Building construction temporary facility early warning management and control system based on Internet

Abstract

The invention relates to the technical field of building construction early warning, and particularly discloses a building construction temporary facility early warning management and control system based on the Internet, environment data of a construction site are collected in real time through a multi-dimensional environment sensing equipment array, and a surface dust accumulation distribution image of a windproof dust suppression net is synchronously obtained through an image collection device; performing multi-source environmental parameter fusion modeling according to the environmental data to obtain a construction site environmental parameter fusion model; performing dynamic efficiency evaluation on the dust suppression capability of the windproof dust suppression net to obtain a dust suppression efficiency evaluation coefficient of the windproof dust suppression net; and when the dust suppression efficiency evaluation coefficient is lower than a preset threshold value, an atomization spraying system is triggered automatically, and a corresponding spraying instruction is generated automatically, so that the atomization spraying system is controlled. Through real-time monitoring, dynamic evaluation and automatic control, the use efficiency of the windproof dust suppression net can be effectively improved, raised dust emission of a construction site is reduced, and environment-friendly compliance of construction activities is ensured.

Description

Technical Field

[0001] The present invention belongs to the technical field of building construction early warning, and relates to an Internet-based building construction temporary facility early warning management and control system. Background Art

[0002] During construction, wind and dust suppression nets, as a crucial temporary facility, play a significant role in reducing dust on construction sites, improving air quality, and protecting the surrounding environment. Real-time monitoring of the status and effectiveness of wind and dust suppression nets not only ensures their effectiveness during construction but also enables timely identification and resolution of potential issues, minimizing the impact of dust on the surrounding environment and residents.

[0003] However, existing technologies for dust suppression management using windbreak and dust suppression nets still have several technical deficiencies: current mainstream solutions rely on isolated data from anemometers or dust concentration sensors, lacking the simultaneous collection and fusion analysis of multi-dimensional parameters such as dust thickness, damage morphology, and porosity changes. This results in incomplete input features for the performance evaluation model and an inability to accurately quantify key performance indicators such as air permeability attenuation. For example, wind speed data alone cannot distinguish the cause of a drop in dust interception efficiency—whether it is dust blockage that causes a drop in air permeability, or the expansion of damage that causes airflow to escape. Existing performance evaluation models are typically constructed based on static parameters (such as initial porosity and design height), and do not incorporate dynamic attenuation factors such as the time-varying increase in dust thickness and porosity degradation caused by material aging. As a result, the deviation between the evaluation coefficient and the actual dust suppression capacity accumulates over time, making it impossible to support accurate maintenance decisions. Summary of the Invention

[0004] In view of the above problems existing in the prior art, the present invention provides an Internet-based early warning and control system for temporary facilities in construction, which is used to solve the above technical problems.

[0005] In order to achieve the above-mentioned and other purposes, the technical solutions adopted by the present invention are as follows: The present invention provides an Internet-based early warning and control system for temporary construction facilities. The system includes a parameter model construction module, a dust suppression efficiency evaluation module, and a spray instruction generation module. The modules are connected via wired and / or wireless connections to achieve data transmission between the modules. Parameter model construction module: This module collects environmental data from the construction site in real time through a multi-dimensional environmental sensing device array, and simultaneously obtains images of dust accumulation and distribution on the surface of the wind and dust suppression net through an image acquisition device. Based on this environmental data, multi-source environmental parameter fusion modeling is performed to obtain a construction site environmental parameter fusion model. Dust suppression efficiency evaluation module: Based on the construction site environmental parameter fusion model, the main axis direction of pollution diffusion is predicted; combined with the average amount of dust emitted per unit time at the construction site, the dust suppression ability of the windbreak and dust suppression net is dynamically evaluated to obtain the dust suppression efficiency evaluation coefficient of the windbreak and dust suppression net; Spray instruction generation module: When the dust suppression efficiency evaluation coefficient is lower than the preset threshold, the atomizing spray system is automatically triggered, and the corresponding spray instruction is autonomously generated to control the atomizing spray system.

[0006] For example, the environmental data of the construction site includes wind speed data, wind direction vector data, and dust concentration data, wherein the specific collection process includes the following steps: Step S11: deploy multiple sets of wind speed sensors, wind direction sensors, and dust concentration sensors at the construction site, and upload the collected environmental data to a cloud server in real time; Step S12: Using a high-resolution camera mounted on a wind and dust suppression net support, periodically capture images of surface dust accumulation, perform binary segmentation on the images, extract the contours of the dust-covered area, calculate the amount of dust adhered per unit area based on the mesh size of the wind and dust suppression net, and generate a dust density gradient distribution matrix; Step S13: performing spatial interpolation processing on the wind speed data and wind direction vector data to generate a wind speed vector distribution field; Step S14: Input the dust concentration data, dust density gradient distribution matrix and wind speed vector distribution field into the fusion algorithm model to construct a construction site environmental parameter fusion model.

