Single-variable pollutant emission source inversion method, device, equipment and medium

Abstract

The invention relates to a univariate pollutant emission source inversion method and device, equipment and a medium. According to the method, the advantages of variational assimilation and deep learning are combined, after three-dimensional variational assimilation is performed on pollutant concentration observation data to obtain a more accurate dye concentration reanalysis field, a trained emission inversion model with a prediction function is utilized to perform pollutant emission source inversion prediction, and a more accurate pollutant emission list is obtained. By adopting the method, the requirements of atmospheric pollution forecast and control on a high-time-resolution emission list can be met, and the calculation efficiency is also remarkably improved. In addition, the pollutant emission list obtained based on the method can also be used for improving the pollutant concentration forecast level of an atmospheric chemical mode, and powerful support is provided for atmospheric pollution prevention, control, forecast and early warning.

Description

Technical Field

[0001] The present application relates to the field of atmospheric environment technology, and in particular to a method, device, equipment and medium for inversion of single variable pollutant emission sources. Background Art

[0002] An accurate air pollution emission inventory is a prerequisite for air pollution control. Typical emission inventories are usually compiled using a "bottom-up" approach, but the compilation process relies on a large amount of emission statistics, and then it is necessary to assume that hourly factors are used to differentiate emissions over time, which further increases the error, making it difficult to update the emission inventory and has considerable uncertainty.

[0003] The main "top-down" methods currently used to constrain "bottom-up" emission estimates are ensemble Kalman filtering and four-dimensional variational assimilation. However, these methods are complex to model and require a lot of computing resources, and their practical applications are limited. In recent years, deep learning methods have been applied to air pollution prediction and have achieved good results. However, average emissions are usually obtained on a monthly or daily time scale, which has limited applicability when estimating emissions during short-term pollution events. Therefore, it is urgent to propose an inversion method that can provide an emission inventory with higher temporal resolution. Summary of the invention

[0004] Based on this, it is necessary to provide a single variable pollutant emission source inversion method, device, equipment and medium to address the above technical problems, the purpose of which is to achieve high time resolution emission source inversion on the basis of improving calculation efficiency and accuracy.

[0005] A single variable pollutant emission source inversion method, the method comprising: Obtain single variable pollutant concentration observation data with short life cycle characteristics in the study area, meteorological reanalysis grid data and prior emission source data; The enhanced disturbance strategy, the weakened disturbance strategy and the cross-level adjustment disturbance strategy are respectively used to disturb the prior emission source data to obtain the disturbed emission source data; Using atmospheric chemistry models and disturbed emission source data, the meteorological field and pollutant concentration field in the study area are simulated hourly to obtain meteorological element data sets and pollutant concentration data sets with hourly resolution. The meteorological element dataset, pollutant concentration dataset and prior emission source data are input into a pre-built emission inversion model based on deep learning for training to obtain a trained emission inversion model; The three-dimensional variational assimilation method is used to assimilate the hourly pollutant concentration observation data in the study area to generate the pollutant concentration reanalysis field; The meteorological element dataset, pollutant concentration reanalysis field and prior emission source data are input into the trained emission inversion model for prediction to obtain the emission increment field. By superimposing the emission increment field on the prior emission field, the pollutant emission inventory with hourly resolution in the study area is obtained.

[0006] In one embodiment, after obtaining single variable pollutant concentration observation data with short life cycle characteristics in the study area, meteorological reanalysis grid data and prior emission source data, it also includes: The pollutant concentration observation data are quality controlled and averaged according to the study area and grid resolution of the WRF-Chem model.

[0007] In one embodiment, the enhanced disturbance strategy, the weakened disturbance strategy, and the cross-level adjustment disturbance strategy are respectively used to disturb the prior emission source data to obtain the disturbed emission source data, including: According to the actual increase, decrease and normal emission ratio of pollution sources, the grid ratio of increased emission, reduced emission and normal emission in the study area is set every hour; Create a set of random numbers that conform to the normal distribution and have the same dimension as the prior emission source data, perform Gaussian filtering on the random numbers, and perform initial perturbations by multiplying the emissions in each type of grid by random numbers that obey different means and standard deviations and conform to the normal distribution, to obtain the perturbed emissions in the enhanced emission grid, weakened emission grid, and normal emission grid; The enhanced disturbance strategy is adopted, that is, based on the initial disturbance, the disturbed emissions in all grids are multiplied by a random number greater than 1 generated from a uniform distribution to obtain an enhanced emission inventory for simulating emission source enhancement; The strategy of reducing disturbance is adopted, that is, based on the initial disturbance, the disturbed emissions in all grids are multiplied by a random number less than 1 generated from a uniform distribution to obtain the reduced emission inventory used to simulate the reduction of emission sources; A cross-level adjustment disturbance strategy is adopted, that is, before the initial disturbance, the emission at each moment in the prior emission source data is adjusted by adding a random number in a set interval. At the same time, the emissions are divided into 8 intervals according to the preset threshold, and the emissions in each interval are perturbed across levels. Then, the emissions in each grid are multiplied by a random number that obeys different means and standard deviations and conforms to the normal distribution, so as to obtain an emission inventory in which the emissions are controlled within the preset range and is used to simulate the migration of existing emission sources and the formation of new emission sources.

