Small-scale gas pollutant trajectory tracing method based on airflow movement

Abstract

The invention provides a small-scale gas pollutant trajectory tracing method based on airflow movement, and relates to the technical field of pollutant monitoring. The method comprises the following steps: extracting and simplifying geometric features of ground attachments of a to-be-monitored plant, constructing a plant geometric model, carrying out grid division to determine a computational domain, carrying out airflow movement track simulation by using the plant geometric model, obtaining a plurality of airflow movement track samples, constructing a plant airflow track data set, and carrying out data processing on the airflow movement track data set. The method comprises the following steps: constructing an atmospheric pollutant airflow trajectory model based on a machine learning algorithm, training and testing the atmospheric pollutant airflow trajectory model by using a factory airflow trajectory data set to obtain a qualified atmospheric pollutant airflow trajectory model, and processing meteorological monitoring data obtained by a monitoring station by using the qualified atmospheric pollutant airflow trajectory model to obtain a meteorological monitoring result. And real-time tracing of the air pollutant track in the factory area is realized. According to the invention, the traceability cost of the gas pollutant track is reduced, the accurate positioning of the pollutant source position is realized, and the problem of difficulty in traceability of the atmospheric pollutants is effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of pollutant monitoring, and in particular to a method for tracing the trajectory of small-scale gaseous pollutants based on airflow motion. Background Art

[0002] The atmospheric environment is a necessary condition for human survival and development. Protecting and improving the quality of the atmospheric environment is of great significance to promoting the development of human society and economy and protecting human health. There are many enterprises in different industries in the industrial park with high concentration, which leads to a wide variety of atmospheric pollutants and high emission intensity. The diffusion of atmospheric pollutants is affected by many factors such as emission source type, geographical environment, and meteorological conditions. Therefore, in the face of sudden complaints and rising monitoring concentrations, the control and management of atmospheric pollutant emissions in industrial parks have fallen into the dilemma of difficult tracing and supervision. Seeking a reliable and feasible method for simulating atmospheric pollutant emissions in industrial parks is extremely important for monitoring and tracing atmospheric pollutant emissions in industrial parks, and has become a hot issue in the current supervision of atmospheric pollution in industrial parks.

[0003] At present, there are two main ways to trace the source of air pollution. One is based on site monitoring. By deploying refined monitoring grids in the region, the changing trend of environmental quality in the region is monitored in real time. When pollution occurs, the pollution source is traced according to the changing trend path of the monitoring data. The other is based on the air quality model. On the basis of simulating regional meteorology, the diffusion of pollutants and the biochemical reactions between various pollutants are obtained by simulating air quality. When the air quality of a certain site is abnormal, the output results and transmission process of the model grid are used to locate the pollution source, so as to trace the pollution source. Both of the above methods require the deployment of a large number of monitoring points, which are costly. At present, there is also a mobile air pollution source tracing method, which uses vehicles equipped with online air monitoring and analysis equipment to move the online air monitoring and analysis equipment to different regional sites to monitor and analyze the atmospheric environment and trace the pollution source. As a temporary pollution monitoring and tracing method, this method also has high requirements for manpower.

[0004] Therefore, it is urgent to propose a gas pollutant trajectory tracing method for industrial parks that saves manpower and equipment costs and improves the tracing accuracy of gas pollutant trajectories. Summary of the invention

[0005] The purpose of the present invention is to address the deficiencies of the above-mentioned prior art and to provide a small-scale gas pollutant trajectory tracing method based on airflow movement. The method constructs a plant airflow trajectory dataset by simulating the airflow movement trajectory under different meteorological conditions. The plant airflow trajectory dataset is used to train and test an atmospheric pollutant airflow trajectory model constructed based on a machine learning method. The atmospheric pollutant airflow trajectory model is used to realize real-time acquisition of the trajectories of gas pollutants in the plant, which greatly saves the equipment cost and manpower cost required for gas pollutant monitoring, realizes accurate acquisition of the source location of gas pollutants, and provides a basis for guiding the formulation of pollutant emission decisions.