[0007] Exemplarily, the operation logic of step S14 is: Step S141: Based on the wind speed vector distribution field generated in step S13, perform Fourier transform frequency domain analysis on the wind speed data sequence, extract the main frequency characteristics of wind speed fluctuations and energy spectrum density parameters, and construct a wind speed dynamic change representation matrix; Step S142: performing sliding window filtering on the dust concentration data, and calculating the first-order derivative and second-order derivative of the concentration change with time in adjacent time windows to obtain concentration dynamic change rate data; Step S143: performing binary segmentation on the dust adhesion density distribution image, extracting the dust coverage area contour, calculating the dust adhesion amount per unit area based on the grid size of the wind and dust suppression net, and generating a dust density gradient distribution matrix; Step S144: Based on the weighted fusion algorithm, the wind speed frequency domain characteristic parameters, the concentration dynamic change rate data and the dust density gradient distribution matrix are superimposed on the multi-source data to obtain the construction site environmental parameter fusion model, in which the weight coefficient is dynamically adjusted by the particle swarm optimization algorithm.

[0008] For example, based on the construction site environmental parameter fusion model, the pollution diffusion main axis direction is predicted, and the prediction logic is: Step S21: extracting the average values ​​of the x- and y-direction components of the wind speed vector from the wind speed dynamic change representation matrix, that is, summing the x-direction wind speeds of all monitoring points and taking the average value to obtain the horizontal wind speed component. Similarly, calculating the y-direction average value to obtain the vertical wind speed component. Step S22: Analyze the dust accumulation image simultaneously, calculate the horizontal and vertical variation of the dust density at each grid point, and find the direction with the largest variation rate as the main gradient direction of the dust density; Step S23: Calculate the dust concentration change rate, then construct a data matrix containing the wind speed vector, the dust concentration change rate, and the gray density gradient, calculate the covariance change relationship matrix between the three, and find the combined direction with the greatest impact on pollution diffusion through eigendecomposition. This direction is the initial predicted main axis direction of pollution diffusion.

[0009] For example, the dust suppression efficiency evaluation coefficient of the wind and dust suppression net is obtained, and the calculation process is as follows: The dust accumulation distribution image on the surface of the windbreak and dust suppression net was segmented using a multi-scale convolutional neural network to calculate the dust coverage rate. Binocular stereo vision technology was used to reconstruct the three-dimensional morphology of the dust accumulation, extract the thickness distribution data, and calculate the thickness variation coefficient. The U-Net++ deep learning model was used to identify damaged areas of the net and calculate the damage rate of the wind and dust suppression net. The porosity of the intact area was analyzed through microscopic images, and a porosity degradation model was constructed based on the usage time of the wind and dust suppression net. Based on the dust coverage rate and thickness variation coefficient, the air permeability attenuation factor is constructed; at the same time, the damage rate and porosity degradation results are combined to generate the effective air permeability area correction coefficient; The theoretical dust suppression efficiency is calculated based on the cosine square value of the angle between the main axis of pollution diffusion and the optimal layout direction of the dust suppression net. The air permeability attenuation factor, the effective ventilation area correction coefficient and the average amount of dust emitted per unit time at the construction site are superimposed. Finally, the dust suppression efficiency evaluation coefficient of the wind-proof and dust-suppressing net in the range of 0 to 1 is generated by normalizing the historical data.

[0010] Exemplarily, the spray instruction includes a jet direction range and an atomization pattern of the atomizing spray system.

[0011] Exemplarily, step S31: Based on the prediction result of the pollution diffusion main axis direction, the real-time wind speed sensor data and the physical layout topology of the dust suppression network are integrated to perform the following process: Step S311: determining a basic deflection angle of the jet direction according to the cosine value of the angle between the pollution diffusion main axis direction and the standard layout direction of the dust suppression net; Step S312: Use the sliding window algorithm to calculate the standard deviation of wind speed fluctuations in the previous 10 minutes. For every 1m / s increase in wind speed fluctuation, dynamically compensate the angle of the jet direction to generate the final jet direction interval. Step S32: Based on the spatial distribution density of the pollution diffusion main axis and the dust suppression efficiency evaluation coefficient, the following process is performed: Step S321: performing morphological dilation processing on the high-density continuous area in the direction of the pollution diffusion main axis, eliminating discrete noise points and extracting the diffusion corridor boundary; Step S322: spatially superimpose the diffusion corridor and the area of ​​dust accumulation coverage to generate a high-risk area for dust accumulation-diffusion coupling; Step S323: Calculate the ratio of the area of ​​the coupling high-risk area to the entire network, and calculate the weight coefficient based on the current average value of dust emissions; Step S324: Determine the atomization mode of the atomization spray system based on the weight coefficient.

