Atmospheric pollution source tracing system and method based on unmanned aerial vehicle

Abstract

The invention discloses an atmospheric pollution source tracing system and method based on an unmanned aerial vehicle. The system comprises an airborne atmospheric monitoring module used for acquiring atmospheric data; the ground station system is used for monitoring whether the flight state of the unmanned aerial vehicle is normal or not, sending state information to the unmanned aerial vehicle control platform and receiving and storing atmosphere data; the airborne atmosphere monitoring module is loaded on the unmanned aerial vehicle, the unmanned aerial vehicle is controlled to fly through the unmanned aerial vehicle control platform, and the unmanned aerial vehicle control platform controls the unmanned aerial vehicle in real time according to state information; the data processing module is used for preprocessing the atmosphere data; and the traceability analysis module is used for carrying out pollution source traceability analysis according to the preprocessed atmosphere data to obtain the position of the pollution source. According to the invention, accurate pollution source positioning and pollutant diffusion simulation provide a scientific basis for the formulation and execution of environmental policies, and the air pollution can be more effectively controlled and reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of pollutant source tracing, and in particular relates to an air pollution source tracing system and method based on an unmanned aerial vehicle. Background Art

[0002] In atmospheric environmental governance, tracing the source of atmospheric pollutants has always been a difficult problem. At present, there is still a lack of effective technical means to accurately locate the source of atmospheric pollution emissions, resulting in blind spots in the monitoring and supervision of atmospheric pollutant emissions, and the inability to effectively control atmospheric pollutant emissions. Among the existing atmospheric pollution source tracing technical methods, atmospheric pollutant monitoring and tracing are mainly carried out through remote sensing, ground-based monitoring stations and mobile monitoring vehicles. Among them, remote sensing monitoring has low cost and wide range, but there are problems of poor monitoring effectiveness and low accuracy; while the use of ground monitoring equipment for monitoring has problems of high cost and uneven distribution of monitoring points; mobile monitoring vehicles are a mobile monitoring method that can overcome the problem of few monitoring points and can verify and trace the source of atmospheric pollutants to a certain extent, but the urban underlying surface causes the ground wind field to be more complex, so it is impossible to accurately trace the source of pollutants through the monitoring results of mobile monitoring vehicles. Summary of the invention

[0003] In order to solve the above technical problems, the present invention proposes an air pollution source tracing system and method based on drones to solve the problems existing in the above-mentioned prior art.

[0004] To achieve the above objectives, the present invention provides an air pollution source tracing system based on a drone, comprising:

[0005] UAV control platform, data processing module, source tracing analysis module, airborne atmosphere monitoring module and ground station system;

[0006] The onboard atmospheric monitoring module is used to obtain atmospheric data; the ground station system is used to monitor whether the UAV flight status is normal and send status information to the UAV control platform, and receive atmospheric data for storage;

[0007] The airborne atmosphere monitoring module is loaded on a UAV, and the UAV is controlled to fly by the UAV control platform, and the UAV control platform controls the UAV in real time according to the status information;

[0008] The data processing module is used to pre-process the atmospheric data;

[0009] The source tracing analysis module is used to perform pollution source tracing analysis based on the preprocessed atmospheric data to obtain the location of the pollution source.

[0010] Optionally, the airborne atmosphere monitoring module is used to obtain atmospheric data through sensor equipment, and the sensor equipment includes a fine particulate matter sensor, a temperature sensor, a humidity sensor, an air pressure sensor, an odorous gas sensor, a toxic gas sensor, a combustible gas sensor and a conventional gas sensor.

[0011] Optionally, two conventional gas sensors are provided, one is installed at the central part of the bottom of the drone, and the other is installed at the end of a detection rod, and the other end of the detection rod is fixedly connected to the bottom of the drone.

[0012] Optionally, the drone control platform includes a data acquisition unit and a route planning unit;

[0013] The data acquisition unit is used to delineate the area to be detected, obtain an electronic map of the area to be detected, and extract building distribution information and terrain information;

[0014] The route planning unit is used to divide the area to be detected into several sub-areas according to functional characteristics, divide the sub-areas into grids, and use the grid centers as initial data collection points; obtain historical detection data, prioritize the grids in each sub-area based on the historical detection data, and generate an initial flight route; perform obstacle avoidance processing on the initial flight route according to the UAV flight height limit, building distribution information and terrain information to obtain the final flight route.