[0008] In one embodiment, the initial perturbation is performed by multiplying the emissions in each type of grid by random numbers that have different means and standard deviations and conform to normal distribution, including: The emissions within the enhanced emission grid are initially perturbed by multiplying a random number with a mean greater than 1, the emissions within the weakened emission grid are initially perturbed by multiplying a random number with a mean less than 1, and the emissions within the normal emission grid are initially perturbed by multiplying a random number with a mean equal to 1.

[0009] In one embodiment, the emission in each interval is subjected to cross-level adjustment disturbance, including: Two types of grid proportions are set in the study area. For one type of grid, a random number is directly added to the emission in the grid to make a disturbance, so that the emission value after disturbance changes in magnitude; for the other type of grid, the emission in the grid is multiplied by a random number to make a disturbance; When disturbing emissions in different intervals, the situations at different times are taken into account. For moments when a priori emissions are overestimated, emissions are reduced. For moments when a priori emissions are underestimated, new emission sources are added.

[0010] In one embodiment, the meteorological field and pollutant concentration field in the study area are forward simulated hourly using the atmospheric chemistry model and the disturbed emission source data to obtain a meteorological element dataset and a pollutant concentration dataset with hourly resolution, including: Based on the prior emission source data, hourly forward simulations were performed during the study period, with the simulation time window set to 1 hour, and the hourly simulation results were used as the background field; Taking the background field at the initial moment of the study as the initial field, the atmospheric chemistry model and the emission inventory obtained by three perturbation strategies are used to perform hourly forward simulations on the meteorological field and pollutant concentration field in the study area to obtain meteorological element datasets and pollutant concentration datasets with hourly resolution.

[0011] In one embodiment, a meteorological element dataset, a pollutant concentration dataset, and a priori emission source data are input into a pre-built emission inversion model based on deep learning for training to obtain a trained emission inversion model, including: The hourly-resolution meteorological element dataset, pollutant concentration dataset and prior emission source data obtained by forward simulation are used as input features and input into the deep learning-based emission inversion model for training. The training target is defined as the difference in hourly emissions between the disturbed emission source data and the prior emission source data, that is, the emission increment. The model inversion accuracy is evaluated by comparing the difference between the emissions predicted by the model and the emission source data after the disturbance, until a trained emission inversion model that meets the preset inversion accuracy is obtained.

[0012] A single variable pollutant emission source inversion device, the device comprising: The data acquisition module is used to obtain single-variable pollutant concentration observation data with short life cycle characteristics in the study area, meteorological reanalysis grid data and prior emission source data; A disturbance module is used to disturb the prior emission source data by using an enhanced disturbance strategy, a reduced disturbance strategy and a cross-level adjustment disturbance strategy to obtain disturbed emission source data; The forward simulation module is used to use the atmospheric chemical model and the disturbed emission source data to perform hourly forward simulation of the meteorological field and pollutant concentration field in the study area to obtain hourly resolution meteorological element data sets and pollutant concentration data sets; The model training module is used to input the meteorological element dataset, the pollutant concentration dataset and the prior emission source data into the pre-built emission inversion model based on deep learning for training, so as to obtain a trained emission inversion model; The variational assimilation module is used to assimilate the hourly pollutant concentration observation data in the study area using a three-dimensional variational assimilation method to generate a pollutant concentration reanalysis field; The emission inversion module is used to input the meteorological element dataset, pollutant concentration reanalysis field and prior emission source data into the trained emission inversion model for prediction, obtain the emission increment field, and obtain the pollutant emission inventory with hourly resolution in the study area by superimposing the emission increment field onto the prior emission field.

