early warning module is used to calculate the cumulative dust exposure of workers based on the location data of workers and the dust concentration distribution data of the construction area obtained by the dust concentration spatial distribution module, and to issue early warning information based on the cumulative dust exposure.

[0126] Preferably, the preprocessing includes:

[0127] First, monitor the dust cloud map based on the measured dust concentration data, and take the plane where the breathing zone is located in the dust cloud map as a cross section, and then grid the cross section;

[0128] Second, calculate the average dust concentration at each grid point as the grid dust concentration value;

[0129] Third, window filtering is applied to the dust concentration value of the grid.

[0130] Fourth, the dust concentration values ​​of the grid after window filtering are standardized to obtain the pre-processed measured dust concentration data.

[0131] Preferably, after the cross-section is meshed, interpolation is used to fill in the missing dust concentration data.

[0132] In another specific embodiment, we will further illustrate the implementation of the present invention by taking into account a specific example of tunnel monitoring and construction.

[0133] I. Hardware Installation and Debugging

[0134] First, the actual locations of the monitoring points need to be determined. In this example, based on the actual site conditions, we set up five monitoring points, one on the excavation trolley and one on the secondary lining trolley. The monitoring points on the excavation trolley face the tunnel exit, and the monitoring points on the secondary lining trolley face the tunnel face. The air inlets of the dust monitors all face the windward side. Three monitoring points can be evenly distributed on the tunnel wall between the excavation trolley and the secondary lining trolley. The height of the monitoring points is the breathing zone height, and the approximate distances of the five monitoring points from the tunnel face are 40m, 90m, 160m, 230m, and 300m.

[0135] The personnel positioning equipment is installed with the positioning base station installed on the wall at the tunnel exit, about 1.5m above the ground. The positioning tag is sewn into the protective clothing of the workers or carried by them. The positioning system obtains the distance between the tag and the base station.

[0136] The gateway is mainly used to receive and forward data sent by dust sensors and positioning base stations. To ensure the reliability of data transmission, the gateway is installed on the secondary lining trolley, which is close to the construction area. The installation location is placed in a relatively stable area according to the actual situation.

[0137] Finally, the hardware devices and cloud server were debugged. By configuring the hardware devices, it was ensured that the data from the five dust sensors and personnel location information could be obtained normally by the cloud server.

[0138] II. Establishing a model relating actual measurement data to dust concentration distribution

[0139] Numerical simulation of dust transport patterns during tunnel blasting was used to obtain dust distribution data in the construction area within 60 minutes after blasting, with 1-minute time intervals, resulting in 60 sets of dust concentration distribution data. Assuming the direction along the tunnel exit is X, the direction connecting the two sides of the tunnel is Y, and the direction from the bottom to the top of the tunnel is Z, a breathing zone plane at a height of 1.5m was selected for analysis, i.e., Z=1.5m. Due to the presence of equipment in some areas, resulting in blank values ​​in the exported results, Kriging interpolation was used to fill these blanks without affecting the overall data. Figure 3 The distribution of dust concentration is shown when Y = 1m, 3m, 5m, and 7m. It can be seen that the dust concentration distribution along the X direction follows a consistent pattern, initially increasing and then decreasing. Therefore, to reduce computational complexity, the data along the Y direction is averaged, retaining only the X direction data. For data with significant fluctuations, filtering is applied. To ensure the data trend remains unchanged, a moving average with a window length of 3 is used for filtering, and this filtering operation is repeated twice. Finally, 60 sets of dust concentration distribution data along the X direction are obtained.

[0140] The processed simulation data was fitted. After comparing different fitting function forms, based on the fitting effect, we finally determined to fit the simulation results with a rational function containing 5 parameters as shown in equation (1). The function parameters were solved by nonlinear least squares method. For the fitted curve, RMSE and R were used. 2 The fitting effect was evaluated. Finally, the parameters of 60 fitted curves were obtained. As shown in Table 1 below.

[0141] (1)

[0142] In equation (1), Indicates a position at a certain moment. Dust concentration at the location, mg / m³ 3 .

[0143] Table 1. Record of Fitted Curve Parameters

[0144]

[0145] Using a peak-finding algorithm, the maximum dust concentration in the construction area at different times is found and recorded. and its location A total of 60 sets of data. Figure 4 The spatial distribution patterns of dust concentration at four different times are shown. Realdata refers to the simulated data after filtering, fitted data refers to the fitted characteristic curve, and peaks refers to the peak dust concentration at a certain time obtained by the peak finding algorithm.

[0146] Using the obtained patterns, the relationship between the dust concentration at actual monitoring points and the spatial distribution of dust concentration is sought. That is, the five parameters of the fitting function in equation (2) are calculated using the dust concentration at actual monitoring points. A Gaussian process regression model is established using MATLAB, and then machine learning is performed using the 60 sets of data obtained in the above steps to obtain a dust concentration peak prediction model. The peak dust concentration is then calculated when the monitoring data are c1, c2, c3, c4, and c5. Similarly, the monitoring point data and peak concentration are regressed with the function parameters to obtain a dust concentration spatial distribution characteristic parameter prediction model, which is used to calculate The five models are p1_Model, p2_Model, p3_Model, q1_Model, and q2_Model.