[0015] Optionally, the data processing module includes a verification unit, a preprocessing unit, and a marking unit;

[0016] The verification unit is used to verify the integrity of the atmospheric data and the validity of the timestamp, and pre-process the data after the verification;

[0017] The preprocessing unit deletes abnormal values ​​in the atmospheric data and fills in missing values ​​by linear interpolation, and then performs normalization processing;

[0018] The marking unit is used to determine whether there is a difference between the gas concentration values ​​obtained by two conventional gas sensors, and if there is a difference, determine whether it exceeds a set threshold, and if so, mark it.

[0019] Optionally, the traceability analysis module includes an information extraction unit and an analysis unit;

[0020] The information extraction unit is used to obtain meteorological information within a time range corresponding to the atmospheric data;

[0021] The analysis unit is used to analyze the location of the pollution source according to the meteorological information and the pre-processed atmospheric data.

[0022] Optionally, the analysis unit includes a distribution range determination subunit, a simulation subunit, and a judgment subunit;

[0023] The distribution range determination subunit is used to import the pre-processed atmospheric data into the geographic information system, perform spatial distribution analysis on the atmospheric data, generate a pollutant concentration distribution map, identify abnormal areas, and preliminarily determine the distribution range of the pollution source;

[0024] The simulation subunit is used to select an atmospheric diffusion model, input the meteorological information and the atmospheric data into the atmospheric diffusion model, and generate a simulation data set; based on the grid where the data collection point corresponding to the marked data is located, obtain the minimum distance between the grid and the edge point of the distribution range, and determine whether the minimum distance meets the requirements based on the simulation data set. If it does not meet the requirements, the corresponding grid is included in the distribution range of the pollution source preliminarily determined.

[0025] The judgment subunit is used to construct a pollution source positioning model, train the pollution source positioning model based on a simulation data set, and obtain the pollution source location through the trained pollution source positioning model and the meteorological information corresponding to the distribution range of the pollution source preliminarily determined and the preprocessed atmospheric data.

[0026] Optionally, an initial pollution source positioning model is constructed based on a convolutional neural network, a residual connection is introduced into the initial pollution source positioning model, and a batch normalization layer is added after each convolution layer.

[0027] The present invention also provides a method for tracing the source of air pollution based on a drone, comprising:

[0028] Acquire atmospheric data through fine particle sensors, temperature sensors, humidity sensors, air pressure sensors, odorous gas sensors, toxic gas sensors, combustible gas sensors and conventional gas sensors;

[0029] Verify and preprocess the atmospheric data, and mark the corresponding data by comparing the gas concentration values ​​obtained by two conventional gas sensors;

[0030] Importing the pre-processed atmospheric data into a geographic information system, performing spatial distribution analysis on the atmospheric data, generating a pollutant concentration distribution map, identifying abnormal areas, and preliminarily determining the distribution range of pollution sources;

[0031] Acquire meteorological information within a time range corresponding to the atmospheric data; select an atmospheric diffusion model, input the meteorological information and the atmospheric data into the atmospheric diffusion model, and generate a simulated data set; based on the grid where the data collection point corresponding to the marked data is located, obtain the minimum distance between the grid and the edge point of the distribution range, and determine whether the minimum distance meets the requirements based on the simulated data set. If it does not meet the requirements, include the corresponding grid in the distribution range of the initially determined pollution source;

[0032] A pollution source positioning model is constructed, and the pollution source positioning model is trained based on a simulated data set, and the location of the pollution source is obtained through the trained pollution source positioning model and the meteorological information corresponding to the distribution range of the pollution source preliminarily determined and the pre-processed atmospheric data.

[0033] Optionally, an initial pollution source positioning model is constructed based on a convolutional neural network, a residual connection is introduced into the initial pollution source positioning model, and a batch normalization layer is added after each convolution layer.

[0034] Compared with the prior art, the present invention has the following advantages and technical effects:

[0035] The present invention uses drones as monitoring platforms to quickly cover large areas. Compared with traditional ground monitoring stations or mobile monitoring vehicles, drones can more flexibly enter hard-to-reach areas and achieve more extensive monitoring. The multi-type sensors carried by drones can simultaneously monitor a variety of atmospheric pollutants, including fine particulate matter, temperature, humidity, air pressure and various gases, enhancing the diversity and accuracy of data collection, which provides rich data support for a comprehensive understanding of atmospheric pollution conditions. The ground station system can monitor the flight status of the drone in real time and promptly feedback information to the drone control platform, ensuring the safety of the flight mission and the continuity of data collection. Through the verification unit, preprocessing unit and marking unit of the data processing module, the collected atmospheric data are effectively cleaned and preprocessed, the data quality is improved, and a reliable data basis is provided for subsequent analysis; the source tracing analysis module combines meteorological information and preprocessed atmospheric data, uses the geographic information system to perform spatial distribution analysis, and generates a pollutant concentration distribution map, so as to more accurately identify the distribution range of abnormal areas and pollution sources; through the atmospheric diffusion model, the diffusion process of pollutants is simulated, and the pollution source is analyzed more deeply in combination with the simulated data set, which improves the prediction ability and accuracy of tracing the source; the pollution source location model constructed by the convolutional neural network improves the model performance through the residual connection and batch normalization layer, so that the pollution source location is more accurate. The accurate pollution source location and pollutant diffusion simulation in the present invention provide a scientific basis for the formulation and implementation of environmental policies, which helps to more effectively control and reduce atmospheric pollution. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0037] Figure 1 A schematic diagram of the system structure of an embodiment of the present invention;