[0013] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented: Obtain single variable pollutant concentration observation data with short life cycle characteristics in the study area, meteorological reanalysis grid data and prior emission source data; The enhanced disturbance strategy, the weakened disturbance strategy and the cross-level adjustment disturbance strategy are respectively used to disturb the prior emission source data to obtain the disturbed emission source data; Using atmospheric chemistry models and disturbed emission source data, the meteorological field and pollutant concentration field in the study area are simulated hourly to obtain meteorological element data sets and pollutant concentration data sets with hourly resolution. The meteorological element dataset, pollutant concentration dataset and prior emission source data are input into a pre-built emission inversion model based on deep learning for training to obtain a trained emission inversion model; The three-dimensional variational assimilation method is used to assimilate the hourly pollutant concentration observation data in the study area to generate the pollutant concentration reanalysis field; The meteorological element dataset, pollutant concentration reanalysis field and prior emission source data are input into the trained emission inversion model for prediction to obtain the emission increment field. By superimposing the emission increment field on the prior emission field, the pollutant emission inventory with hourly resolution in the study area is obtained.

[0014] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps: Obtain single variable pollutant concentration observation data with short life cycle characteristics in the study area, meteorological reanalysis grid data and prior emission source data; The enhanced disturbance strategy, the weakened disturbance strategy and the cross-level adjustment disturbance strategy are respectively used to disturb the prior emission source data to obtain the disturbed emission source data; Using atmospheric chemistry models and disturbed emission source data, the meteorological field and pollutant concentration field in the study area are simulated hourly to obtain meteorological element data sets and pollutant concentration data sets with hourly resolution. The meteorological element dataset, pollutant concentration dataset and prior emission source data are input into a pre-built emission inversion model based on deep learning for training to obtain a trained emission inversion model; The three-dimensional variational assimilation method is used to assimilate the hourly pollutant concentration observation data in the study area to generate the pollutant concentration reanalysis field; The meteorological element dataset, pollutant concentration reanalysis field and prior emission source data are input into the trained emission inversion model for prediction to obtain the emission increment field. By superimposing the emission increment field on the prior emission field, the pollutant emission inventory with hourly resolution in the study area is obtained.

[0015] The above-mentioned single variable pollutant emission source inversion method, device, equipment and medium combine the advantages of variational assimilation and deep learning. After performing three-dimensional variational assimilation on the pollutant concentration observation data to obtain a more accurate pollutant concentration reanalysis field, the trained emission inversion model with prediction function is used to perform pollutant emission source inversion prediction to obtain a more accurate pollutant emission inventory. And compared with the prior art, the emission inversion model trained in this application can predict pollutant emissions with hourly resolution, which greatly reduces the consumption of computing resources and shortens the calculation time, so that this application has the ability to solve the emission estimation problem in short-term pollution events, which can not only meet the needs of atmospheric pollution forecasting and control for high time resolution emission inventories, but also significantly improve the calculation efficiency. In addition, the pollutant emission inventory obtained based on this method can also be used to improve the pollutant concentration forecast level of atmospheric chemistry models, and provide strong support for atmospheric pollution prevention and control and forecast and warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of a flow chart of a single variable pollutant emission source inversion method in an embodiment; Figure 2 is the SO predicted by the emission inversion model in one embodiment 2 Schematic representation of the correlation between emissions and “real” emissions; Figure 3 In one embodiment, SO2 is calculated based on the MEIC_2016 emission inventory and the inversion of the present application at 17:00 UTC (universal standard time) on January 21, 2023. 2 Scatter diagram of the concentration interpolated to the ground station and the observed concentration at the station; Figure 4 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0018] In one embodiment, Figure 1 As shown, a single variable pollutant emission source inversion method is provided, comprising the following steps: Step 1: Obtain single variable pollutant concentration observation data, meteorological reanalysis grid data and prior emission source data with short life cycle characteristics in the study area.

[0019] Specifically, after obtaining the single-variable pollutant concentration observation data with short life cycle characteristics in the study area, the meteorological reanalysis grid data and the prior emission source data, it is also necessary to perform quality control on the pollutant concentration observation data and, based on the study area and grid resolution of the WRF-Chem model, average the pollutant concentration observation data that fall within the same grid.

[0020] Through quality control, the accuracy and reliability of observation data can be effectively observed, and the actual situation of pollutant concentrations in the study area can be more realistically reflected, laying a reliable data foundation for subsequent data analysis and prediction. Among them, the WRF-Chem model was developed by the NOAA (National Oceanic and Atmospheric Administration) Forecast System Laboratory (FSL). It is a regional air quality model that couples the meteorological model (WRF) and the chemical model (Chem) online. Averaging the observed data through the WRF-Chem model helps to simplify the data structure and reduce the complexity and discreteness of the data.

[0021] Step 2: Use the enhanced disturbance strategy, weakened disturbance strategy and cross-level adjustment disturbance strategy to perturb the prior emission source data to obtain the disturbed emission source data.

[0022] Among them, the enhanced disturbance strategy, weakened disturbance strategy and cross-level adjustment disturbance strategy are used for the enhancement, weakening, addition and migration of emission sources, which help to provide a more comprehensive and realistic emission source simulation. Specifically, three different disturbances are performed on the prior emission source data, including the following steps: First, according to the actual increase, reduction and normal emission ratio of pollution sources, the grid ratio of enhanced emission, weakened emission and normal emission in the study area is set every hour.

