Method and system for analyzing pollution sources at a construction site

Abstract

The application provides a construction site pollution source analysis method and system, determines the atmospheric pollution source intensity information of each monitoring area of the construction site based on a preset Gaussian diffusion model and environmental monitoring data and meteorological data, detects and identifies a video stream through a pre-trained target detection model, determines the comprehensive score of the construction site within a preset time period based on a preset multi-dimensional scoring model, environmental monitoring data, detection and identification results and atmospheric pollution source intensity information, determines the key pollution information of the construction site based on the atmospheric pollution source intensity information and satellite image data and construction progress information of the construction site, performs pollution cause analysis based on the environmental monitoring data and historical construction data of the construction site, and determines the pollution source analysis result of the construction site based on the comprehensive score, key pollution information and pollution cause analysis result. The application can meet the high real-time pollution source analysis demand of the actual construction situation of the construction site due to the rapid change.

Description

Technical Field

[0001] This invention relates to the field of pollution source analysis technology, and in particular to a method and system for analyzing pollution sources at construction sites. Background Technology

[0002] Currently, the analysis of pollution sources at construction sites often relies on manual monitoring, which suffers from problems such as untimely monitoring, missed detections, and delayed analysis. This makes it unsuitable for real-time analysis of pollution sources during construction periods, such as tower crane operations and earthmoving, hindering timely responses from site managers. Therefore, real-time pollution source analysis during construction periods is needed, tailored to the specific conditions of the construction site, to meet the high real-time requirements arising from the rapid changes in construction conditions. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide a method and system for analyzing pollution sources at construction sites, so as to alleviate the above-mentioned problems existing in related technologies.

[0004] In a first aspect, embodiments of the present invention provide a method for analyzing pollution sources at a construction site, comprising: acquiring video streams of the construction site within a preset time period using multiple target cameras deployed at the construction site, and acquiring environmental monitoring data of the construction site within the preset time period using multiple sensors deployed at the construction site; wherein the target cameras are integrated cameras combining visible light and infrared sensors, and the environmental monitoring data includes atmospheric particulate matter concentration data and noise data at each monitoring point within the preset time period; determining the atmospheric pollution source intensity information of each monitoring area at the construction site within the preset time period based on a preset Gaussian diffusion model, the environmental monitoring data, and meteorological data of the construction site within the preset time period; and through... A pre-trained target detection model detects and identifies the video stream, and determines a comprehensive score for the construction site within a preset time period based on a preset multi-dimensional scoring model, the environmental monitoring data, the detection and identification results, and the atmospheric pollution source intensity information. Based on the atmospheric pollution source intensity information, satellite imagery data of the construction site, and construction progress information, key pollution information for the construction site is determined. This key pollution information includes key operational periods and key operational areas that generate pollution. Based on the environmental monitoring data and historical construction data of the construction site, a pollution cause analysis is performed, and based on the comprehensive score, the key pollution information, and the pollution cause analysis results, the pollution source analysis results for the construction site are determined.

[0005] Secondly, embodiments of the present invention also provide a construction site pollution source analysis system, comprising: multiple target cameras and multiple sensors deployed at the construction site, an Internet of Things (IoT) platform, and a cloud server. The IoT platform is connected to the multiple target cameras, the multiple sensors, and the cloud server. The target cameras are integrated cameras combining visible light and infrared sensors. The cloud server is equipped with a preset Gaussian diffusion model, a preset multidimensional scoring model, and a pre-trained target detection model. The cloud server is also connected to a meteorological database, a satellite image database, and a construction database. The multiple target cameras are used to collect video streams from the construction site within a preset time period. The multiple sensors are used to collect environmental monitoring data from the construction site within the preset time period. The environmental monitoring data includes atmospheric particulate matter concentration data and noise data at each monitoring point within the preset time period. The cloud server is used to: acquire the video streams and environmental monitoring data through the IoT platform, and acquire meteorological data from the meteorological database within the preset time period at the construction site; based on the preset Gaussian diffusion model... The model, environmental monitoring data, and meteorological data are used to determine the intensity information of air pollution sources in each monitoring area of ​​the construction site within a preset time period. The cloud server is also used to: detect and identify the video stream using the target detection model, and determine a comprehensive score for the construction site within a preset time period based on a preset multi-dimensional scoring model, the environmental monitoring data, the detection and identification results, and the air pollution source intensity information. The cloud server is also used to: obtain satellite image data of the construction site from the satellite image database, and obtain construction progress information of the construction site from the construction database; determine key pollution information of the construction site based on the air pollution source intensity information, the satellite image data, and the construction progress information; wherein, the key pollution information includes key operation periods and key operation areas that generate pollution; the cloud server is also used to: obtain historical construction data of the construction site from the construction database; perform pollution cause analysis based on the environmental monitoring data and the historical construction data, and determine the pollution source analysis results of the construction site based on the comprehensive score, the key pollution information, and the pollution cause analysis results.

[0006] This invention provides a method and system for analyzing pollution sources at construction sites. Based on a preset Gaussian diffusion model and environmental monitoring and meteorological data from the construction site within a preset time period, it determines the intensity information of atmospheric pollution sources in each monitoring area of ​​the construction site within that preset time period. Then, a pre-trained target detection model detects and identifies video streams from the construction site within the preset time period. Based on a preset multi-dimensional scoring model, environmental monitoring data, detection and identification results, and atmospheric pollution source intensity information, a comprehensive score for the construction site within the preset time period is determined. Next, based on the atmospheric pollution source intensity information, satellite imagery data of the construction site, and construction progress information, key pollution information is determined. Then, pollution cause analysis is performed based on environmental monitoring data and historical construction data from the construction site. Finally, based on the comprehensive score, key pollution information, and pollution cause analysis results, the pollution source analysis results for the construction site are determined. This method enables automated real-time analysis of pollution sources at construction sites, improving efficiency and accuracy compared to existing methods that rely on manual monitoring. It can meet the high real-time pollution source analysis needs arising from the rapid changes in actual construction conditions at construction sites.