[0147] Currently, the dust concentration distribution in the construction area can be obtained by monitoring the dust concentration at actual monitoring points. Figure 5 This is a predicted image showing the dust concentration distribution in the construction area 12 minutes after tunnel blasting.

[0148] III. Calculation of Dust Collection and Early Warning

[0149] First, the dust concentration data from five monitoring points are input into the Predict_Peaks_Model to obtain the peak concentration at the current moment. Then, the already trained p1_Model, p2_Model, p3_Model, q1_Model, and q2_Model are used to calculate... And use constraint equations to Optimization calculations are performed. The position x0 of the worker from the tunnel exit is obtained through the personnel positioning system. The cumulative tunnel advance L provided by the tunnel construction information center is then retrieved to determine the distance between the worker and the tunnel face. Calculate the dust concentration c around the worker using equation (1). If a worker is at position The time spent there is If the average respiratory volume per minute is Q (L / min), then in Within a given time period, the surrounding dust concentration was The cumulative dust exposure of workers while working in the tunnel can be calculated using the following formula:

[0150] (7)

[0151] In formula (7), CDE represents the cumulative dust exposure of the worker, mg; Q is the human respiratory rate per minute, L / min, which is generally taken as 7 to 8; For the breathing intensity correction factor, a value of 1 to 3 is recommended; For time, in min.

[0152] Now, let's assume a scenario: 15 minutes after the blast, a worker walks from the secondary lining trolley (approximately 300m from the working face) to the invert arch (approximately 60m from the working face) to inspect the equipment. The inspection time is 5 minutes, and the walking speed is 1m / s. Calculations show that this process lasts from 15 minutes to 24 minutes after the blast. The specific process for calculating the dust exposure is as follows:

[0153] The relevant information obtained from the dust sensor and positioning system is shown in the table below:

[0154] Table 2 Dust sensor monitoring information and personnel location information

[0155]

[0156] Table 3 shows the calculation results of five parameters representing the dust concentration distribution characteristics, the dust concentration at the worker's location, and the amount of dust exposure, including the correction factor for breathing intensity when the worker is walking. Take 1.2 as the breathing intensity correction factor during maintenance. Take 1.5. The data calculated from the table shows that the worker's cumulative dust exposure during this period is 1.709 mg. The "Occupational Exposure Limits for Hazardous Factors in the Workplace Part 1: Chemical Hazardous Factors" (GBZ 2.1—2019) stipulates that for free SiO2 content between 10% and 50%, its time-weighted average permissible concentration shall not exceed 1 mg / m³. 3 (Total dust), 0.7 mg / m³ 3 (Dust exposure). Since the standard only provides the time-weighted average allowable concentration calculated on a daily basis, it is necessary to calculate the worker's cumulative dust exposure for one day, and then use the allowable concentration given in the standard to calculate the maximum cumulative dust exposure. The two are compared, and if the worker's cumulative dust exposure for the day exceeds the maximum cumulative dust exposure, an alarm is issued.

[0157] Table 3. Dust Spatial Distribution Characteristics and Cumulative Dust Accumulation Calculation Table

[0158]

[0159] Furthermore, in another embodiment of the present invention, the solution can be implemented by a device, which may include corresponding modules that perform one or more steps in the various embodiments described above. Therefore, each or more steps in the various embodiments described above can be performed by a corresponding module, and the electronic device may include one or more of these modules. A module may be one or more hardware modules specifically configured to perform a corresponding step, or implemented by a processor configured to perform a corresponding step, or stored in a computer-readable medium for implementation by a processor, or implemented through some combination thereof.

[0160] This device can be implemented using a bus architecture. A bus architecture can include any number of interconnect buses and bridges, depending on the specific application of the hardware and overall design constraints. The bus connects various circuits, including one or more processors, memory, and / or hardware modules. The bus can also connect various other circuits such as peripherals, voltage regulators, power management circuitry, external antennas, etc.

[0161] A bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Component Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, this diagram uses only one connection line, but this does not imply that there is only one bus or one type of bus.

[0162] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of this solution includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which the embodiments of this solution pertain. The processor performs the various methods and processes described above. For example, the method embodiments of this solution can be implemented as software programs tangibly contained in a machine-readable medium, such as memory. In some embodiments, part or all of the software program can be loaded and / or installed via memory and / or a communication interface. When the software program is loaded into memory and executed by the processor, one or more steps of the methods described above can be performed. Alternatively, in other embodiments, the processor can be configured to perform one of the methods described above by any other suitable means (e.g., by means of firmware).

[0163] The logic and / or steps represented in the flowchart or otherwise described herein may be specifically implemented in any readable storage medium for use by, or in conjunction with, an instruction execution system, apparatus or device (such as a computer-based system, a processor-included system or other system that can fetch and execute instructions from, an instruction execution system, apparatus or device).