[0038] Figure 2 The figure is a schematic diagram of a method flow of an embodiment of the present invention. DETAILED DESCRIPTION

[0039] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0040] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0041] Embodiment 1

[0042] like Figure 1 As shown, this embodiment provides an air pollution source tracing system based on a drone, including:

[0043] UAV control platform, data processing module, source tracing analysis module, airborne atmosphere monitoring module and ground station system;

[0044] The onboard atmospheric monitoring module is used to obtain atmospheric data; the ground station system is used to monitor whether the UAV flight status is normal and send status information to the UAV control platform, and receive atmospheric data for storage;

[0045] Specifically, the ground station determines whether the drone data transmission is normal based on the data sent by the drone, and determines whether the drone's current flight status is appropriate based on meteorological information. The drone control platform adjusts the route in real time based on the flight status to ensure safety and continuity of data collection.

[0046] In some specific embodiments, the airborne atmosphere monitoring module is used to obtain atmospheric data through sensor equipment, and the sensor equipment includes fine particulate matter sensors, temperature sensors, humidity sensors, air pressure sensors, odorous gas sensors, toxic gas sensors, combustible gas sensors and conventional gas sensors.

[0047] In some specific embodiments, two conventional gas sensors are provided, one is installed at the center of the bottom of the drone, and the other is installed at the end of a detection rod, and the other end of the detection rod is fixedly connected to the bottom of the drone.

[0048] The airborne atmospheric monitoring module is mounted on the UAV, and the UAV is controlled by the UAV control platform. The UAV control platform controls the UAV in real time based on the status information.

[0049] In some specific embodiments, the drone control platform includes a data acquisition unit and a route planning unit;

[0050] The data acquisition unit is used to delineate the area to be inspected, obtain the electronic map of the area to be inspected, and extract the building distribution information and terrain information;

[0051] The route planning unit is used to divide the area to be inspected into several sub-areas according to functional characteristics, divide the sub-areas into grids, and use the grid center as the initial data collection point; obtain historical detection data, prioritize the grids in each sub-area based on the historical detection data, and generate an initial flight route; perform obstacle avoidance processing on the initial flight route according to the drone's flight height limit, building distribution information and terrain information to obtain the final flight route.

[0052] Specifically, a fixed-wing UAV was selected for data collection.

[0053] Specifically, the flight altitude and speed of the drone should be determined based on monitoring requirements and sensor performance. The flight altitude should ensure that the sensor can accurately detect the concentration of pollutants on the ground or in low-altitude areas; the flight speed should be moderate to ensure the continuity and accuracy of data collection.

[0054] The data processing module is used to pre-process the atmospheric data;

[0055] In some specific embodiments, the data processing module includes a verification unit, a preprocessing unit, and a marking unit;

[0056] The verification unit is used to verify the integrity of the atmospheric data and the validity of the timestamp, and pre-process the data after verification;

[0057] The preprocessing unit deletes outliers in the atmospheric data and fills in missing values ​​through linear interpolation, and then performs normalization;

[0058] The marking unit is used to determine whether there is a difference in the gas concentration values ​​obtained by two conventional gas sensors. If there is a difference, it is determined whether it exceeds a set threshold. If it exceeds, it is marked.

[0059] The source tracing analysis module is used to perform pollution source tracing analysis based on the preprocessed atmospheric data to obtain the location of the pollution source.

[0060] In some specific embodiments, the traceability analysis module includes an information extraction unit and an analysis unit;

[0061] The information extraction unit is used to obtain meteorological information within a time range corresponding to the atmospheric data;

[0062] The analysis unit is used to analyze the location of pollution sources based on meteorological information and pre-processed atmospheric data.