[0007] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained in accordance with the structures particularly pointed out in the description, claims and drawings.

[0008] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0009] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0010] Figure 1 This is a flowchart illustrating a method for analyzing pollution sources at a construction site, as described in an embodiment of the present invention.

[0011] Figure 2 This is a schematic diagram of the construction site pollution source analysis system in an embodiment of the present invention. Detailed Implementation

[0012] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0013] Currently, the analysis of pollution sources at construction sites often relies on manual monitoring, which has problems such as untimely monitoring, missed detection, and delayed analysis. It is difficult to apply to the real-time analysis of pollution sources during construction periods such as tower crane operations and earthwork transportation, making it difficult for site managers to respond in a timely manner based on the pollution source analysis results.

[0014] Based on this, the present invention provides a method and system for analyzing pollution sources at construction sites, which can alleviate the above-mentioned problems existing in related technologies.

[0015] To facilitate understanding of this embodiment, a detailed description of a method for analyzing pollution sources at construction sites disclosed in this invention will be provided first. (See [link to relevant documentation]). Figure 1 As shown, the method may include the following steps:

[0016] Step S102: Video streams of the construction site are collected within a preset time period by multiple target cameras deployed at the construction site, and environmental monitoring data of the construction site within a preset time period are collected by multiple sensors deployed at the construction site.

[0017] The target camera can be an integrated camera that combines a visible light sensor and an infrared sensor, and the environmental monitoring data can include atmospheric particulate matter concentration data and noise data at each monitoring point within a preset time period.

[0018] For example, see Figure 2 As shown, multiple target cameras 201 and multiple sensors 202 can be deployed at the construction site, and the Internet of Things platform 203 can be connected to the multiple target cameras 201, multiple sensors 202 and cloud server 204 respectively, thereby forming a construction site pollution source analysis system. This system can collect video streams of the construction site within a preset time period through multiple target cameras 201 deployed at different locations, and collect atmospheric particulate matter concentration data and noise data of the construction site within a preset time period through multiple sensors 202 deployed at different locations.

[0019] Step S104: Based on the preset Gaussian diffusion model, environmental monitoring data, and meteorological data of the construction site within a preset time period, determine the atmospheric pollution source intensity information of each monitoring area of ​​the construction site within the preset time period.

[0020] Following the previous example, see Figure 2 As shown, a preset Gaussian diffusion model can be pre-deployed on the cloud server 204, and the cloud server 204 can be connected to the meteorological database 205. This allows the cloud server 204 to obtain atmospheric particulate matter concentration data and noise data collected by multiple sensors 202 at the construction site within a preset time period from the IoT platform 203, and meteorological data of the construction site within a preset time period from the meteorological database 205. This facilitates the cloud server 204 in determining the atmospheric pollution source intensity information of each monitoring area at the construction site within a preset time period based on the preset Gaussian diffusion model and the obtained atmospheric particulate matter concentration data, noise data, and meteorological data.

[0021] Step S106: The video stream is detected and identified using a pre-trained target detection model, and the comprehensive score of the construction site within a preset time period is determined based on a preset multi-dimensional scoring model, environmental monitoring data, detection and identification results, and atmospheric pollution source intensity information.

[0022] Following the previous example, see Figure 2 As shown, a pre-set multi-dimensional scoring model and a pre-trained target detection model can be pre-deployed on the cloud server 204. This allows the cloud server 204 to acquire video streams of the construction site collected by multiple target cameras 201 within a preset time period, as well as atmospheric particulate matter concentration data and noise data of the construction site collected by multiple sensors 202 within a preset time period from the IoT platform 203. Subsequently, the cloud server 204 can detect and identify the acquired video streams through the target detection model and determine the comprehensive score of the construction site within the preset time period based on the pre-set multi-dimensional scoring model, the obtained atmospheric particulate matter concentration data and noise data, the detection and identification results, and the atmospheric pollution source intensity information.

[0023] Step S108: Based on the atmospheric pollution source intensity information, satellite imagery data of the construction site, and construction progress information, determine the key pollution information of the construction site.

[0024] Key pollution information may include critical operating periods and critical operating areas that generate pollution.

[0025] Following the previous example, see Figure 2 As shown, the cloud server 204 can be pre-connected to the satellite image database 206 and the construction database 207 respectively, so that the cloud server 204 can obtain satellite image data of the construction site from the satellite image database 206 and obtain construction progress information of the construction site from the construction database 207. Then, the cloud server 204 can determine the key operation period and key operation area that generate pollution at the construction site based on the obtained atmospheric pollution source intensity information, the obtained satellite image data and construction progress information.

[0026] Step S110: Based on environmental monitoring data and historical construction data at the construction site, conduct pollution cause analysis, and determine the pollution source analysis results at the construction site based on comprehensive score, key pollution information and pollution cause analysis results.

[0027] Following the previous example, see Figure 2 As shown, the cloud server 204 can obtain atmospheric particulate matter concentration data and noise data collected by multiple sensors 202 at the construction site within a preset time period from the IoT platform 203, and obtain historical construction data of the construction site from the construction database 207. Then, the cloud server 204 can perform pollution cause analysis based on the obtained atmospheric particulate matter concentration data, noise data, and historical construction data. Finally, the cloud server 204 can use the obtained comprehensive score, key pollution information, and pollution cause analysis results to generate a pollution source analysis report for the construction site as the pollution source analysis result. The pollution source analysis report may include at least one part or all of the comprehensive score, key pollution information, and pollution cause analysis results.

[0028] This invention provides a method for analyzing pollution sources at construction sites. Based on a preset Gaussian diffusion model and environmental monitoring and meteorological data from the construction site within a preset time period, it determines the intensity information of atmospheric pollution sources in each monitoring area of ​​the construction site within that preset time period. Then, a pre-trained target detection model detects and identifies video streams from the construction site within the preset time period. Based on a preset multi-dimensional scoring model, environmental monitoring data, detection and identification results, and atmospheric pollution source intensity information, a comprehensive score for the construction site within the preset time period is determined. Next, based on the atmospheric pollution source intensity information, satellite imagery data of the construction site, and construction progress information, key pollution information is determined. Then, based on environmental monitoring data and historical construction data of the construction site, pollution cause analysis is performed. Finally, based on the comprehensive score, key pollution information, and pollution cause analysis results, the pollution source analysis results for the construction site are determined. This method enables automated real-time analysis of pollution sources at construction sites, improving efficiency and accuracy compared to existing methods that rely on manual monitoring. It can meet the high real-time pollution source analysis needs arising from the rapid changes in actual construction conditions at construction sites.