[0164] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0165] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for monitoring dust exposure levels of personnel in tunnel construction areas, characterized in that, The method includes: S1. Obtain measured dust concentration data from multiple monitoring points, and preprocess the measured dust concentration data to obtain preprocessed measured dust concentration data. S2. Based on the preprocessed measured dust concentration data, obtain the peak value of the dust concentration; the peak value is calculated using a peak prediction model; the peak prediction model is established as follows: S21. Obtain dust concentration data at different time points from multiple monitoring points, and preprocess the dust concentration data to form a basic database. S22. Filter and standardize the data in the basic database to form standardized data; S23. Establish a fitting function based on standardized data: Indicates a position at a certain moment. Dust concentration at the location, Indicates characteristic parameters; S24. Based on Gaussian regression, establish the relationship between standardized data and dust concentration peak value, thereby enabling the prediction of dust concentration peak value based on standardized data to obtain the peak value of dust concentration. S3. Based on the peak value estimate and the preprocessed measured dust concentration data, the dust concentration distribution data of the construction area is obtained through a dust concentration spatial distribution model; the method for obtaining the dust concentration distribution data of the construction area is as follows: S31. Based on Gaussian regression, establish the relationship between standardized data, peak estimates, and characteristic parameters p1, p2, p3, q1, and q2 to form a spatial distribution model of dust concentration: in, They are respectively The estimated value, This represents the peak estimate; S32. Based on the spatial distribution model of dust concentration and the fitting function, the characteristic parameters corresponding to the measured dust concentration data are solved in reverse. S33. Based on the characteristic parameters corresponding to the reverse solution and the measured dust concentration data, the dust concentration distribution data of the construction area is obtained; S4. Based on the location data of workers in the construction area and the dust concentration distribution data of the construction area, calculate the cumulative dust exposure of workers.

2. The method according to claim 1, characterized in that, Step S32 further includes: Substituting the relationship between position x and the feature parameters into the fitting function, the feature parameters corresponding to the measured dust concentration data are solved in reverse. The relationship between position x and the feature parameters is as follows: 。 3. The method according to claim 1, characterized in that, The Gaussian regression is achieved by establishing a Gaussian regression model: First, establish the Gaussian process: Where the random variable is the function at position a. The value; The defined mean function; Let covariance function be used. For hyperparameters; Secondly, training output With predicted output The joint distribution is: Predicted output The distribution under the observed values ​​is as follows: Where K represents the covariance, For training input, For predicting input; Secondly, utilize Solving for the maximum hyperparameter ; Finally, the hyperparameters obtained from the solution will be... Substitute into the predicted output The distribution under the observed values ​​yields the predicted output. The mean is The variance is .

4. The method according to claim 1, characterized in that, The method for calculating the cumulative dust collection amount is as follows: Where CDE represents the cumulative dust exposure of the workers, Q is the average respiratory rate per minute, and t is time. This is a correction factor for breathing intensity. This is an estimate of the dust concentration at the location of the worker, obtained from dust concentration distribution data of the construction area.

5. The method according to claim 1, characterized in that, The method further includes: S5. Based on the cumulative dust exposure of workers and in conjunction with occupational dust exposure restriction requirements, issue early warning information.

6. The method according to claim 1, characterized in that, The preprocessing in S1 includes: S11. Establish a dust cloud map based on the measured dust concentration data, and take the plane where the breathing zone is located in the dust cloud map as a cross section, and then grid the cross section. S12. Calculate the average dust concentration at each grid point as the grid dust concentration value; S13. Apply window filtering to the dust concentration value of the grid; S14. Standardize the grid dust concentration values ​​after window filtering to obtain preprocessed measured dust concentration data.

7. A dust exposure level monitoring system for personnel in tunnel construction areas, characterized in that, The system is used to perform the method for monitoring personnel dust exposure levels in tunnel construction areas as described in claim 1, the system comprising: The monitoring point module is deployed within the tunnel construction area to obtain measured dust concentration data. The preprocessing module preprocesses the measured dust concentration data to obtain preprocessed measured dust concentration data. The basic database module is used to establish a basic database based on the preprocessed measured dust concentration data; The peak prediction module builds a peak prediction model based on the data stored in the basic database module to obtain the peak estimate. The dust concentration spatial distribution module establishes a dust concentration spatial distribution model based on preprocessed measured dust concentration data and peak value estimates to obtain dust concentration distribution data in the construction area. The worker positioning module is used to acquire worker location data; The early warning module is used to calculate the cumulative dust exposure of workers based on the location data of workers and the dust concentration distribution data of the construction area obtained by the dust concentration spatial distribution module, and to issue early warning information based on the cumulative dust exposure.

8. A dust exposure level monitoring device for personnel in tunnel construction areas, characterized in that, The device includes at least a memory and a processor; the processor calls instructions stored in the memory to execute the method for monitoring personnel dust exposure levels in tunnel construction areas as described in any one of claims 1-6.

Images