[0023] Then, a set of random numbers that conform to the normal distribution and have the same dimension as the prior emission source data is created, and the random numbers are Gaussian filtered. The emissions in each type of grid are initially perturbed by random numbers that obey different means and standard deviations and conform to the normal distribution, and the perturbed emissions in the enhanced emission grid, weakened emission grid, and normal emission grid are obtained. Among them, the emissions in the enhanced emission grid correspond to multiplying the random numbers with a mean greater than 1 for initial perturbation, the emissions in the weakened emission grid correspond to multiplying the random numbers with a mean less than 1 for initial perturbation, and the emissions in the normal emission grid correspond to multiplying the random numbers with a mean equal to 1 for initial perturbation.

[0024] An enhanced disturbance strategy is adopted, that is, on the basis of the initial disturbance, the disturbed emissions in all grids are multiplied by a random number greater than 1 generated from a uniform distribution to obtain an enhanced emission inventory for simulating emission source enhancement.

[0025] A disturbance reduction strategy is adopted, that is, based on the initial disturbance, the disturbed emissions in all grids are multiplied by a random number less than 1 generated from a uniform distribution to obtain a reduced emission inventory for simulating the reduction of emission sources.

[0026] A cross-level adjustment disturbance strategy is adopted, that is, before the initial disturbance, the emission at each moment in the prior emission source data is adjusted by adding a random number in a set interval. At the same time, the emissions are divided into 8 intervals according to the preset threshold, and the emissions in each interval are perturbed across levels. Then, the emissions in each grid are multiplied by a random number that obeys different means and standard deviations and conforms to the normal distribution, so as to obtain an emission inventory in which the emissions are controlled within the preset range and is used to simulate the migration of existing emission sources and the formation of new emission sources.

[0027] Among them, cross-level adjustment disturbances include: Two types of grid proportions are set in the study area. For one type of grid, a random number is directly added to the emission in the grid to make a disturbance, so that the emission value after disturbance changes in magnitude; for the other type of grid, the emission in the grid is multiplied by a random number to make a disturbance; When disturbing emissions in different intervals, the situations at different times are taken into consideration. For moments when the prior emissions are overestimated, emissions are reduced. For moments when the prior emissions are underestimated, new emission sources are added. This can reduce high-value emissions and increase a certain proportion of low-value emissions, simulating the migration of existing emission sources and the formation of new emission sources.

[0028] Step 3: Use the atmospheric chemistry model and the disturbed emission source data to perform hourly forward simulation of the meteorological field and pollutant concentration field in the study area to obtain hourly resolution meteorological element datasets and pollutant concentration datasets.

[0029] Specifically, based on the prior emission source data, hourly forward simulations were performed during the study period, with the simulation time window set to 1 hour, and the hourly simulation results were used as the background field. Then, with the background field at the initial moment of the study as the initial field, the atmospheric chemistry model and the emission inventory obtained using three perturbation strategies were used to perform hourly forward simulations on the meteorological field and pollutant concentration field in the study area, obtaining hourly resolution meteorological element datasets and pollutant concentration datasets.

[0030] Step 4: Input the meteorological element dataset, pollutant concentration dataset and prior emission source data into a pre-built deep learning-based emission inversion model for training to obtain a trained emission inversion model.

[0031] Specifically, the hourly-resolution meteorological element data set, pollutant concentration data set and prior emission source data obtained by forward simulation are used as input features and input into the emission inversion model based on deep learning for training. The training target is defined as the difference between the perturbed emission source data and the prior emission source data per hour, i.e., the emission increment. The model inversion accuracy is evaluated by comparing the difference between the model-predicted emissions and the perturbed emission source data until a trained emission inversion model that meets the preset inversion accuracy is obtained. In this embodiment, the U-Net model is specifically selected as the emission inversion model.

[0032] Step 5: Use the three-dimensional variational assimilation method to assimilate the hourly pollutant concentration observation data in the study area to generate a pollutant concentration reanalysis field.

[0033] Step 6: Input the meteorological element dataset, pollutant concentration reanalysis field, and prior emission source data into the trained emission inversion model for prediction to obtain the emission increment field. By superimposing the emission increment field on the prior emission field, the pollutant emission inventory with hourly resolution in the study area is obtained.