[0029] As one possible implementation, the meteorological data mentioned above may include wind speed data at the construction site within a preset time period, and the atmospheric pollution source intensity information may include the atmospheric pollution source intensity range of the corresponding monitoring area within the preset time period. Based on this, step S104 (i.e., determining the atmospheric pollution source intensity information of each monitoring area at the construction site within a preset time period based on the preset Gaussian diffusion model, environmental monitoring data, and meteorological data at the construction site within the preset time period) may include: determining the diffusion parameters of the preset Gaussian diffusion model corresponding to each monitoring point based on wind speed data; calculating the atmospheric pollution source intensity value of each monitoring point within the preset time period using the diffusion parameters and atmospheric particulate matter concentration data through the preset Gaussian diffusion model; and determining the atmospheric pollution source intensity range corresponding to the same monitoring area based on the atmospheric pollution source intensity values ​​corresponding to all monitoring points within the same monitoring area.

[0030] For example, the following Gaussian diffusion model can be used:

[0031]

[0032] in, For spatial points atmospheric pollutant concentration (unit: mg / m³) 3 ), Source strength (unit: mg / s), The average wind speed at the leak height (unit: m / s). , These are expressed as concentration standard deviations. axis, Diffusion parameters on the axis, Effective height for atmospheric pollutant leakage (unit: m); under windy conditions, , , The regression index is the horizontal diffusion parameter. For the regression index of vertical diffusion parameters, These are the regression coefficients for the lateral diffusion parameter. These are the regression coefficients for the vertical diffusion parameters. The horizontal distance (m) from the downwind side of the exhaust pipe.

[0033] Following the previous example, see Figure 2 As shown, multiple monitoring points can be deployed in each monitoring area, and each monitoring point is used to monitor the atmospheric particulate matter concentration at its corresponding location through a corresponding sensor 202; the cloud server 204 can obtain the atmospheric particulate matter concentration within a preset time period detected by the corresponding sensor 202 of each monitoring point through the Internet of Things platform 203. The cloud server 204 can also obtain the wind speed corresponding to each monitoring point within a preset time period from the meteorological database 205; after obtaining the atmospheric particulate matter concentration and wind speed corresponding to each monitoring point within the preset time period, the cloud server 204 can use the wind speed corresponding to each monitoring point to calculate the diffusion parameter corresponding to each monitoring point (i.e., , Furthermore, the cloud server 204 can then assign each monitoring point its corresponding... , and Substitute these values ​​into the expression of the Gaussian diffusion model above to calculate the atmospheric pollution source intensity value corresponding to each monitoring point. ;

[0034] Following the previous example, see Figure 2 As shown, the atmospheric pollution source intensity values ​​corresponding to all monitoring points in each monitoring area are obtained on cloud server 204. Then, the cloud server 204 can calculate the atmospheric pollution source intensity values ​​corresponding to all monitoring points within the same monitoring area. The minimum and maximum values, and then the cloud server 204 can output the atmospheric pollution source intensity values ​​corresponding to each monitoring area. The minimum and maximum values ​​are used as the lower and upper limits of the corresponding atmospheric pollution source intensity range, respectively, so that the cloud server 204 can obtain the atmospheric pollution source intensity range corresponding to each monitoring area.

[0035] As one possible implementation, the video stream may include visible light images and infrared images at multiple different times within a preset time period; based on this, the detection and recognition of the video stream by using a pre-trained target detection model in step S106 may include: performing image segmentation and target detection on the visible light images and infrared images at the same time using the target detection model to identify target regions and target objects, and obtaining the category and pixel information of each target region and each target object within the preset time period.

[0036] The target area can include areas with dust particles, areas covered by green netting, areas where materials are stacked, etc. The target object can include materials, personnel, etc. The pixel information can include the coordinates of the center position of the pixel of the detection frame of the corresponding target area or target object, the size of the detection frame, the outline, and other information.

[0037] Following the previous example, see Figure 2As shown, a pre-trained YOLOv7 model can be pre-deployed on the cloud server 204. After acquiring visible light and infrared images of the construction site from multiple target cameras 201 within a preset time period from the IoT platform 203, the cloud server 204 can use the YOLOv7 model to perform image segmentation and target detection on the visible light and infrared images at the same time to identify target areas and objects. This allows the cloud server 204 to obtain information such as the category, pixel center coordinates of the detection box, detection box size, and contour for each target area and object within the preset time period. Because the YOLOv7 model can utilize the E-ELAN module to enhance feature diversity, its use ensures high recognition accuracy under complex lighting conditions, thus guaranteeing accurate identification of dust particle areas, green netting covered areas, material stacking areas, materials, and personnel even under complex lighting conditions.

[0038] As one possible implementation, the comprehensive score of the construction site within a preset time period in step S106 above, based on a preset multidimensional scoring model, environmental monitoring data, detection and identification results, and air pollution source intensity information, may include:

[0039] Step A1 involves preprocessing the environmental monitoring data and using a pre-set multi-dimensional scoring model to determine the basic indicator scores of the construction site within a preset time period.

[0040] Preprocessing can include noise reduction and normalization.

[0041] For example, the preprocessed environmental monitoring data may include standardized atmospheric particulate matter concentration data and standardized noise data within a preset time period. After preprocessing (including noise reduction and standardization) the atmospheric particulate matter concentration data and noise data of each monitoring point at the construction site within the preset time period to obtain standardized atmospheric particulate matter concentration data and standardized noise data within the preset time period, the standardized atmospheric particulate matter concentration data and standardized noise data can be input into a preset multidimensional scoring model for statistical analysis, so as to determine the basic index score based on the statistical analysis results through the preset multidimensional scoring model. The statistical analysis results may include the average atmospheric particulate matter concentration, average noise, and abnormal duration within the preset time period. The abnormal duration may include the first abnormal duration where the atmospheric particulate matter concentration is greater than a preset concentration threshold and the second abnormal duration where the noise is greater than a preset noise threshold.