[0006] To achieve the above object, the present invention adopts the following technical solution:

[0007] A method for tracing the trajectory of small-scale gas pollutants based on airflow motion comprises the following steps:

[0008] Step 1: Select the plant area to be monitored, obtain the GIS, terrain data and surface building data of the plant area, extract and simplify the geometric features of the ground attachments in the plant area, and construct a geometric model of the plant area;

[0009] Step 2: Mesh the geometric model of the plant area and determine the calculation domain;

[0010] Step 3, using the factory geometry model to simulate the airflow trajectory and construct a factory airflow trajectory dataset;

[0011] Step 4: construct an air pollutant airflow trajectory model based on a machine learning algorithm, and use the airflow trajectory samples in the plant airflow trajectory data set to train and test the air pollutant airflow trajectory model to obtain an air pollutant airflow trajectory model that passes the test;

[0012] Step 5: Set up monitoring stations in the factory area, use the monitoring stations to obtain meteorological monitoring data and input them into the atmospheric pollutant airflow trajectory model that has passed the test, and use the atmospheric pollutant airflow trajectory model to determine the atmospheric pollutant airflow trajectory in the factory area in real time.

[0013] Preferably, the atmospheric pollutants include VOCs, PM2.5, PM10, H2S and SO2.

[0014] Preferably, in step 1, the ground attachments in the factory area are merged to eliminate gaps between adjacent ground attachments, thereby simplifying the geometric features of the ground attachments in the factory area.

[0015] Preferably, in step 2, the horizontal area terrain data and attachment data of the plant area geometric model are obtained to generate the underlying surface of the plant area geometric model, and according to the height of the monitoring equipment, the boundary height of the plant area geometric model is set and the side surface is generated, the calculation domain of the plant area geometric model is determined and meshing is performed.

[0016] Preferably, the step 3 includes the following steps:

[0017] Step 3.1, set the wind speed and wind direction corresponding to the meteorological conditions, use the plant geometry model to simulate the airflow motion trajectory in the calculation domain under different wind speed and wind direction conditions, obtain the airflow motion trajectory corresponding to different meteorological conditions, and generate airflow motion trajectory samples;

[0018] Step 3.2, construct a factory airflow trajectory dataset including a training set and a test set, and randomly assign each airflow motion trajectory sample to the training set and the test set of the factory airflow trajectory dataset.

[0019] Preferably, in step 3, the wind speed is set according to a wind speed calculation interval, and the minimum wind speed and the maximum wind speed in the wind speed calculation interval are obtained according to local meteorological monitoring data over the years;

[0020] The wind directions include east, southeast, south, southwest, west, northwest, north and northeast.

[0021] Preferably, in step 3, the airflow motion trajectory simulation is based on the RNG k-ε model that takes into account the vortex effect in turbulence;

[0022] In the RNG k-ε model that takes into account the vortex effect in turbulence, the transport equations for turbulent kinetic energy k and dissipation rate ε are:

[0023]

[0024] in,

[0025]

[0026] Where t is time; ρ is fluid density; k is turbulent kinetic energy; x i is the distance in the i direction; u i is the velocity vector in the i direction; x l is the distance in the l direction; α k is the reciprocal of the effective Prandtl number of the turbulent kinetic energy k; μ eff is the effective viscosity coefficient, μ eff =μ+μ t , μ is the fluid viscosity coefficient, μ t is the turbulent dynamic viscosity coefficient; x j is the distance in the j direction; G kis the term for turbulent kinetic energy generated by the mean velocity gradient; G b is the term for the turbulent kinetic energy generated by buoyancy; ε is the dissipation rate; Y M is the contribution of pulsating expansion to the total dissipation rate in compressible turbulence; C 1ε , C 2ε , C 3ε are model constants determined by experiments; R and η are intermediate calculation quantities; C μ is a constant; η0 and β are fixed parameters, where η0 is 4.377 and β is 0.012; S is the modulus of the mean strain rate tensor.

[0027] Preferably, the machine learning algorithm adopts random forest, decision tree, support vector machine or K nearest neighbor method.