[0012] As described above, the present invention provides an Internet-based early warning and control system for temporary construction facilities, which has at least the following beneficial effects: The present invention provides an Internet-based early warning and control system for temporary construction facilities, which collects environmental data of the construction site in real time through a multi-dimensional environmental sensing device array, and simultaneously obtains the surface dust accumulation distribution image of the wind and dust suppression net. This real-time collection and fusion of multi-source data can comprehensively reflect the environmental conditions of the construction site, including meteorological factors such as wind speed, wind direction, humidity, temperature, and the actual use of the wind and dust suppression net. By establishing a fusion model of construction site environmental parameters, the diffusion trend of dust can be scientifically analyzed and predicted, thereby providing data support for subsequent dust suppression measures. This real-time monitoring and modeling method helps construction units to promptly understand the dust situation on site, respond quickly, and ensure the effectiveness of dust suppression measures.

[0013] The dust suppression effectiveness evaluation module uses an environmental parameter fusion model to predict the direction of the main axis of pollution diffusion and, combined with the average amount of dust emitted per unit time, dynamically evaluates the dust suppression capabilities of windbreak and dust suppression nets. This process not only quantifies the nets' actual dust suppression effectiveness but also enables timely adjustments to construction site dust suppression strategies based on the evaluation results. Through dynamic monitoring and evaluation, construction units can flexibly adjust dust suppression measures under varying environmental conditions, thereby improving overall dust suppression efficiency. This scientific evaluation mechanism can effectively reduce the impact of construction on the surrounding environment and mitigate the threat posed by dust to resident health.

[0014] When the dust suppression effectiveness evaluation coefficient falls below a preset threshold, the spray command generation module automatically triggers the atomizing spray system and generates the corresponding spray command. This automated control design significantly improves construction site response speed and management efficiency. When dust concentration exceeds the safety threshold, the system quickly activates the spray device to spray atomizing water, reducing airborne dust concentration. This automated control method not only reduces the burden of manual operation but also ensures that effective dust suppression measures are implemented at critical moments, maximizing protection for the surrounding environment and the health of residents.

[0015] The embodiments of the present invention can effectively improve the use efficiency of wind and dust suppression nets through real-time monitoring, dynamic evaluation and automatic control, reduce dust emissions at construction sites, and ensure the environmental compliance of construction activities. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 Schematic diagram of the connection of various modules of the system of the present invention. DETAILED DESCRIPTION

[0018] The above contents described below in conjunction with the implementation of the present invention are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the claims, they shall fall within the scope of protection of the present invention.

[0019] Example 1 See also Figure 1 As shown, an Internet-based early warning and control system for temporary construction facilities includes a parameter model construction module, a dust suppression efficiency evaluation module, and a spray instruction generation module, wherein each module is connected by wired and / or wireless connections to achieve data transmission between the modules; Parameter model construction module: This module collects environmental data from the construction site in real time through a multi-dimensional environmental sensing device array, and simultaneously obtains images of dust accumulation and distribution on the surface of the wind and dust suppression net through an image acquisition device. Based on this environmental data, multi-source environmental parameter fusion modeling is performed to obtain a construction site environmental parameter fusion model. The environmental data at the construction site include wind speed data, wind direction vector data and dust concentration data, among which wind speed data include but are not limited to wind speed amplitude, maximum instantaneous wind speed, etc.; wind direction vector data include but are not limited to horizontal wind direction angle, wind direction standard deviation, vertical wind component, etc.; dust concentration data include but are not limited to basic concentration, maximum instantaneous concentration, etc.

[0020] The specific collection process includes the following steps: Step S11: deploy multiple sets of wind speed sensors, wind direction sensors, and dust concentration sensors at the construction site, and upload the collected environmental data to a cloud server in real time; Step S12: Using a high-resolution camera mounted on a wind and dust suppression net support, periodically capture images of surface dust accumulation, perform binary segmentation on the images, extract the contours of the dust-covered area, calculate the amount of dust adhered per unit area based on the mesh size of the wind and dust suppression net, and generate a dust density gradient distribution matrix; Step S13: performing spatial interpolation processing on the wind speed data and wind direction vector data to generate a wind speed vector distribution field; Step S14: Input the dust concentration data, dust density gradient distribution matrix and wind speed vector distribution field into the fusion algorithm model to construct a construction site environmental parameter fusion model.