[0034] Three-dimensional variational assimilation (3D-VAR) is a data assimilation method that combines observational data from different sources (such as data from meteorological observation stations, satellite remote sensing data, etc.) with the background field of numerical models (such as atmospheric circulation models, etc.) to obtain a more accurate estimate of the state of geophysical systems such as the atmosphere and ocean. By performing three-dimensional variational assimilation on pollutant concentration observation data, a more accurate pollutant concentration reanalysis field can be obtained, and then the pollutant concentration reanalysis field is input into the trained emission inversion model for pollutant emission source inversion prediction, making the model prediction more accurate.

[0035] In a specific embodiment, to optimize the sulfur dioxide (SO2) produced by fireworks during the Spring Festival from January 20 to January 23, 2023, 2 ) emissions as an example, a single variable pollutant emission source inversion method provided in this application is used to obtain SO 2 Emission analysis site. The research area is China, and the specific steps include: 1. Collect SO from ground monitoring stations from January 11 to February 9, 2022 and from January 20 to January 23, 2023 2 The concentration data were collected and quality controlled. At the same time, the FNL reanalysis data during the study period were obtained. The FNL reanalysis data are global reanalysis data provided by the National Centers for Environmental Prediction (NECP) / National Center for Atmospheric Research (NCAR). The 2016 Multi-resolution Emission Inventory of China (MEIC_2016) compiled by Tsinghua University was used as the prior emission source data. According to the research area and grid resolution of the WRF-Chem model, the SO 2 The concentration data were averaged.

[0036] 2. Regarding SO in MEIC_2016 2 Considering that only using the MEIC inventory in January 2016 for forward simulation will lead to insufficient sample diversity, thus limiting the ability of the deep learning model to learn various features under different conditions, to address this limitation and enrich the training dataset, three perturbation strategies are used to adjust the SO in MEIC_2016. 2 For emissions, each method generated a 30-day emissions inventory from January 11 to February 9, 2022. The specific steps are as follows: 2.1. Initial disturbance: Firstly, according to the ratio of actual emission increase, emission reduction and normal emission, the grid ratio of enhanced emission, weakened emission and normal emission in the study area is set every hour.

[0037] Then, a set of random numbers with normal distribution and the same dimension as the prior emission source data is created. To ensure the correlation of disturbances in adjacent grids, the random numbers are Gaussian filtered and the SO 2 The emissions are multiplied by random numbers with different means and standard deviations and in accordance with normal distribution for initial perturbation, and the perturbed SO in the enhanced emission grid, weakened emission grid and normal emission grid are obtained. 2 Emissions.

[0038] The set grid ratio and random number information are shown in Table 1.

[0039] Table 1 Grid ratio and random number information

[0040] Specifically, the random number Select from three sets of numbers with the same dimension as the emission that conform to the Gaussian distribution and , after the initial perturbation, the grid points The disturbance SO 2 Emissions Expressed as ; in, and represents the horizontal and vertical coordinates of the grid points, Represents SO from MEIC_2016 list 2 Emission, for different classification grid points, Taken from different : ; in, N represents a normal distribution, is the mean, standard deviation It is taken from The random number in the interval is chosen to introduce more variability while ensuring the actual disturbance level. Gaussian filtering is performed to ensure the spatial consistency of the disturbance between adjacent grid points.

[0041] 2.2 Enhanced perturbation strategy: Based on the initial perturbation, the perturbation SO in all grids is enhanced. 2 Emissions Multiply it by a value drawn from a uniform distribution in the interval Random number , and obtain the enhanced emission inventory used to simulate emission source enhancement .

[0042] 2.3. Disturbance reduction strategy: Based on the initial disturbance, all disturbances in the grid are reduced. 2 Emissions Multiply it by a value drawn from a uniform distribution in the interval Random number , and obtain the abatement emission inventory used to simulate emission source abatement .

[0043] 2.4. Cross-level adjustment disturbance strategy: Before the initial disturbance, the emission level of each moment in the prior emission source data is adjusted by adding a random number in a set interval, and the SO 2 The emissions are divided into 8 intervals, expressed as ; in, Indicates SO 2 Emissions; For SO in 8 intervals 2 After the emissions are adjusted across levels, the SO 2 The emission is multiplied by a random number with different means and standard deviations and in accordance with the normal distribution, and the emission is controlled within 0-600 mol / km 2 / h and is used to simulate the migration of existing emission sources and the formation of new emission sources. Specifically, the calculation formula for the cross-level adjustment disturbance strategy is as follows: ; in, is taken from the interval The random number is used to adjust the emission between different intervals so that Compared with the emissions after the initial disturbance, the magnitude of the increase or decrease is achieved. In order to ensure that the emissions have actual physical meaning (emissions cannot be negative), Constrained to be a number greater than or equal to 0. This adjustment is intended to simulate the migration of existing emission sources and the emergence of new emission sources.