[0028] Preferably, the step 4 includes the following steps:

[0029] Step 4.1, using the airflow trajectory samples in the training set to train the air pollutant airflow trajectory model, the training air pollutant airflow trajectory model generates the air pollutant airflow trajectory according to the meteorological conditions, and obtains the trained air pollutant airflow trajectory model;

[0030] Step 4.2, inputting the airflow trajectory samples in the test set into the trained air pollutant airflow trajectory model, and using the trained air pollutant airflow trajectory model to generate air pollutant airflow trajectories according to the meteorological conditions in the airflow trajectory samples;

[0031] Step 4.3, compare the airflow trajectory of atmospheric pollutants generated by the trained airflow trajectory model with the airflow trajectory of atmospheric pollutants in the airflow motion trajectory sample, calculate the error value between the airflow trajectory of atmospheric pollutants generated by the trained airflow trajectory model and the airflow trajectory of atmospheric pollutants in the airflow motion trajectory sample, and compare them with the preset accuracy value; if the error value is less than the preset accuracy value, the trained airflow trajectory model is tested to be qualified, and a qualified airflow trajectory model is obtained, and step 5 is entered; otherwise, the trained airflow trajectory model is adjusted, and the training is continued in step 4.1.

[0032] Preferably, the atmospheric pollutant airflow trajectory calculation period is 1 hour, a plurality of source coordinates marked with longitude and latitude are provided within the calculation period, and the atmospheric pollutant airflow trajectory model calculates the atmospheric pollutant airflow trajectory once every 5 minutes.

[0033] Beneficial technical effects brought by the present invention:

[0034] (1) The present invention aims to solve the problem that industrial parks have a large number of industries and a high concentration of atmospheric pollutants, resulting in a wide variety of atmospheric pollutants, high emission intensity, and diffusion that is easily affected by multiple factors such as emission source type, geographical environment, and meteorological conditions. A small-scale gas pollutant trajectory tracing method based on airflow movement is proposed, which achieves accurate acquisition of the airflow trajectory of atmospheric pollutants. By accurately determining the source location of atmospheric pollutants, the problem of difficulty in tracing the source of pollutants in the process of atmospheric pollutant emission control and management is effectively solved.

[0035] (2) The present invention proposes a small-scale gas pollutant trajectory tracing method based on airflow movement. This method can accurately obtain the airflow trajectory of atmospheric pollutants by using only the meteorological monitoring data obtained at the monitoring point, and trace the source trajectory of atmospheric pollutants based on the airflow trajectory of atmospheric pollutants. This overcomes the problems of the high cost of setting up refined monitoring grids when tracing the source of atmospheric pollutants based on monitoring stations in the prior art and the large amount of human resources required for mobile monitoring, thereby greatly saving the equipment cost and labor cost required for gas pollutant monitoring.

[0036] (3) The present invention proposes a small-scale gas pollutant trajectory tracing method based on airflow movement, which fully considers the impact of buildings in the factory area on the airflow trajectory of atmospheric pollutants. It not only improves the accuracy of monitoring the airflow trajectory of atmospheric pollutants in the factory area, but also realizes the rapid tracing of the source location of atmospheric pollutants. It is conducive to guiding the establishment of an air pollution prevention and control system and has a positive role in promoting atmospheric ecological environmental protection and sustainable development. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 The present invention is a flow chart of a method for tracing the trajectory of small-scale gas pollutants based on airflow motion. DETAILED DESCRIPTION

[0038] It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0039] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0040] In the present invention, the directions or positional relationships indicated by terms such as "upper", "lower", "bottom", "top", etc. are based on the directions or positional relationships shown in the accompanying drawings. They are relational words determined only for the convenience of describing the structural relationships of the various parts or elements of the present invention. They do not specifically refer to any part or element in the present invention and cannot be understood as limitations on the present invention.

[0041] In the present invention, terms such as "connected" and "connection" should be understood in a broad sense, indicating that the connection can be fixed, integral or detachable; it can be directly connected or indirectly connected through an intermediate medium. Relevant scientific research or technical personnel in this field can determine the specific meaning of the above terms in the present invention according to specific circumstances, and they should not be understood as limiting the present invention.

[0042] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0043] This embodiment takes a certain factory as an example to explain in detail a small-scale gas pollutant trajectory tracing method based on airflow motion disclosed in the present invention. The small-scale gas pollutant trajectory tracing method based on airflow motion proposed in the present invention is used to trace the air pollutants (including VOCs, PM2.5, PM10, H2S and SO2) in the factory. Figure 1 As shown, the specific steps include:

[0044] Step 1: Select the plant area to be monitored, obtain the GIS, terrain data and surface building data of the plant area, extract the geometric characteristics of the ground attachments in the plant area, and build a geometric model of the plant area by merging the surface attachments in the plant area, eliminating the gaps between adjacent ground attachments, and simplifying the geometric characteristics of the ground attachments in the plant area without affecting the overall flow state of atmospheric pollutants in the plant area.