[0042] Following the previous example, see Figure 2 As shown, the calculation process for the basic indicators can be carried out as follows: Cloud server 204 obtains the sensor 202 corresponding to each monitoring point on the construction site (including PM) from IoT platform 203. 2.5 Sensors, PM10 PM2.5 concentration collected over a period of time by sensors, noise sensors, etc. 2.5 Concentration data, PM 10 Concentration data and noise data, etc., are collected and denoised to obtain standardized PM2.5. 2.5 Concentration data, standardized PM 10 Concentration data and standardized noise data, etc.; then the cloud server 204 statistically analyzes the standardized PM2.5 levels during this period. 2.5 Concentration data, standardized PM 10 The average, maximum, and minimum values ​​of both concentration data and standardized noise data were calculated, and the standardized PM2.5 concentrations for this period were statistically analyzed. 2.5 Concentration data, standardized PM 10 The concentration data and the standardized noise data each correspond to the abnormal moments and durations exceeding the respective thresholds; subsequently, cloud server 204 represents the standardized PM... 2.5 Concentration data, standardized PM 10 The average value and anomaly duration of the concentration data and standardized noise data are each assigned corresponding weights. Then, the cloud server 204 calculates the standardized values ​​according to the following formula and performs a weighted average to calculate the basic indicator score of the construction site during this period:

[0043]

[0044] in, Basic indicator scores, , and Standardized PM 2.5 Concentration data, standardized PM 10 The average values ​​corresponding to both concentration data and standardized noise data. , and Standardized PM 2.5 Concentration data, standardized PM 10 The duration of anomalies corresponding to both concentration data and standardized noise data. , , , , and They are respectively , , , , and Each has its own assigned weight.

[0045] Step A2: Determine the process indicator score of the construction site within a preset time period using a preset multidimensional scoring model that incorporates category and pixel information.

[0046] For example, the target area may include a dust particle area, a green netting covered area, and a material storage area, and the target object may include personnel. After obtaining the pixel information (including the coordinates of the center position of the detection frame pixel, the size of the detection frame, the outline, etc.) of the dust particle area, the green netting covered area, the material storage area, and the personnel, a preset multidimensional scoring model can be used to determine the dust particle coverage rate of the construction site within a preset time period using the pixel information of the dust particle area, the green netting coverage rate of the construction site within a preset time period using the pixel information of the green netting covered area, the proportion of illegally stored material area of ​​the construction site within a preset time period using the pixel information of the material storage area, and whether each person on the construction site is wearing a safety helmet within a preset time period using the pixel information of the personnel. The preset multidimensional scoring model can also be used to count the number of personnel not wearing safety helmets within a preset time period. Finally, the preset multidimensional scoring model can be used to determine the process indicator score using the dust particle coverage rate, the green netting coverage rate, the proportion of illegally stored material area, and the number of personnel.

[0047] Following the previous example, see Figure 2 As shown, the calculation process of the process indicators can be performed as follows: Cloud server 204 uses the YOLOv7 model deployed on it to detect and identify the video stream obtained from IoT platform 203 to obtain information such as the center coordinates of the pixel position of the detection frame, the size of the detection frame, and the outline of the dust particle area, the green net coverage area, the material stacking area, and the personnel; then, cloud server 204 uses the multi-dimensional scoring model deployed on it to calculate the total area of ​​all dust particle areas using the center coordinates of the pixel position of the detection frame, the size of the detection frame, and the outline of the dust particle area. and the total area of ​​all monitored areas The dust particle coverage rate was calculated using a multidimensional scoring model. Then, cloud server 204 uses a multi-dimensional scoring model to calculate the total area of ​​all green net-covered regions by utilizing information such as the center coordinates of the pixels in the detection boxes, the size of the detection boxes, and their outlines. The green network coverage rate was calculated using a multi-dimensional scoring model. Subsequently, cloud server 204 uses a multi-dimensional scoring model to calculate the total area of ​​all illegally stacked materials by utilizing information such as the center coordinates of the pixels in the detection frame, the size of the detection frame, and its outline. and the total area of ​​all material storage areas. The proportion of illegally stacked materials was calculated using a multi-dimensional scoring model. Subsequently, cloud server 204 uses a multi-dimensional scoring model to determine whether each person is wearing a safety helmet by utilizing information such as the pixel center coordinates of the detection box, the size of the detection box, and the outline. The multi-dimensional scoring model also counts the number of people not wearing safety helmets. Subsequently, cloud server 204 used a multi-dimensional scoring model to apply pre-set scoring criteria and the obtained dust particle coverage rate. Green net coverage rate Percentage of area where materials are illegally piled up Number of personnel Calculate the process indicator scores.

[0048] The above scoring criteria may include: a maximum score of 100 points, and dust particle coverage rate. Deduct 5 points for every 10% increase in green network coverage. Deduct 10 points for every 10% decrease; the number of people not wearing safety helmets. For each additional violation, 5 points will be deducted; the percentage of illegally stacked materials will be affected. For every 10% increase, deduct 10 points. Therefore, the process indicator score... .

[0049] In practical applications, the target detection model (such as the YOLOv7 model mentioned above) can be updated quarterly to adapt to seasonal changes.

[0050] Step A3: Calculate the average value between the upper and lower limits of the intensity range of air pollution sources in each monitoring area, and use the area and average value of each monitoring area to determine the derived index score of the construction site within the preset time period through a preset multidimensional scoring model.

[0051] For example, after obtaining the average value between the upper and lower limits of the intensity range of air pollution sources in each monitoring area, a first weight can be assigned to the average value of each monitoring area according to the area size of the monitoring area through a preset multidimensional scoring model. The product value of the average value of each monitoring area and its corresponding first weight can be calculated through the preset multidimensional scoring model. Then, the derived index score can be determined by using the product value corresponding to each monitoring area through the preset multidimensional scoring model.