[0044] 3. Based on the MEIC_2016 emission inventory, hourly forward simulations were performed during the study period. The simulation time window was set to 1 hour, and the hourly simulation results were used as the background field. Subsequently, the background field at the initial moment of the study was used as the initial field, and three versions of the emission inventory obtained based on the three perturbation strategies were used to perform hourly forward simulations of the meteorological field and pollutant concentration field from January 11 to February 9, 2022. Through the above process, three different forward simulation results other than the background field were obtained every hour during the study period, and the differences were entirely caused by the different emission inventories used.

[0045] 4. Using the simulated hourly meteorological element data set (including surface pressure, boundary layer height, 2-meter water vapor mixing ratio, 10-meter wind direction and speed, and 2-meter temperature), SO 2 The concentration and prior emission source data are used as input features to train the emission inversion model. The training target is defined as the hourly SO2 difference between the perturbed emission source data and the prior emission source data. 2 The difference in emissions, i.e., the emission increment, is intended to allow the emission inversion model to learn and establish SO 2 The mapping relationship between emission increments and related independent variables. The perturbed emissions from January 20 to January 23, 2023 are generated using the same perturbation method as the training dataset. Assuming that the perturbed emissions of these four days are “real” emissions, the WRF_Chem model is used for forward simulation. The simulated meteorological element data and SO 2 The concentration was used as a feature input into the trained model to estimate the hourly SO 2 Emission increment. SO obtained by comparing the model inversion 2 The difference between the emissions and the “true” emissions is used to evaluate the inversion accuracy of the emissions inversion model. Figure 2 The SO predicted by the emission inversion model in the test set is shown 2 The correlation between emissions and “real” emissions shows that the SO2 predicted by the emission inversion model 2 The emissions agree very well with the “true” emissions, with a slope of 0.99, indicating that the model reproduces the perturbed SO 2 Emissions.

[0046] 5. Apply the three-dimensional variational assimilation method to convert the hourly surface SO 2 Observational data are assimilated into the model background field to optimize simulation results and generate SO 2 Concentration reanalysis field.

[0047] 6. Reanalyze the generated SO 2 The concentration reanalysis field, hourly meteorological element data, and prior emission source data are input into the trained emission inversion model to generate the corresponding SO 2emission increment field, and then SO 2 The emission increment field is superimposed on the prior emission field to obtain the final predicted SO 2 Emissions.

[0048] Finally, if Figure 3 As shown in the figure, SO2 is inverted by WRF-Chem model based on MEIC_2016 emission inventory and this application. 2 The concentration was simulated and analyzed, and the ground station observation data were used to verify the SO 2 Simulation results of concentration. Figure 3 The simulated SO under the two emission inventories at 17:00 UTC on January 21, 2023 are shown. 2 The scatter plot of the concentration interpolated to the ground station and the observed concentration at the station. Figure 3 It can be seen that the emission sources obtained by the method proposed in this application have a significant improvement effect on the forecast, and can well reproduce the SO during the concentrated period of fireworks setting off during the Spring Festival. 2 The peak concentration method has the ability to solve the emission estimation problem in short-term pollution events (large-scale fireworks shows or forest fires), which not only meets the needs of atmospheric pollution forecasting and control for high-temporal-resolution emission inventories, but also significantly reduces computing resource consumption and computing time. In addition, the emission inventory obtained based on this method can also be used to improve the pollutant concentration forecast level of atmospheric chemistry models, providing strong support for atmospheric pollution prevention, control, governance, forecasting and early warning.

[0049] In one embodiment, a single variable pollutant emission source inversion device is provided, comprising: The data acquisition module is used to obtain single-variable pollutant concentration observation data with short life cycle characteristics in the study area, meteorological reanalysis grid data and prior emission source data; A disturbance module is used to disturb the prior emission source data by using an enhanced disturbance strategy, a reduced disturbance strategy and a cross-level adjustment disturbance strategy to obtain disturbed emission source data; The forward simulation module is used to use the atmospheric chemical model and the disturbed emission source data to perform hourly forward simulation of the meteorological field and pollutant concentration field in the study area to obtain hourly resolution meteorological element data sets and pollutant concentration data sets; The model training module is used to input the meteorological element dataset, the pollutant concentration dataset and the prior emission source data into the pre-built emission inversion model based on deep learning for training, so as to obtain a trained emission inversion model; The variational assimilation module is used to assimilate the hourly pollutant concentration observation data in the study area using a three-dimensional variational assimilation method to generate a pollutant concentration reanalysis field; The emission inversion module is used to input the meteorological element dataset, pollutant concentration reanalysis field and prior emission source data into the trained emission inversion model for prediction, obtain the emission increment field, and obtain the pollutant emission inventory with hourly resolution in the study area by superimposing the emission increment field onto the prior emission field.