[0021] After capturing images of surface dust accumulation, the present embodiment first performs distortion correction to eliminate edge stretching caused by wide-angle lenses. The image is then segmented using adaptive threshold binarization. A segmentation threshold is determined based on grayscale histogram analysis. Areas with pixel values ​​above the threshold are identified as dust accumulation areas. Continuous contour boundaries are extracted, and noise points with areas smaller than the set threshold are filtered out. The segmented dust-covered areas are then divided into blocks using the standard grid size of the dust suppression net. The percentage of dust pixels within each grid is calculated (for example, a grid with 50% dust pixels is considered to have a level of 0.5 accumulation). A dust density gradient distribution matrix is ​​then constructed based on spatial location. The discrete point data collected by the wind speed sensor is spatially expanded using inverse distance weighted interpolation. The inverse distance weights of all sensor data within a 50-meter radius around each interpolation point are calculated, and a continuous wind speed scalar field covering the entire construction area is generated by superposition. Simultaneously, based on the horizontal azimuth and pitch angle measurements of the three-dimensional wind direction sensor, the wind direction is decomposed into two orthogonal components: east-west and south-north. The same interpolation algorithm is used to construct wind direction vector component fields for each direction. To achieve spatially coordinated expression of wind speed and direction, the scalar wind speed field is vector-combined with the two directional component fields to generate a vector distribution field that maintains continuity in both wind speed magnitude and direction. A terrain obstruction correction factor is incorporated into the interpolation process: For areas obstructed by buildings or stockpiles, a wind resistance attenuation factor is calculated based on the height and spacing of the obstructions, dynamically reducing the interpolation weight in that area to avoid inflated predictions.

[0022] The operation logic of step S14 is: Step S141: Based on the wind speed vector distribution field generated in step S13, perform Fourier transform frequency domain analysis on the wind speed data sequence, extract the main frequency characteristics of wind speed fluctuations and energy spectrum density parameters, and construct a wind speed dynamic change representation matrix; Step S142: performing sliding window filtering on the dust concentration data, and calculating the first-order derivative and second-order derivative of the concentration change with time in adjacent time windows to obtain concentration dynamic change rate data; Step S143: performing binary segmentation on the dust adhesion density distribution image, extracting the dust coverage area contour, calculating the dust adhesion amount per unit area based on the grid size of the wind and dust suppression net, and generating a dust density gradient distribution matrix; Step S144: Based on the weighted fusion algorithm, the wind speed frequency domain characteristic parameters, the concentration dynamic change rate data and the dust density gradient distribution matrix are superimposed on the multi-source data to obtain the construction site environmental parameter fusion model, in which the weight coefficient is dynamically adjusted by the particle swarm optimization algorithm.

[0023] This embodiment of the present invention first captures wind speed amplitude data in 10-minute time windows. A fast Fourier transform is performed on the discrete wind speed sequence within each time window, converting the time-domain wind speed fluctuations into a frequency-domain energy spectrum distribution. By traversing the peak points in the spectrum, frequency bands whose energy proportion exceeds a set threshold are selected as dominant frequency features, and the energy integral of each frequency band is calculated as the energy spectrum density parameter. To further quantify the dynamic characteristics of wind speed, the amplitude, energy density, and frequency band width corresponding to the dominant frequency are normalized in three dimensions to construct a wind speed dynamic variation matrix with time windows as rows and frequency domain features as columns. Dynamic sliding window filtering and rate of change analysis are performed on the dust concentration data. Using a sliding window mechanism with a time window length of 5 minutes and a sliding step size of 1 minute, the original concentration series is subjected to median filtering of the data within the window. After filtering, the mean change in concentration between adjacent time windows is calculated as the first-order derivative to represent the concentration change rate. The curvature change of the concentration data within the window is also calculated as the second-order derivative to represent the acceleration of change. For example, when the first-order derivative remains positive and the second-order derivative rises at an accelerated rate within a certain period of time, it is determined that there is a sudden increase in the risk of dust diffusion in the area. An adaptive threshold algorithm is used to perform binary segmentation on the grayscale image of the dust adhesion density distribution image. The segmentation threshold is dynamically adjusted according to the grayscale mean and variance of the local area of ​​the image to accurately separate the dust adhesion area from the clean mesh surface. After segmentation, the edge tracking algorithm is used to extract the continuous contour boundary of the dust-covered area and eliminate noise points with an area of ​​less than 10 pixels. Based on the actual grid size of the wind and dust control net, the image is divided into a regular grid array, the proportion of dust pixels in each grid is counted, and the dust density gradient distribution matrix is ​​constructed according to the row and column positions. A dynamic weighted multi-channel fusion algorithm is used to superimpose the wind speed frequency domain feature matrix, dust concentration change rate data, and dust density gradient distribution matrix in three-dimensional space. The weight allocation mechanism is implemented through the particle swarm optimization algorithm: the basic weights of wind speed, concentration, and dust density are initialized to 0.4, 0.3, and 0.3, respectively; the mean square error between the model prediction results and the measured data is used as the optimization target, and the objective function is minimized by iteratively adjusting the weight coefficient, ultimately obtaining the dynamic weight combination that maximizes the model accuracy.