[0052] Following the previous example, see Figure 2 As shown, the calculation process for the derived indicators can be performed as follows: Cloud server 204 calculates the upper limit of the intensity range of air pollution sources corresponding to each monitoring area. and lower limit average And the average value obtained for each monitoring area Input the multidimensional scoring model to score each monitored area according to its size. The average value corresponding to each monitoring area Assign corresponding weights The larger the monitoring area, the higher the average value. It will be assigned a larger weight Then, cloud server 204 calculates the average value corresponding to each monitoring area using a multi-dimensional scoring model. Its assigned weight product value Then, the derived index scores were calculated using a multidimensional scoring model. .

[0053] Step A4: Determine the comprehensive score by using the basic indicator score, process indicator score, and derived indicator score through a preset multidimensional scoring model.

[0054] For example, after obtaining the basic indicator score, process indicator score, and derived indicator score of the construction site within a preset time period, a corresponding second weight can be assigned to each of the basic indicator score, process indicator score, and derived indicator score according to the environmental state of the construction site within the preset time period. Then, the first product value of the basic indicator score and its corresponding second weight, the second product value of the process indicator score and its corresponding second weight, and the third product value of the derived indicator score and its corresponding second weight are calculated by the preset multidimensional scoring model. Finally, the comprehensive score of the construction site within the preset time period is determined by using the first product value, the second product value, and the third product value.

[0055] Following the previous example, see Figure 2 As shown, the calculation process for the comprehensive score can be carried out as follows: The basic indicator scores of the construction site over a period of time are calculated using a multi-dimensional scoring model deployed on cloud server 204. Process indicator scores and derived indicator scores Subsequently, cloud server 204 uses a multi-dimensional scoring model to assess the environmental conditions of the construction site within a preset time period. for , and Each is assigned a corresponding weight ( , and The corresponding weights are respectively , and ), different environmental conditions Corresponding to include , and With different weight combinations, cloud server 204 then calculates the first product value using a multidimensional scoring model. Second product value and the third product value Then, the overall score is calculated. .

[0056] As an application example, a construction site mainly has a normal state within a preset time period (which can be denoted as...). ), warning status (can be recorded as) ) and emergency (can be written as These three environmental states can then be dynamically adjusted based on the environmental state. , and The respective weights: under normal conditions on the construction site In this case, , and The corresponding weights are respectively , and ,For example , and The values ​​can be 0.5, 0.3, and 0.2 respectively; under the condition of being in a warning state at the construction site, , and The corresponding weights are respectively , and ,For example , and The values ​​can be 0.3, 0.2, and 0.5 respectively; in case of an emergency at the construction site, , and The corresponding weights are respectively , and ,For example , and The values ​​can be 0.4, 0.4, and 0.2 respectively.

[0057] As one possible implementation, the aforementioned satellite imagery data may include the geographical locations corresponding to each monitoring area, and the aforementioned construction progress information may include multiple different work periods (such as tower crane operation periods, earthwork transportation periods, etc.) and the work areas corresponding to each work period (such as tower crane operation areas, earthwork transportation areas, etc.); based on this, the aforementioned step S108 (i.e., determining the key pollution information at the construction site based on atmospheric pollution source intensity information and satellite imagery data and construction progress information at the construction site) may include:

[0058] Step a1: Based on the intensity range and geographical location of the air pollution sources corresponding to each monitoring area, generate an air pollution heat map of the construction site within a preset time period.

[0059] Among them, the atmospheric pollution heat map can characterize the geographical distribution of the intensity of atmospheric pollution sources at the construction site.

[0060] Following the previous example, see Figure 2 As shown, when the cloud server 204 obtains the atmospheric pollution source intensity range corresponding to each monitoring area at multiple different times within a certain period and acquires satellite image data of the construction site from the satellite image database 206, the cloud server 204 obtains the geographical location corresponding to each monitoring area from the satellite image data and associates the atmospheric pollution source intensity range and geographical location corresponding to the same monitoring area at each time within this period. Then, it uses the associated information of each monitoring area at each time within this period to generate atmospheric pollution heat maps corresponding to multiple different times within this period. The atmospheric pollution heat map corresponding to each time can reflect the geographical location distribution of the monitoring area at the construction site and the atmospheric pollution source intensity range corresponding to each monitoring area at that time through special identifiers (such as color identifiers, font identifiers, etc.).

[0061] Step a2 involves binding the atmospheric pollution heat map and construction progress information of the construction site within the same time period to obtain the bound information of the construction site in multiple different time periods, and determining the key operation periods and key operation areas based on the bound information.

[0062] Following the previous example, see Figure 2As shown, after the cloud server 204 obtains the atmospheric pollution heat maps corresponding to multiple different times within a certain period, the cloud server 204 obtains the construction progress information of the construction site from the construction database 207 (including multiple different work periods and the work areas corresponding to each work period). Then, the cloud server 204 matches the time information corresponding to each atmospheric pollution heat map with the work periods included in the construction progress information to form a corresponding heat map set of atmospheric pollution heat maps that match the same work period. Then, the cloud server 204 binds each heat map set with its corresponding work period and work area. After that, the cloud server 204 analyzes the obtained bound information to determine the key work periods and key work areas that generate pollution at the construction site (for example, the key work period can be the work period when the lower limit of the atmospheric pollution source intensity range exceeds the preset source intensity threshold, and the key work area can be all work areas corresponding to the corresponding key work period).

[0063] As one possible implementation, the aforementioned historical construction data may include historical violation records, historical construction logs, and historical equipment usage records at the construction site. Based on this, the pollution cause analysis in step S110 based on environmental monitoring data and historical construction data at the construction site may include: inputting environmental monitoring data, historical violation records, historical construction logs, and historical equipment usage records into a pre-trained Bayesian network model for pollution cause analysis, so as to output multiple probability values ​​corresponding to different pollution causes as pollution cause analysis results through the Bayesian network model.