[0045] Step 2: Mesh the plant geometry model and determine the computational domain.

[0046] In this embodiment, the underlying surface of the plant area geometric model is generated by acquiring the horizontal area terrain data and attachment data of the plant area geometric model, and combined with the installation height of the monitoring equipment being 1.5 to 2 meters, the boundary height of the plant area geometric model in the vertical direction is set to 10 meters, and the side surfaces are generated on the boundary of the plant area geometric model, so that the area enclosed by the underlying surface and the side surfaces is used as the calculation domain of the plant area geometric model, and the calculation domain of the plant area geometric model is evenly divided into multiple hexahedral grids.

[0047] Step 3: Use the factory geometry model to simulate the airflow trajectory and construct a factory airflow trajectory dataset, which specifically includes the following sub-steps:

[0048] Step 3.1, set the wind speed and wind direction corresponding to the meteorological conditions, use the plant geometry model to simulate the airflow motion trajectory in the calculation domain under different wind speed and wind direction conditions, obtain the airflow motion trajectory corresponding to different meteorological conditions, and generate airflow motion trajectory samples.

[0049] In this embodiment, the wind speed is set according to the wind speed calculation interval, and the minimum wind speed and the maximum wind speed of the wind speed calculation interval are obtained according to the local meteorological monitoring data over the years; the wind direction includes east, southeast, south, southwest, west, northwest, north and northeast.

[0050] The airflow trajectory simulation is based on the RNG k-ε model that takes into account the vortex effect in turbulence, where the transport equations for turbulent kinetic energy k and dissipation rate ε are:

[0051]

[0052] in,

[0053]

[0054] Where t is time; ρ is fluid density; k is turbulent kinetic energy; x i is the distance in the i direction; u i is the velocity vector in the i direction; x l is the distance in the l direction; α k is the reciprocal of the effective Prandtl number of the turbulent kinetic energy k; μ eff is the effective viscosity coefficient, μ eff =μ+μ t , μ is the fluid viscosity coefficient, μ t is the turbulent dynamic viscosity coefficient; x j is the distance in the j direction; G k is the term for turbulent kinetic energy generated by the mean velocity gradient; G b is the term for the turbulent kinetic energy generated by buoyancy; ε is the dissipation rate; Y M is the contribution of pulsating expansion to the total dissipation rate in compressible turbulence; C 1ε , C 2ε , C 3ε are model constants determined by experiments; R and η are intermediate calculation quantities; C μ is a constant; η0 and β are fixed parameters, where η0 is 4.377 and β is 0.012; S is the modulus of the mean strain rate tensor.

[0055] Step 3.2, construct a plant airflow trajectory dataset including a training set and a test set, and randomly assign each airflow trajectory sample to the training set and the test set of the plant airflow trajectory dataset. In this embodiment, the number of airflow trajectory samples in the training set is not less than the number of airflow trajectory samples in the test set.

[0056] Step 4: construct an air pollutant airflow trajectory model based on a machine learning algorithm, and use the airflow motion trajectory samples in the factory airflow trajectory dataset to train and test the air pollutant airflow trajectory model to obtain an air pollutant airflow trajectory model that passes the test.

[0057] Among them, the machine learning algorithms include random forest, decision tree, support vector machine and K nearest neighbor method.

[0058] In this embodiment, the machine learning algorithm adopts a random forest algorithm, and the model structure of the atmospheric pollutant airflow trajectory model is constructed based on a convolutional neural network (CNN). The convolutional neural network is common knowledge to those skilled in the art, and its internal network structure can be obtained from the prior art and will not be repeated in this embodiment.

[0059] The step 4 specifically includes the following sub-steps:

[0060] Step 4.1, using the airflow trajectory samples in the training set to train the air pollutant airflow trajectory model, the training air pollutant airflow trajectory model generates the air pollutant airflow trajectory according to the meteorological conditions, and obtains the trained air pollutant airflow trajectory model.