[0024] Dust suppression efficiency evaluation module: Based on the construction site environmental parameter fusion model, the main axis direction of pollution diffusion is predicted; combined with the average amount of dust emitted per unit time at the construction site, the dust suppression ability of the windbreak and dust suppression net is dynamically evaluated to obtain the dust suppression efficiency evaluation coefficient of the windbreak and dust suppression net; Based on the construction site environmental parameter fusion model, the main axis direction of pollution diffusion is predicted. The prediction logic is: Step S21: extracting the average values ​​of the x- and y-direction components of the wind speed vector from the wind speed dynamic change representation matrix, that is, summing the x-direction wind speeds of all monitoring points and taking the average value to obtain the horizontal wind speed component. Similarly, calculating the y-direction average value to obtain the vertical wind speed component. Step S22: Analyze the dust accumulation image simultaneously, calculate the horizontal and vertical variation of the dust density at each grid point, and find the direction with the largest variation rate as the main gradient direction of the dust density; Step S23: Calculate the dust concentration change rate, then construct a data matrix containing the wind speed vector, the dust concentration change rate, and the gray density gradient, calculate the covariance change relationship matrix between the three, and find the combined direction with the greatest impact on pollution diffusion through eigendecomposition. This direction is the initial predicted main axis direction of pollution diffusion.

[0025] This embodiment of the present invention retrieves a wind speed dynamic change representation matrix from a cloud server, extracts raw x-axis wind speed data from all monitoring points, performs an arithmetic average calculation on the wind speed values ​​at each monitoring point to obtain the horizontal average wind speed component, and simultaneously performs the same processing on the y-axis wind speed data to obtain the vertical average wind speed component, thereby forming planar wind field vector feature data. Next, the preprocessed dust density gradient distribution matrix is ​​called up, and the horizontal and vertical grayscale gradients are calculated for each 1cm×1cm pixel grid. The square root of the sum of the squares of the horizontal and vertical gradient amplitudes is taken to obtain the comprehensive gradient intensity. The grid cell with the largest intensity value is selected, and the gradient direction of this cell is used as the main gray density gradient direction. During this process, the Sobel operator is used for gradient detection. The dust concentration sensor data is then subjected to sliding window processing, and the absolute difference ratio of the concentration change is calculated within a 5-minute time window as the dynamic concentration change rate. When constructing the multidimensional data matrix, the horizontal wind speed component, vertical wind speed component, concentration change rate, horizontal gray density gradient value, and vertical gray density gradient value are stored in the matrix column vectors in a time-aligned manner. The covariance change relationship matrix is ​​obtained by calculating the covariance values ​​between each pair of column vectors. The eigendecomposition thereof adopts Jacobi rotation iteration method, and the eigenvector corresponding to the maximum eigenvalue is taken as the main axis direction of pollution diffusion.

[0026] The dust suppression efficiency evaluation coefficient of the wind and dust suppression net is obtained, and the calculation process is as follows: The surface dust accumulation distribution image of the windbreak and dust suppression net was segmented using a multi-scale convolutional neural network to calculate the dust coverage rate. The three-dimensional morphology of the dust was reconstructed using binocular stereo vision technology, and the thickness distribution data was extracted to calculate the thickness variation coefficient. The U-Net++ deep learning model was used to identify damaged areas of the net and calculate the damage rate of the wind and dust suppression net. The porosity of the intact area was analyzed through microscopic images, and a porosity degradation model was constructed based on the usage time of the wind and dust suppression net. Based on the dust coverage rate and thickness variation coefficient, the air permeability attenuation factor is constructed; at the same time, the damage rate and porosity degradation results are combined to generate the effective air permeability area correction coefficient; The theoretical dust suppression efficiency is calculated based on the cosine square value of the angle between the main axis of pollution diffusion and the optimal layout direction of the dust suppression net. The air permeability attenuation factor, the effective ventilation area correction coefficient and the average amount of dust emitted per unit time at the construction site are superimposed. Finally, the dust suppression efficiency evaluation coefficient of the wind-proof and dust-suppressing net in the range of 0 to 1 is generated by normalizing the historical data.

[0027] The embodiment of the present invention uses multi-angle high-definition camera devices deployed around the wind and dust control net to collect dust accumulation distribution images on the net surface in real time. A multi-scale convolutional neural network is used to segment the image: first, the original image is histogram equalized to enhance the contrast, and then texture features of different scales are extracted through multi-level convolution kernels, and finally a binary mask image is output, where the white area represents dust accumulation and the black area represents clean area. The percentage of pixels in the dust accumulation area is the dust coverage rate. In order to obtain three-dimensional thickness data, binocular stereo vision technology is used. The left and right cameras are used to synchronously shoot images, and the feature point matching algorithm is used to calculate the disparity map. The camera calibration parameters are combined to reconstruct a three-dimensional point cloud model of the dust surface, from which the thickness distribution is extracted and the thickness variation coefficient is calculated. The calculation logic of the thickness variation coefficient is the ratio of the standard deviation of the thickness values ​​of all sampling points to the average thickness.