[0064] Following the previous example, see Figure 2 As shown, a pre-trained Bayesian network model can be deployed on the cloud server 204 in advance. The cloud server 204 obtains the PM data of each monitoring point over a period of time from the IoT platform 203. 2.5 Concentration data, PM 10 After obtaining concentration data, noise data, and other information, and retrieving historical violation records, historical construction logs, and historical equipment usage records from the construction database 207, the cloud server 204 can then use the obtained PM data. 2.5 Concentration data, PM 10 Concentration data, noise data, historical violation records, historical construction logs, and historical equipment usage records are input into a Bayesian network model to perform pollution cause analysis calculations, so that the probability values ​​corresponding to multiple different pollution causes can be obtained through the output of the Bayesian network model.

[0065] The computational process of using Bayesian network models for pollution causation analysis can be mainly divided into the following parts:

[0066] (1) Identify key nodes.

[0067] Key nodes can specifically include pollution incidents, equipment status, construction activities, management measures, and environmental monitoring.

[0068] (2) Variable definition and discretization.

[0069] For historical violation records, the variable can be defined as the number of violations, and categorized into 0 violations, 1-3 violations, and more than 3 violations; for historical equipment usage records, the variable can be defined as the equipment's service life, and categorized into less than 3 years, 3-5 years, and more than 5 years; for PM... 2.5 For concentration data, variables can be defined as air quality levels, which can be categorized as excellent, good, lightly polluted, heavily polluted, etc. For historical construction logs, variables can be defined as job type and number of workers, with job type categorized as earthwork, structure, decoration, etc.

[0070] (3) Calculate the probability of pollution causes.

[0071] When new environmental monitoring data (such as PM2.5) is received 2.5 Concentration data, PM 10 When processing new environmental monitoring data (such as concentration data and noise data), along with historical violation records, construction logs, and equipment usage records from the construction site, the Bayesian network model is input into a pre-trained model. The model then performs the following operations: identifying key nodes (e.g., pollution events, equipment status, construction activities, management measures, environmental monitoring, etc.); determining the number of violations based on historical violation records (e.g., 0 times, 1-3 times, more than 3 times, etc.); determining the equipment's service life based on historical equipment usage records (e.g., less than 3 years, 3-5 years, more than 5 years, etc.); and determining the PM2.5 concentration data. 2.5 Concentration data is used to determine air quality levels (e.g., excellent, good, lightly polluted, heavily polluted, etc.); based on historical construction logs, the type of work (e.g., earthwork, structure, decoration, etc.) and the number of workers are determined; based on key nodes, number of violations, equipment service life, air quality levels, work types, and the number of workers, the probability value corresponding to each of the preset multiple different pollution causes is calculated.

[0072] For example, when PM is detected 2.5 When the concentration suddenly increased, the Bayesian network model calculated that the main causes of the phenomenon included equipment failure, illegal operation, and meteorological factors. The probabilities of equipment failure, illegal operation, and meteorological factors causing the phenomenon were 68%, 25%, and 7%, respectively.

[0073] Because Bayesian network models are particularly well-suited for handling uncertainties in the causes of construction site pollution, they can integrate quantitative environmental monitoring data and qualitative management information (i.e., historical construction data) to output specific probability values, providing decision support for precise pollution control at construction sites.

[0074] In practical applications, when performing the above steps of collecting environmental monitoring data of the construction site within a preset time period by deploying multiple sensors on the construction site, the collection frequency of the corresponding sensors at each monitoring point can be dynamically adjusted according to whether the environmental monitoring data meets the preset conditions.

[0075] Whether the environmental monitoring data meets the preset conditions may include: the concentration of atmospheric particulate matter within the preset time period reaches the level of heavy pollution in air quality, and the noise within the preset time period reaches the level of heavy pollution in noise pollution.

[0076] For example, a sensor (such as PM) can be set for each monitoring point. 2.5 Sensors, PM 10 Sensors, noise sensors, etc., by default collect relevant environmental monitoring data (such as PM2.5) at a frequency of once per minute. 2.5 Concentration data, PM 10 (Concentration data and noise data, etc.) If the environmental monitoring data collected by a certain sensor reaches the level of severe pollution, the frequency of the sensor collecting its corresponding environmental monitoring data can be adjusted to a frequency of once every 5 seconds.

[0077] Furthermore, corresponding atmospheric pollution source intensity thresholds can be preset. This allows for dynamic adjustment of the frequency at which atmospheric particulate matter concentration sensors in each monitoring area collect data based on whether the lower and / or upper limits of the atmospheric pollution source intensity range corresponding to each monitoring area exceed the preset thresholds. For example, if the lower and / or upper limits of the atmospheric pollution source intensity range corresponding to a target monitoring area exceed the preset thresholds, the frequency at which atmospheric particulate matter concentration sensors at all monitoring points in the target monitoring area collect data will be increased. Meanwhile, atmospheric particulate matter concentration sensors in monitoring areas where neither the lower nor upper limit of the atmospheric pollution source intensity range exceeds the preset thresholds will collect data at the default frequency.

[0078] The beneficial effects of the above-mentioned methods for analyzing pollution sources at construction sites can be mainly reflected in the following aspects:

[0079] 1) Automated environmental monitoring improves efficiency and accuracy compared to manual monitoring.

[0080] 2) It enables automated scoring at construction sites, improving the efficiency and accuracy of pollution assessment at construction sites;

[0081] 3) It enables automated pollution cause analysis, improving the efficiency and accuracy of tracing the source of pollution incidents at construction sites;

[0082] 4) It enables the automated identification of critical operation periods and areas that generate pollution, providing guidance for pollution prevention and control at construction sites.