[0050] For the specific definition of the single variable pollutant emission source inversion device, please refer to the definition of the single variable pollutant emission source inversion method above, which will not be repeated here. Each module in the above-mentioned single variable pollutant emission source inversion device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0051] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a single variable pollutant emission source inversion method is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0052] Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0053] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented: Obtain single variable pollutant concentration observation data with short life cycle characteristics in the study area, meteorological reanalysis grid data and prior emission source data; The enhanced disturbance strategy, the weakened disturbance strategy and the cross-level adjustment disturbance strategy are respectively used to disturb the prior emission source data to obtain the disturbed emission source data; Using atmospheric chemistry models and disturbed emission source data, the meteorological field and pollutant concentration field in the study area are simulated hourly to obtain meteorological element data sets and pollutant concentration data sets with hourly resolution. The meteorological element dataset, pollutant concentration dataset and prior emission source data are input into a pre-built emission inversion model based on deep learning for training to obtain a trained emission inversion model; The three-dimensional variational assimilation method is used to assimilate the hourly pollutant concentration observation data in the study area to generate the pollutant concentration reanalysis field; The meteorological element dataset, pollutant concentration reanalysis field and prior emission source data are input into the trained emission inversion model for prediction to obtain the emission increment field. By superimposing the emission increment field on the prior emission field, the pollutant emission inventory with hourly resolution in the study area is obtained.

[0054] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented: Obtain single variable pollutant concentration observation data with short life cycle characteristics in the study area, meteorological reanalysis grid data and prior emission source data; The enhanced disturbance strategy, the weakened disturbance strategy and the cross-level adjustment disturbance strategy are respectively used to disturb the prior emission source data to obtain the disturbed emission source data; Using atmospheric chemistry models and disturbed emission source data, the meteorological field and pollutant concentration field in the study area are simulated hourly to obtain meteorological element data sets and pollutant concentration data sets with hourly resolution. The meteorological element dataset, pollutant concentration dataset and prior emission source data are input into a pre-built emission inversion model based on deep learning for training to obtain a trained emission inversion model; The three-dimensional variational assimilation method is used to assimilate the hourly pollutant concentration observation data in the study area to generate the pollutant concentration reanalysis field; The meteorological element dataset, pollutant concentration reanalysis field and prior emission source data are input into the trained emission inversion model for prediction to obtain the emission increment field. By superimposing the emission increment field on the prior emission field, the pollutant emission inventory with hourly resolution in the study area is obtained.

[0055] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0056] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0057] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A single variable pollutant emission source inversion method, characterized in that: The method comprises: Obtain single variable pollutant concentration observation data with short life cycle characteristics in the study area, meteorological reanalysis grid data and prior emission source data; The prior emission source data are disturbed by using an enhanced disturbance strategy, a weakened disturbance strategy, and a cross-level adjustment disturbance strategy to obtain disturbed emission source data; Using the atmospheric chemistry model and the emission source data after the disturbance, the meteorological field and pollutant concentration field in the study area are forward simulated every hour to obtain the meteorological element data set and pollutant concentration data set with hourly resolution; Inputting the meteorological element dataset, the pollutant concentration dataset and the prior emission source data into a pre-built emission inversion model based on deep learning for training, thereby obtaining a trained emission inversion model; The three-dimensional variational assimilation method is used to assimilate the hourly pollutant concentration observation data in the study area to generate the pollutant concentration reanalysis field; The meteorological element dataset, pollutant concentration reanalysis field and prior emission source data are input into the trained emission inversion model for prediction to obtain the emission increment field. By superimposing the emission increment field onto the prior emission field, a pollutant emission inventory with hourly resolution in the study area is obtained.

2. The method according to claim 1, characterized in that After obtaining the single variable pollutant concentration observation data with short life cycle characteristics in the study area, meteorological reanalysis grid data and prior emission source data, it also includes: The pollutant concentration observation data are quality controlled, and the pollutant concentration observation data falling within the same grid are averaged according to the study area and grid resolution of the WRF-Chem model.