[0028] The embodiment of the present invention constructs a damage recognition model based on the U-Net++ deep learning architecture. The specific construction logic is: using a transfer learning strategy, fine-tuning training is performed on the pre-trained model through a labeled damage image dataset (including cracks, tears, holes, etc.). After the model outputs the semantic segmentation result of the damaged area, the proportion of damaged pixels to the total pixels is counted to generate the damage rate. The microscopic image of the undamaged area is further magnified to analyze the pore structure, and the dynamic threshold segmentation method is used to distinguish pores from entities. The current porosity is obtained by counting the proportion of pore area; a porosity degradation model is established in combination with the cumulative usage time of the dust suppression net (the initial porosity decays exponentially over time, and the decay rate is calibrated through material aging experiments).

[0029] The air permeability attenuation factor in this embodiment of the present invention is determined by both the dust coverage and the thickness variation coefficient: the higher the dust coverage and the more uneven the thickness distribution, the greater the decline in air permeability (specifically, the coverage affects the attenuation rate according to a nonlinear power relationship, while the thickness variation coefficient is scaled and adjusted using a tangent function). The effective air permeability area correction factor is calculated based on the damage rate and porosity degradation results: first, the theoretical air permeability area is reduced by the damage rate, and then the quadratic attenuation effect of porosity degradation on micropore permeability is added to generate the effective air permeability correction factor.

[0030] The embodiment of the present invention extracts the theoretical dust suppression efficiency benchmark value of the windproof and dust suppression net from the database, and performs step-by-step weight superposition on it with the real-time monitored air permeability attenuation factor and the effective ventilation area correction coefficient (the air permeability attenuation factor reflects the influence of dust accumulation and blockage, and the ventilation area correction coefficient represents the damage and pore degradation effect). At the same time, a compensation weight coefficient is introduced according to the dynamic changes of the average value of dust emissions per unit time at the construction site. For example, when the dust emissions exceed the set threshold, the dynamic weight coefficient increases the compensation amplitude for the efficiency attenuation according to a logarithmic relationship. After the multi-factor coupling operation, the initial efficiency evaluation coefficient is output, and then based on the extreme value interval consisting of the maximum efficiency value and the minimum efficiency value in the past three years extracted from the historical monitoring database, the initial efficiency evaluation coefficient is converted into a standardized windproof and dust suppression net dust suppression efficiency evaluation coefficient in the range of 0 to 1 through linear mapping.

[0031] Spray instruction generation module: When the dust suppression efficiency evaluation coefficient is lower than the preset threshold, the atomizing spray system is automatically triggered, and the corresponding spray instruction is autonomously generated to control the atomizing spray system.

[0032] The spray instruction includes the jet direction range and atomization pattern of the atomizing spray system.

[0033] Step S31: Based on the prediction result of the pollution diffusion main axis direction, the real-time wind speed sensor data and the physical layout topology of the dust suppression network are integrated to perform the following process: Step S311: Determine the basic deflection angle of the jet direction based on the cosine value of the angle between the pollution diffusion axis and the standard layout direction of the dust suppression net. That is, the larger the angle, the greater the angle at which the jet direction needs to be deflected upstream of the pollution source, and the deflection amplitude is inversely proportional to the cosine value. Step S312: Use the sliding window algorithm to calculate the standard deviation of wind speed fluctuations in the previous 10 minutes. For every 1m / s increase in wind speed fluctuation, dynamically compensate the angle of the jet direction to generate the final jet direction interval. Step S32: Based on the spatial distribution density of the pollution diffusion main axis and the dust suppression efficiency evaluation coefficient, the following process is performed: Step S321: performing morphological dilation processing on the high-density continuous area in the direction of the pollution diffusion main axis, eliminating discrete noise points and extracting the diffusion corridor boundary; Step S322: spatially superimpose the diffusion corridor and the area of ​​dust accumulation coverage to generate a high-risk area for dust accumulation-diffusion coupling; Step S323: Calculate the ratio of the area of ​​the coupling high-risk area to the entire network, and calculate the weight coefficient based on the current average value of dust emissions; Step S324: Determine the atomization mode of the atomization spray system based on the weight coefficient.