[0083] Based on the above-mentioned method for analyzing pollution sources at construction sites, this invention also provides a system for analyzing pollution sources at construction sites. (See attached document.) Figure 2 As shown, the system may include: multiple target cameras 201 and multiple sensors 202 deployed at the construction site, as well as an Internet of Things (IoT) platform 203 and a cloud server 204. The IoT platform 203 is connected to the multiple target cameras 201, the multiple sensors 202 and the cloud server 204 respectively. The target cameras 201 are integrated cameras with visible light and infrared sensors. The cloud server 204 is equipped with a preset Gaussian diffusion model, a preset multidimensional scoring model and a pre-trained target detection model. The cloud server 204 is also connected to a meteorological database 205, a satellite image database 206 and a construction database 207.

[0084] The multiple target cameras 201 can be used to collect video streams from the construction site within a preset time period;

[0085] The multiple sensors 202 can be used to collect environmental monitoring data at the construction site within a preset time period; wherein, the environmental monitoring data includes atmospheric particulate matter concentration data and noise data at each monitoring point within the preset time period;

[0086] The cloud server 204 can be used to: acquire the video stream and the environmental monitoring data through the Internet of Things platform 203, and acquire the meteorological data of the construction site within a preset time period from the meteorological database 205; and determine the atmospheric pollution source intensity information of each monitoring area of ​​the construction site within a preset time period based on the preset Gaussian diffusion model, the environmental monitoring data, and the meteorological data.

[0087] The cloud server 204 can also be used to: detect and identify the video stream through the target detection model, and determine the comprehensive score of the construction site within a preset time period based on the preset multi-dimensional scoring model, the environmental monitoring data, the detection and identification results, and the atmospheric pollution source intensity information;

[0088] The cloud server 204 can also be used to: obtain satellite image data of the construction site from the satellite image database 206, and obtain construction progress information of the construction site from the construction database 207; and determine key pollution information of the construction site based on the atmospheric pollution source intensity information, the satellite image data, and the construction progress information; wherein, the key pollution information includes key operation periods and key operation areas that generate pollution.

[0089] The cloud server 204 can also be used to: obtain historical construction data from the construction database 207; perform pollution cause analysis based on the environmental monitoring data and the historical construction data; and determine the pollution source analysis results at the construction site based on the comprehensive score, the key pollution information, and the pollution cause analysis results.

[0090] The construction site pollution source analysis system provided in this embodiment of the invention has the same implementation principle and technical effects as the aforementioned construction site pollution source analysis method embodiment. For the sake of brevity, any parts not mentioned in the construction site pollution source analysis system embodiment can be referred to the corresponding content in the aforementioned construction site pollution source analysis method embodiment.

[0091] Unless otherwise specifically stated, the relative steps, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention.

[0092] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0093] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0094] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered 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 analyzing pollution sources at construction sites, characterized in that, include: Video streams of the construction site are collected within a preset time period by multiple target cameras deployed at the construction site, and environmental monitoring data of the construction site within a preset time period are collected by multiple sensors deployed at the construction site; wherein, the target cameras are integrated cameras with visible light sensors and infrared sensors, and the environmental monitoring data includes atmospheric particulate matter concentration data and noise data of each monitoring point within the preset time period. Based on a preset Gaussian diffusion model, the environmental monitoring data, and meteorological data from the construction site within a preset time period, the atmospheric pollution source intensity information for each monitoring area at the construction site within the preset time period is determined; wherein, the atmospheric pollution source intensity information includes the range of atmospheric pollution source intensity for the corresponding monitoring area within the preset time period; The video stream is detected and identified by a pre-trained target detection model to obtain the category and pixel information of each target region and each target object within a preset time period; The environmental monitoring data is preprocessed, and the basic index scores of the construction site within a preset time period are determined using the preprocessed environmental monitoring data through a preset multidimensional scoring model; wherein, the preprocessing includes noise reduction and standardization; The process indicator score of the construction site within a preset time period is determined by using the category and pixel information through the preset multidimensional scoring model. Calculate the average value between the upper and lower limits of the intensity range of air pollution sources in each monitoring area, and use the area and average value of each monitoring area to determine the derived index score of the construction site within a preset time period using the preset multidimensional scoring model; The comprehensive score of the construction site within a preset time period is determined by using the basic index score, the process index score, and the derived index score through the preset multidimensional scoring model. Based on the atmospheric pollution source intensity information, satellite imagery data of the construction site, and construction progress information, key pollution information at the construction site is determined; wherein, the key pollution information includes key operation periods and key operation areas that generate pollution; Pollution cause analysis is conducted based on the environmental monitoring data and historical construction data at the construction site, and the pollution source analysis results at the construction site are determined based on the comprehensive score, the key pollution information, and the pollution cause analysis results.

2. The method for analyzing pollution sources at construction sites according to claim 1, characterized in that, The meteorological data includes wind speed data at the construction site during a preset time period; based on a preset Gaussian diffusion model, the environmental monitoring data, and the meteorological data at the construction site during the preset time period, the intensity information of air pollution sources in each monitoring area of ​​the construction site during the preset time period is determined, including: Based on the wind speed data, the diffusion parameters of the preset Gaussian diffusion model corresponding to each monitoring point are determined. The atmospheric pollution source intensity value of each monitoring point within a preset time period is calculated using the preset Gaussian diffusion model with the diffusion parameters and the atmospheric particulate matter concentration data. Based on the atmospheric pollution source intensity values ​​of all monitoring points in the same monitoring area, the atmospheric pollution source intensity range corresponding to that monitoring area is determined.

3. The method for analyzing pollution sources at construction sites according to claim 2, characterized in that, The video stream includes visible light images and infrared images at multiple different times within a preset time period; the video stream is detected and identified using a pre-trained target detection model, including: The target detection model performs image segmentation and target detection on visible light and infrared images at the same time to identify target areas and target objects, and obtains the category and pixel information of each target area and each target object within a preset time period; wherein, the target area includes at least one of dust particle area, green netting covered area and material stacking area, and the target object includes materials and / or personnel.