[0063] In some specific embodiments, the analysis unit includes a distribution range determination subunit, a simulation subunit, and a judgment subunit;

[0064] The distribution range determination subunit is used to import the pre-processed atmospheric data into the geographic information system, perform spatial distribution analysis on the atmospheric data, generate pollutant concentration distribution maps, identify abnormal areas by comparing concentration data at different times and locations, and preliminarily determine the distribution range of pollution sources;

[0065] The simulation subunit is used to select an atmospheric diffusion model, input meteorological information and atmospheric data into the atmospheric diffusion model, and generate a simulated data set; based on the grid where the data collection point corresponding to the marked data is located, the minimum distance between the grid and the edge point of the distribution range is obtained, and based on the simulated data set, it is determined whether the minimum distance meets the requirements. If it does not meet the requirements, the corresponding grid will be included in the distribution range of the pollution source preliminarily determined.

[0066] Specifically, the route of the next cruise is obtained based on the distribution range of the pollution sources preliminarily determined and the final drone flight route confirmed this time, and the detection route is continuously optimized and adjusted.

[0067] Specifically, the meteorological data includes wind speed, wind direction, temperature, etc., and the atmospheric diffusion model may adopt a Gaussian diffusion model.

[0068] The judgment subunit is used to construct a pollution source location model, which is trained based on a simulated data set. The location of the pollution source is obtained by combining the trained pollution source location model with the meteorological information corresponding to the distribution range of the pollution source and the preprocessed atmospheric data.

[0069] In some specific embodiments, an initial pollution source positioning model is constructed based on a convolutional neural network, a residual connection is introduced into the initial pollution source positioning model, and a batch normalization layer is added after each convolution layer.

[0070] Specifically, the mean square error loss function is used as the loss function for model training. Assume that (x i ,y i ) is the actual location of the pollution source, (y' i ,x' i ) is the position predicted by the model, then the mean square error loss function can be expressed as:

[0071]

[0072] This embodiment also provides a method for tracing air pollution sources based on a drone, including:

[0073] Acquire atmospheric data through fine particle sensors, temperature sensors, humidity sensors, air pressure sensors, odorous gas sensors, toxic gas sensors, combustible gas sensors and conventional gas sensors;

[0074] Verify and preprocess the atmospheric data, and mark the corresponding data by comparing the gas concentration values ​​obtained by two conventional gas sensors;

[0075] Import the pre-processed atmospheric data into the geographic information system, conduct spatial distribution analysis on the atmospheric data, generate pollutant concentration distribution maps, identify abnormal areas, and preliminarily determine the distribution range of pollution sources;

[0076] Obtain meteorological information within the time range corresponding to the atmospheric data; select an atmospheric diffusion model, input the meteorological information and atmospheric data into the atmospheric diffusion model, and generate a simulated data set; based on the grid where the data collection point corresponding to the marked data is located, obtain the minimum distance between the grid and the edge point of the distribution range, and determine whether the minimum distance meets the requirements based on the simulated data set. If not, include the corresponding grid in the distribution range of the initially determined pollution source;

[0077] A pollution source location model is constructed and trained based on a simulated data set. The location of the pollution source is obtained by combining the trained pollution source location model with the meteorological information corresponding to the distribution range of the pollution source and the preprocessed atmospheric data.

[0078] Specifically, an initial pollution source positioning model is constructed based on a convolutional neural network, a residual connection is introduced into the initial pollution source positioning model, and a batch normalization layer is added after each convolution layer.

[0079] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An atmospheric pollution source tracing system based on drones, It is characterized in that The following steps are involved: UAV control platform, data processing module, source tracing analysis module, airborne atmosphere monitoring module and ground station system; The onboard atmospheric monitoring module is used to obtain atmospheric data; the ground station system is used to monitor whether the UAV flight status is normal and send status information to the UAV control platform, and receive atmospheric data for storage; The airborne atmosphere monitoring module is loaded on a UAV, and the UAV is controlled to fly by the UAV control platform, and the UAV control platform controls the UAV in real time according to the status information; The data processing module is used to pre-process the atmospheric data; The source tracing analysis module is used to perform pollution source tracing analysis based on the preprocessed atmospheric data to obtain the location of the pollution source.

2. The air pollution source tracing system based on drone according to claim 1, It is characterized in that The airborne atmosphere monitoring module is used to obtain atmospheric data through sensor equipment, and the sensor equipment includes a fine particle sensor, a temperature sensor, a humidity sensor, an air pressure sensor, a malodorous gas sensor, a toxic gas sensor, a combustible gas sensor and a conventional gas sensor.

3. The air pollution source tracing system based on drone according to claim 2, It is characterized in that The conventional gas sensors are provided with two, one is installed at the central part of the bottom of the drone, and the other is installed at the end of the detection rod, and the other end of the detection rod is fixedly connected to the bottom of the drone.