[0061] Step 4.2, input the airflow trajectory samples in the test set into the trained air pollutant airflow trajectory model, and use the trained air pollutant airflow trajectory model to generate the air pollutant airflow trajectory according to the meteorological conditions in the airflow trajectory samples.

[0062] Step 4.3, compare the airflow trajectory of atmospheric pollutants generated by the trained airflow trajectory model with the airflow trajectory of atmospheric pollutants in the airflow motion trajectory sample, calculate the error value between the airflow trajectory of atmospheric pollutants generated by the trained airflow trajectory model and the airflow trajectory of atmospheric pollutants in the airflow motion trajectory sample, and compare them with the preset accuracy value; if the error value is less than the preset accuracy value, the trained airflow trajectory model is tested to be qualified, and a qualified airflow trajectory model is obtained, and step 5 is entered; otherwise, the trained airflow trajectory model is adjusted, and the training is continued in step 4.1.

[0063] Step 5: Set up monitoring stations in the factory area, and set up monitoring equipment at the monitoring stations. The installation height of the monitoring equipment is 1.5 to 2 meters. Use the monitoring stations to obtain meteorological monitoring data and input them into the atmospheric pollutant airflow trajectory model that has passed the test. Use the atmospheric pollutant airflow trajectory model to determine the atmospheric pollutant airflow trajectory in the factory area in real time, so as to accurately locate the source of atmospheric pollutants in the factory area.

[0064] In this embodiment, the air pollutant airflow trajectory calculation cycle is 1 hour, and 12 source coordinates marked with longitude and latitude are set in one calculation cycle. The air pollutant airflow trajectory model calculates the air pollutant airflow trajectory every 5 minutes. At the same time, when there is a new polluted air mass flowing through the monitoring device, the monitoring device is used to obtain the meteorological data of the current time and the previous hour, and the longitude and latitude data and gas phase data of the monitoring device are used to call the air pollutant airflow trajectory model for reverse tracing, with a calculation step of 5 minutes to obtain the source trajectory of the air pollutants in the previous hour.

[0065] In summary, the method of the present invention can quickly determine the airflow trajectory of atmospheric pollutants in real time, accurately determine the source location of atmospheric pollutants, effectively solve the problem of difficulty in tracing the source of atmospheric pollutants within the factory area, greatly reduce the manpower and material resources required for atmospheric pollution monitoring, and is conducive to protecting the sustainable development of the atmospheric ecological environment.

[0066] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not a limitation of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the protection scope of the present invention.

Claims

1. A small-scale gas pollutant trajectory tracing method based on airflow motion, characterized in that: The following steps are involved: Step 1: Select the plant area to be monitored, obtain the GIS, terrain data and surface building data of the plant area, extract and simplify the geometric features of the ground attachments in the plant area, and construct a geometric model of the plant area; Step 2: Mesh the geometric model of the plant area and determine the calculation domain; Step 3, using the factory geometry model to simulate the airflow trajectory and construct a factory airflow trajectory dataset; Step 4: construct an air pollutant airflow trajectory model based on a machine learning algorithm, and use the airflow trajectory samples in the plant airflow trajectory data set to train and test the air pollutant airflow trajectory model to obtain an air pollutant airflow trajectory model that passes the test; Step 5: Set up monitoring stations in the factory area, use the monitoring stations to obtain meteorological monitoring data and input them into the atmospheric pollutant airflow trajectory model that has passed the test, and use the atmospheric pollutant airflow trajectory model to determine the atmospheric pollutant airflow trajectory in the factory area in real time.

2. The method for tracing the trajectory of small-scale gas pollutants based on airflow motion according to claim 1 is characterized in that: The air pollutants include VOCs, PM2.5, PM10, H2S and SO2.

3. The method for tracing the trajectory of small-scale gas pollutants based on airflow motion according to claim 1 is characterized in that: In the step 1, the ground attachments in the factory area are merged to eliminate the gaps between adjacent ground attachments and simplify the geometric features of the ground attachments in the factory area.

4. The method for tracing the trajectory of small-scale gas pollutants based on airflow motion according to claim 1 is characterized in that: In step 2, the horizontal area terrain data and attachment data of the plant area geometric model are obtained to generate the underlying surface of the plant area geometric model. According to the height of the monitoring equipment, the boundary height of the plant area geometric model is set and the side surface is generated. The calculation domain of the plant area geometric model is determined and meshing is performed.