4. The method for analyzing pollution sources at construction sites according to claim 3, characterized in that, The preprocessed environmental monitoring data includes standardized atmospheric particulate matter concentration data and standardized noise data within a preset time period. The pre-processed environmental monitoring data is used to determine the basic indicator scores of the construction site within a preset time period through the pre-defined multi-dimensional scoring model, including: The standardized atmospheric particulate matter concentration data and the standardized noise data are input into the preset multidimensional scoring model for statistical analysis, so as to determine the basic index score based on the statistical analysis results through the preset multidimensional scoring model; wherein, the statistical analysis results include the average atmospheric particulate matter concentration, average noise and abnormal duration within a preset time period, and the abnormal duration includes the first abnormal duration when the atmospheric particulate matter concentration is greater than a preset concentration threshold and the second abnormal duration when the noise is greater than a preset noise threshold.

5. The method for analyzing pollution sources at construction sites according to claim 3, characterized in that, The target area includes areas with dust particles, areas covered by green netting, and areas where materials are piled up; the target objects include personnel; the process indicator scores of the construction site within a preset time period are determined using the preset multidimensional scoring model based on the categories and pixel information, including: The dust particle coverage rate of the construction site within a preset time period is determined by using the pixel information of the dust particle area through the preset multidimensional scoring model. The green net coverage rate of the construction site within a preset time period is determined by using the pixel information of the green net coverage area through the preset multidimensional scoring model. The preset multidimensional scoring model is used to determine the percentage of illegally stacked material area at the construction site within a preset time period by utilizing the pixel information of the material stacking area. The preset multidimensional scoring model is used to determine whether each person on the construction site is wearing a safety helmet within a preset time period using the pixel information of the personnel, and the preset multidimensional scoring model is used to count the number of people who are not wearing safety helmets within the preset time period. The process indicator score is determined by the preset multidimensional scoring model using the dust particle coverage rate, the green net coverage rate, the proportion of illegally piled material area, and the number of personnel.

6. The method for analyzing pollution sources at construction sites according to claim 3, characterized in that, The pre-defined multi-dimensional scoring model uses the area and average value of each monitoring area to determine the derived index score of the construction site within a pre-defined time period, including: The preset multidimensional scoring model assigns a corresponding first weight to the average value of each monitoring area according to the size of the monitoring area, and calculates the product of the average value of each monitoring area and its corresponding first weight using the preset multidimensional scoring model. Then, the preset multidimensional scoring model uses the product value corresponding to each monitoring area to determine the derived index score.

7. The method for analyzing pollution sources at construction sites according to claim 3, characterized in that, The satellite imagery data includes the geographical locations corresponding to each monitoring area, and the construction progress information includes multiple different work periods and the work area corresponding to each work period; Based on the atmospheric pollution source intensity information, satellite imagery data of the construction site, and construction progress information, key pollution information at the construction site is determined, including: Based on the intensity range and geographical location of air pollution sources corresponding to each monitoring area, an air pollution heat map of the construction site is generated within a preset time period; wherein, the air pollution heat map represents the geographical distribution of the intensity of air pollution sources at the construction site. By binding the atmospheric pollution heat map and construction progress information of the construction site at the same time period, the bound information of the construction site at multiple different time periods is obtained, and the key operation period and the key operation area are determined based on the bound information.

8. The method for analyzing pollution sources at construction sites according to claim 3, characterized in that, The historical construction data includes historical violation records, historical construction logs, and historical equipment usage records at the construction site; based on the environmental monitoring data and the historical construction data at the construction site, pollution cause analysis is performed, including: The environmental monitoring data, historical violation records, historical construction logs, and historical equipment usage records are input into a pre-trained Bayesian network model for pollution cause analysis. The probability values ​​corresponding to multiple different pollution causes are output by the Bayesian network model as the pollution cause analysis results.

9. A construction site pollution source analysis system, characterized in that, include: The system includes multiple target cameras and sensors deployed at the construction site, as well as an IoT platform and a cloud server. The IoT platform is connected to the multiple target cameras, the multiple sensors, and the cloud server. The target cameras are integrated cameras that combine visible light and infrared sensors. The cloud server is equipped with a preset Gaussian diffusion model, a preset multidimensional scoring model, and a pre-trained target detection model. The cloud server is also connected to a meteorological database, a satellite image database, and a construction database. The multiple target cameras are used to collect video streams from the construction site within a preset time period; The multiple sensors are used to collect environmental monitoring data at the construction site within a preset time period; wherein, the environmental monitoring data includes atmospheric particulate matter concentration data and noise data at each monitoring point within the preset time period; The cloud server is used to: acquire the video stream and the environmental monitoring data through the IoT platform, and acquire meteorological data of the construction site within a preset time period from the meteorological database; and determine the atmospheric pollution source intensity information of each monitoring area at the construction site within a preset time period based on the preset Gaussian diffusion model, the environmental monitoring data, and the meteorological data; wherein the atmospheric pollution source intensity information includes the atmospheric pollution source intensity range of the corresponding monitoring area within the preset time period; The cloud server is also used for: detecting and identifying the video stream using the target detection model to obtain the category and pixel information of each target area and each target object within a preset time period; preprocessing the environmental monitoring data and determining the basic index score of the construction site within the preset time period using the preprocessed environmental monitoring data through the preset multidimensional scoring model; wherein, the preprocessing includes noise reduction and standardization; determining the process index score of the construction site within the preset time period using the category and pixel information through the preset multidimensional scoring model; calculating the average value between the upper and lower limits of the atmospheric pollution source intensity range of each monitoring area, and determining the derived index score of the construction site within the preset time period using the area and average value of each monitoring area through the preset multidimensional scoring model; and determining the comprehensive score of the construction site within the preset time period using the basic index score, the process index score, and the derived index score through the preset multidimensional scoring model. The cloud server is also used to: obtain satellite image data of the construction site from the satellite image database, and obtain construction progress information of the construction site from the construction database; and determine key pollution information of the construction site based on the atmospheric pollution source intensity information, the satellite image data, and the construction progress information; wherein, the key pollution information includes key operation periods and key operation areas that generate pollution. The cloud server is also used to: obtain historical construction data from the construction database; perform pollution cause analysis based on the environmental monitoring data and the historical construction data; and determine the pollution source analysis results at the construction site based on the comprehensive score, the key pollution information, and the pollution cause analysis results.

Images