4. The air pollution source tracing system based on drone according to claim 1, It is characterized in that The UAV control platform includes a data acquisition unit and a route planning unit; The data acquisition unit is used to delineate the area to be detected, obtain an electronic map of the area to be detected, and extract building distribution information and terrain information; The route planning unit is used to divide the area to be detected into several sub-areas according to functional characteristics, divide the sub-areas into grids, and use the grid centers as initial data collection points; obtain historical detection data, prioritize the grids in each sub-area based on the historical detection data, and generate an initial flight route; perform obstacle avoidance processing on the initial flight route according to the UAV flight height limit, building distribution information and terrain information to obtain the final flight route.

5. The air pollution source tracing system based on drone according to claim 3, It is characterized in that The data processing module includes a verification unit, a preprocessing unit, and a marking unit; The verification unit is used to verify the integrity of the atmospheric data and the validity of the timestamp, and pre-process the data after the verification; The preprocessing unit deletes abnormal values ​​in the atmospheric data and fills in missing values ​​by linear interpolation, and then performs normalization processing; The marking unit is used to determine whether there is a difference between the gas concentration values ​​obtained by two conventional gas sensors, and if there is a difference, determine whether it exceeds a set threshold, and if so, mark it.

6. The air pollution source tracing system based on drone according to claim 1, It is characterized in that The traceability analysis module includes an information extraction unit and an analysis unit; The information extraction unit is used to obtain meteorological information within a time range corresponding to the atmospheric data; The analysis unit is used to analyze the location of the pollution source according to the meteorological information and the pre-processed atmospheric data.

7. The air pollution source tracing system based on drone according to claim 6, It is characterized in that The analysis unit includes a distribution range determination subunit, a simulation subunit, and a judgment subunit; The distribution range determination subunit is used to import the pre-processed atmospheric data into the geographic information system, perform spatial distribution analysis on the atmospheric data, generate a pollutant concentration distribution map, identify abnormal areas, and preliminarily determine the distribution range of the pollution source; The simulation subunit is used to select an atmospheric diffusion model, input the meteorological information and the atmospheric data into the atmospheric diffusion model, and generate a simulation data set; Based on the grid where the data collection point corresponding to the marked data is located, the minimum distance between the grid and the edge point of the distribution range is obtained, and whether the minimum distance meets the requirements is determined based on the simulated data set. If it does not meet the requirements, the corresponding grid is included in the distribution range of the initially determined pollution source; The judgment subunit is used to construct a pollution source positioning model, train the pollution source positioning model based on a simulation data set, and obtain the pollution source location through the trained pollution source positioning model and the meteorological information corresponding to the distribution range of the pollution source preliminarily determined and the preprocessed atmospheric data.

8. The air pollution source tracing system based on drone according to claim 7, It is characterized in that An initial pollution source positioning model is constructed based on a convolutional neural network, a residual connection is introduced into the initial pollution source positioning model, and a batch normalization layer is added after each convolution layer.

9. A method for tracing the source of air pollution based on drones. It is characterized in that include: Acquire atmospheric data through fine particle sensors, temperature sensors, humidity sensors, air pressure sensors, odorous gas sensors, toxic gas sensors, combustible gas sensors and conventional gas sensors; Verify and preprocess the atmospheric data, and mark the corresponding data by comparing the gas concentration values ​​obtained by two conventional gas sensors; Importing the pre-processed atmospheric data into a geographic information system, performing spatial distribution analysis on the atmospheric data, generating a pollutant concentration distribution map, identifying abnormal areas, and preliminarily determining the distribution range of pollution sources; Obtain meteorological information within the time range corresponding to the atmospheric data; Selecting an atmospheric diffusion model, inputting the meteorological information and the atmospheric data into the atmospheric diffusion model, and generating a simulation data set; Based on the grid where the data collection point corresponding to the marked data is located, the minimum distance between the grid and the edge point of the distribution range is obtained, and whether the minimum distance meets the requirements is determined based on the simulated data set. If it does not meet the requirements, the corresponding grid is included in the distribution range of the initially determined pollution source; A pollution source positioning model is constructed, and the pollution source positioning model is trained based on a simulated data set, and the location of the pollution source is obtained through the trained pollution source positioning model and the meteorological information corresponding to the distribution range of the pollution source preliminarily determined and the pre-processed atmospheric data.

10. The method for tracing air pollution sources based on drones according to claim 9, It is characterized in that An initial pollution source positioning model is constructed based on a convolutional neural network, a residual connection is introduced into the initial pollution source positioning model, and a batch normalization layer is added after each convolution layer.

Images