[0034] In an embodiment of the present invention, a high-precision angle sensor is used to measure the actual angle between the main axis of pollution diffusion and the standard layout direction of the dust suppression net (usually the normal direction of the net body), and the cosine value of the angle between the two is calculated. When the cosine value is smaller, the deviation between the diffusion direction and the normal of the net body is greater. At this time, the direction of the spray jet needs to be offset upstream of the pollution source (i.e., in the opposite direction of the diffusion direction). The offset angle increases nonlinearly with the decrease of the cosine value - when the cosine value is 0.5, the offset angle is set to 15°; if the cosine value is reduced to 0.2, the offset angle increases to 35°, and so on to achieve precise reverse interception. An industrial-grade ultrasonic anemometer is used to collect wind speed data in real time, and the standard deviation of wind speed fluctuations is calculated with a sliding window length period of 10 minutes. When the standard deviation exceeds 1m / s, it is determined that the wind direction stability is insufficient, and the basic deflection angle is dynamically compensated at this time: For every 1 m / s increase in the standard deviation, the jet direction is offset by a threshold value, based on the baseline deflection angle, toward the dominant wind speed direction. A rolling average algorithm suppresses angular oscillations caused by sudden changes in wind speed, ensuring smooth transitions in jet direction adjustment. The final output jet direction interval value is [baseline deflection angle - compensation amount, baseline deflection angle + compensation amount].

[0035] This embodiment of the present invention uses a 5×5 pixel rectangular structuring element to perform a morphological dilation operation on the density map, eliminating discrete noise points and connecting broken areas. An edge-tracing algorithm is then used to extract the diffusion corridor boundaries. The dust coverage heat map and the diffusion corridor spatial distribution map are superimposed to identify overlapping areas. These overlapping areas are then rasterized, and grids that meet the following conditions are marked as high-risk areas for coupling: dust coverage ≥ a set threshold or diffusion corridor density ≥ a set threshold. The proportion of high-risk areas for coupling to the total area of ​​the dust suppression network is calculated. Current dust emission mean data is collected and its ratio to historical baseline emissions is calculated. A comprehensive weight coefficient is then generated: weight = area ratio × emissions ratio.

[0036] According to the weight coefficient, 4 control levels are divided from low to high, and the corresponding atomization modes are set as follows: if the weight coefficient is between 0-0.2, the atomization mode is set to normal mode; if the weight coefficient is between 0.2-0.4, the atomization mode is set to enhanced mode; if the weight coefficient is between 0.4-0.6, the atomization mode is set to precision mode; if the weight coefficient is greater than 0.6, the atomization mode is set to super enhanced mode.

[0037] It should be noted that the intervals and thresholds are set for ease of comparison. The threshold size depends on the amount of sample data and the cardinality set by those skilled in the art for each set of sample data, as long as it does not affect the proportional relationship between the parameter and the quantized value. Furthermore, the above formulas are all dimensionless numerical calculations. These formulas are derived from software simulations of the most recent real-world conditions using large amounts of data. The preset parameters in these formulas are set by those skilled in the art based on actual conditions.

[0038] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0039] It should be understood that determining B based on A does not mean determining B based solely on A. B can also be determined based on A and / or other information.

[0040] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0041] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An Internet-based early warning and control system for temporary construction facilities, characterized in that: include: Parameter model construction module: This module collects environmental data from the construction site in real time through a multi-dimensional environmental sensing device array, and simultaneously obtains images of dust accumulation and distribution on the surface of the wind and dust suppression net through an image acquisition device. Based on this environmental data, multi-source environmental parameter fusion modeling is performed to obtain a construction site environmental parameter fusion model. Dust suppression efficiency evaluation module: Based on the construction site environmental parameter fusion model, the main axis direction of pollution diffusion is predicted; combined with the average amount of dust emitted per unit time at the construction site, the dust suppression ability of the windbreak and dust suppression net is dynamically evaluated to obtain the dust suppression efficiency evaluation coefficient of the windbreak and dust suppression net; Spray instruction generation module: When the dust suppression efficiency evaluation coefficient is lower than the preset threshold, the atomizing spray system is automatically triggered, and the corresponding spray instruction is autonomously generated to control the atomizing spray system.

2. The Internet-based construction temporary facility early warning and control system according to claim 1 is characterized in that: The environmental data at the construction site includes wind speed data, wind direction vector data, and dust concentration data. The specific collection process includes the following steps: Step S11: deploy multiple sets of wind speed sensors, wind direction sensors, and dust concentration sensors at the construction site, and upload the collected environmental data to a cloud server in real time; Step S12: Using a high-resolution camera mounted on a wind and dust suppression net support, periodically capture images of surface dust accumulation, perform binary segmentation on the images, extract the contours of the dust-covered area, calculate the amount of dust adhered per unit area based on the mesh size of the wind and dust suppression net, and generate a dust density gradient distribution matrix; Step S13: performing spatial interpolation processing on the wind speed data and wind direction vector data to generate a wind speed vector distribution field; Step S14: Input the dust concentration data, dust density gradient distribution matrix and wind speed vector distribution field into the fusion algorithm model to construct a construction site environmental parameter fusion model.