5. The method for tracing the trajectory of small-scale gas pollutants based on airflow motion according to claim 1 is characterized in that: The step 3 includes the following steps: Step 3.1, set the wind speed and wind direction corresponding to the meteorological conditions, use the plant geometry model to simulate the airflow motion trajectory in the calculation domain under different wind speed and wind direction conditions, obtain the airflow motion trajectory corresponding to different meteorological conditions, and generate airflow motion trajectory samples; Step 3.2, construct a factory airflow trajectory dataset including a training set and a test set, and randomly assign each airflow motion trajectory sample to the training set and the test set of the factory airflow trajectory dataset.

6. The method for tracing the trajectory of small-scale gas pollutants based on airflow motion according to claim 5 is characterized in that: In step 3, the wind speed is set according to the wind speed calculation interval, and the minimum wind speed and the maximum wind speed in the wind speed calculation interval are obtained according to the local meteorological monitoring data over the years; The wind directions include east, southeast, south, southwest, west, northwest, north and northeast.

7. The method for tracing the trajectory of small-scale gas pollutants based on airflow motion according to claim 5 is characterized in that: In step 3, the airflow trajectory simulation is based on the RNG k-ε model that takes into account the vortex effect in turbulence; In the RNG k-ε model that takes into account the vortex effect in turbulence, the transport equations for turbulent kinetic energy k and dissipation rate ε are: in, Where t is time; ρ is fluid density; k is turbulent kinetic energy; x i is the distance in the i direction; u i is the velocity vector in the i direction; x l is the distance in the l direction; α k is the reciprocal of the effective Prandtl number of the turbulent kinetic energy k; μ eff is the effective viscosity coefficient, μ eff =μ+μ t , μ is the fluid viscosity coefficient, μ t is the turbulent dynamic viscosity coefficient; x j is the distance in the j direction; G k is the term for turbulent kinetic energy generated by the mean velocity gradient; G b is the term for the turbulent kinetic energy generated by buoyancy; ε is the dissipation rate; Y M is the contribution of pulsating expansion to the total dissipation rate in compressible turbulence; C 1ε , C 2ε , C 3ε are model constants determined by experiments; R and η are intermediate calculation quantities; C μ is a constant; η0 and β are fixed parameters, where η0 is 4.377 and β is 0.012; S is the modulus of the mean strain rate tensor.

8. The method for tracing the trajectory of small-scale gas pollutants based on airflow motion according to claim 1 is characterized in that: The machine learning algorithm uses random forest, decision tree, support vector machine or K nearest neighbor method.

9. The method for tracing the trajectory of small-scale gas pollutants based on airflow motion according to claim 8 is characterized in that: The step 4 includes the following steps: Step 4.1, using the airflow trajectory samples in the training set to train the air pollutant airflow trajectory model, the training air pollutant airflow trajectory model generates the air pollutant airflow trajectory according to the meteorological conditions, and obtains the trained air pollutant airflow trajectory model; Step 4.2, inputting the airflow trajectory samples in the test set into the trained air pollutant airflow trajectory model, and using the trained air pollutant airflow trajectory model to generate air pollutant airflow trajectories according to the meteorological conditions in the airflow trajectory samples; Step 4.3, compare the airflow trajectory of atmospheric pollutants generated by the trained airflow trajectory model with the airflow trajectory of atmospheric pollutants in the airflow motion trajectory sample, calculate the error value between the airflow trajectory of atmospheric pollutants generated by the trained airflow trajectory model and the airflow trajectory of atmospheric pollutants in the airflow motion trajectory sample, and compare it with the preset accuracy value; if the error value is less than the preset accuracy value, the trained airflow trajectory model is tested to be qualified, and a qualified airflow trajectory model is obtained, and step 5 is entered; otherwise, the trained airflow trajectory model is adjusted, and the training is continued in step 4.

1.

10. The method for tracing the trajectory of small-scale gas pollutants based on airflow motion according to claim 1 is characterized in that: The calculation period of the air pollutant airflow trajectory is 1 hour. A plurality of source coordinates marked with longitude and latitude are set within the calculation period. The air pollutant airflow trajectory model calculates the air pollutant airflow trajectory once every 5 minutes.

Images