3. The method according to claim 1, characterized in that The prior emission source data are disturbed by using an enhanced disturbance strategy, a weakened disturbance strategy, and a cross-level adjustment disturbance strategy to obtain disturbed emission source data, including: According to the actual increase, decrease and normal emission ratio of pollution sources, the grid ratio of increased emission, reduced emission and normal emission in the study area is set every hour; Create a set of random numbers that conform to the normal distribution and have the same dimension as the prior emission source data, perform Gaussian filtering on the random numbers, and perform initial perturbations by multiplying the emissions in each type of grid by random numbers that obey different means and standard deviations and conform to the normal distribution, to obtain perturbed emissions in enhanced emission grids, weakened emission grids, and normal emission grids; The enhanced disturbance strategy is adopted, that is, based on the initial disturbance, the disturbed emissions in all grids are multiplied by a random number greater than 1 generated from a uniform distribution to obtain an enhanced emission inventory for simulating emission source enhancement; The strategy of reducing disturbance is adopted, that is, based on the initial disturbance, the disturbed emissions in all grids are multiplied by a random number less than 1 generated from a uniform distribution to obtain the reduced emission inventory used to simulate the reduction of emission sources; A cross-level adjustment disturbance strategy is adopted, that is, before the initial disturbance, the emission at each moment in the prior emission source data is adjusted by adding a random number in a set interval. At the same time, the emissions are divided into 8 intervals according to the preset threshold, and the emissions in each interval are perturbed by cross-level adjustment. Then, the emissions in each grid are multiplied by a random number that obeys different means and standard deviations and conforms to the normal distribution, so as to obtain an emission inventory in which the emissions are controlled within the preset range and is used to simulate the migration of existing emission sources and the formation of new emission sources.

4. The method according to claim 3, characterized in that: The initial perturbation is performed by multiplying the emissions in each type of grid by random numbers with different means and standard deviations and in accordance with normal distribution, including: The emissions in the enhanced emission grid are initially disturbed by multiplying a random number whose mean is greater than 1, the emissions in the weakened emission grid are initially disturbed by multiplying a random number whose mean is less than 1, and the emissions in the normal emission grid are initially disturbed by multiplying a random number whose mean is equal to 1.

5. The method according to claim 3, characterized in that: The emissions in each interval are adjusted across levels, including: Two types of grid proportions are set in the study area. For one type of grid, a random number is directly added to the emission in the grid to make a disturbance, so that the emission value after disturbance changes in magnitude; for the other type of grid, the emission in the grid is multiplied by a random number to make a disturbance; When disturbing emissions in different intervals, the situations at different times are taken into account. For moments when the prior emissions are overestimated, emissions are reduced, and for moments when the prior emissions are underestimated, new emission sources are added.

6. The method according to claim 3, characterized in that Using the atmospheric chemistry model and the disturbed emission source data, the meteorological field and pollutant concentration field in the study area are simulated hourly forward to obtain hourly resolution meteorological element data sets and pollutant concentration data sets, including: Based on the prior emission source data, hourly forward simulations were performed during the study period, with the simulation time window set to 1 hour, and the hourly simulation results were used as the background field; Taking the background field at the initial moment of the study as the initial field, the atmospheric chemistry model and the emission inventory obtained by three perturbation strategies are used to perform hourly forward simulations on the meteorological field and pollutant concentration field in the study area to obtain meteorological element datasets and pollutant concentration datasets with hourly resolution.

7. The method according to claim 1, characterized in that The meteorological element dataset, pollutant concentration dataset and prior emission source data are input into a pre-built emission inversion model based on deep learning for training to obtain a trained emission inversion model, including: The hourly-resolution meteorological element dataset, pollutant concentration dataset and prior emission source data obtained by forward simulation are used as input features and input into the deep learning-based emission inversion model for training. The training target is defined as the difference in hourly emissions between the disturbed emission source data and the prior emission source data, that is, the emission increment. The model inversion accuracy is evaluated by comparing the difference between the emissions predicted by the model and the emission source data after the disturbance, until a trained emission inversion model that meets the preset inversion accuracy is obtained.

8. A single variable pollutant emission source inversion device, characterized in that: The device comprises: The data acquisition module is used to obtain single-variable pollutant concentration observation data with short life cycle characteristics in the study area, meteorological reanalysis grid data and prior emission source data; A disturbance module, used to disturb the prior emission source data by using an enhanced disturbance strategy, a reduced disturbance strategy and a cross-level adjustment disturbance strategy to obtain disturbed emission source data; A forward simulation module is used to use the atmospheric chemistry model and the emission source data after the disturbance to perform hourly forward simulation of the meteorological field and pollutant concentration field in the study area to obtain a meteorological element data set and a pollutant concentration data set with hourly resolution; A model training module, used for inputting the meteorological element data set, the pollutant concentration data set and the prior emission source data into a pre-built emission inversion model based on deep learning for training, so as to obtain a trained emission inversion model; The variational assimilation module is used to assimilate the hourly pollutant concentration observation data in the study area using a three-dimensional variational assimilation method to generate a pollutant concentration reanalysis field; The emission inversion module is used to input the meteorological element data set, pollutant concentration reanalysis field and prior emission source data into the trained emission inversion model for prediction, obtain the emission increment field, and obtain the pollutant emission inventory with hourly resolution in the study area by superimposing the emission increment field onto the prior emission field.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images