3. The Internet-based construction temporary facility early warning and control system according to claim 2 is characterized in that: The operation logic of step S14 is: Step S141: Based on the wind speed vector distribution field generated in step S13, perform Fourier transform frequency domain analysis on the wind speed data sequence, extract the main frequency characteristics of wind speed fluctuations and energy spectrum density parameters, and construct a wind speed dynamic change representation matrix; Step S142: performing sliding window filtering on the dust concentration data, and calculating the first-order derivative and second-order derivative of the concentration change with time in adjacent time windows to obtain concentration dynamic change rate data; Step S143: performing binary segmentation on the dust adhesion density distribution image, extracting the dust coverage area contour, calculating the dust adhesion amount per unit area based on the grid size of the wind and dust suppression net, and generating a dust density gradient distribution matrix; Step S144: Based on the weighted fusion algorithm, the wind speed frequency domain characteristic parameters, the concentration dynamic change rate data and the dust density gradient distribution matrix are superimposed on the multi-source data to obtain the construction site environmental parameter fusion model, in which the weight coefficient is dynamically adjusted by the particle swarm optimization algorithm.

4. The Internet-based construction temporary facility early warning and control system according to claim 3 is characterized in that: Based on the construction site environmental parameter fusion model, the main axis direction of pollution diffusion is predicted. The prediction logic is: Step S21: extracting the average values ​​of the x- and y-direction components of the wind speed vector from the wind speed dynamic change representation matrix, that is, summing the x-direction wind speeds of all monitoring points and taking the average value to obtain the horizontal wind speed component. Similarly, calculating the y-direction average value to obtain the vertical wind speed component. Step S22: Analyze the dust accumulation image simultaneously, calculate the horizontal and vertical variation of the dust density at each grid point, and find the direction with the largest variation rate as the main gradient direction of the dust density; Step S23: Calculate the dust concentration change rate, then construct a data matrix containing the wind speed vector, the dust concentration change rate, and the gray density gradient, calculate the covariance change relationship matrix between the three, and find the combined direction with the greatest impact on pollution diffusion through eigendecomposition. This direction is the initial predicted main axis direction of pollution diffusion.

5. The Internet-based construction temporary facility early warning and control system according to claim 4 is characterized in that: The dust suppression efficiency evaluation coefficient of the wind and dust suppression net is obtained, and the calculation process is as follows: The surface dust accumulation distribution image of the windbreak and dust suppression net was segmented using a multi-scale convolutional neural network to calculate the dust coverage rate. The three-dimensional morphology of the dust was reconstructed using binocular stereo vision technology, and the thickness distribution data was extracted to calculate the thickness variation coefficient. A deep learning model was used to identify damaged areas of the net, calculate the damage rate of the wind and dust suppression net, analyze the porosity of the intact area through microscopic images, and construct a porosity degradation model based on the usage time of the wind and dust suppression net. Based on the dust coverage rate and thickness variation coefficient, the air permeability attenuation factor is constructed; at the same time, the damage rate and porosity degradation results are combined to generate the effective air permeability area correction coefficient; The theoretical dust suppression efficiency is calculated based on the cosine square value of the angle between the main axis of pollution diffusion and the optimal layout direction of the dust suppression net. The air permeability attenuation factor, the effective ventilation area correction coefficient and the average amount of dust emitted per unit time at the construction site are superimposed. Finally, the dust suppression efficiency evaluation coefficient of the wind-proof and dust-suppressing net in the range of 0 to 1 is generated by normalizing the historical data.

6. The Internet-based construction temporary facility early warning and control system according to claim 1 is characterized in that: The spray instruction includes the jet direction range and atomization pattern of the atomizing spray system.

7. The Internet-based construction temporary facility early warning and control system according to claim 6, characterized in that: Step S31: Based on the prediction result of the pollution diffusion main axis direction, the real-time wind speed sensor data and the physical layout topology of the dust suppression network are integrated to perform the following process: Step S311: determining a basic deflection angle of the jet direction according to the cosine value of the angle between the pollution diffusion main axis direction and the standard layout direction of the dust suppression net; Step S312: Use the sliding window algorithm to calculate the standard deviation of wind speed fluctuations in the previous 10 minutes. For every 1m / s increase in wind speed fluctuation, dynamically compensate the angle of the jet direction to generate the final jet direction interval. Step S32: Based on the spatial distribution density of the pollution diffusion main axis and the dust suppression efficiency evaluation coefficient, the following process is performed: Step S321: performing morphological dilation processing on the high-density continuous area in the direction of the pollution diffusion main axis, eliminating discrete noise points and extracting the diffusion corridor boundary; Step S322: spatially superimpose the diffusion corridor and the area of ​​dust accumulation coverage to generate a high-risk area for dust accumulation-diffusion coupling; Step S323: Calculate the ratio of the area of ​​the coupling high-risk area to the entire network, and calculate the weight coefficient based on the current average value of dust emissions; Step S324: Determine the atomization mode of the atomization spray system based on the weight coefficient.

Images