WO2023165336A1

WO2023165336A1 - Atmospheric pollutant emission flux processing method, storage medium, and computer terminal

Info

Publication number: WO2023165336A1
Application number: PCT/CN2023/076324
Authority: WO
Inventors: 姚易辰; 陈国栋; 李展; 陈磊; 仲晓辉; 胡媛; 杜飞; 王志斌; 李�昊
Original assignee: 阿里巴巴达摩院(杭州)科技有限公司
Priority date: 2022-03-03
Filing date: 2023-02-16
Publication date: 2023-09-07
Also published as: CN114324780A; CN114324780B

Abstract

An atmospheric pollutant emission flux processing method, a storage medium, and a computer terminal. The method comprises: acquiring an observed value of the concentration of an atmospheric pollutant and an initial emission flux of the atmospheric pollutant (S202), wherein the observed value is used for representing a numerical value obtained by measuring the concentration of a collected atmospheric pollutant by a target station; processing the observed value and the initial emission flux by using an atmospheric propagation model to obtain a target derivative of the concentration of the atmospheric pollutant (S204), wherein the atmospheric propagation model is constructed by means of a deep learning framework, the atmospheric propagation model is used for representing a dynamic propagation process of the atmospheric pollutant in the atmosphere, and the target derivative is used for representing a derivative of the observed value relative to the emission flux; and updating the initial emission flux on the basis of the target derivative to obtain a target emission flux of the atmospheric pollutant (S206). The technical problem in related technology of relatively low monitoring accuracy of the concentration of an atmospheric pollutant is solved.

Description

Atmospheric pollutant emission flux processing method, storage medium and computer terminal

This application claims the priority of the Chinese patent application submitted to the China Patent Office on March 3, 2022, with the application number 202210200399.2, and the application name is "Method for processing air pollutant emission flux, storage medium, and computer terminal", the entire content of which Incorporated in this application by reference.

technical field

The invention relates to the field of air pollutant emission flux processing, in particular to a method for processing air pollutant emission flux, a storage medium and a computer terminal.

Background technique

The greenhouse gas carbon dioxide plays a vital role in the circulation of the Earth's atmosphere. Carbon dioxide absorbs surface radiation, trapping more heat in the atmosphere, which increases temperatures. At present, the monitoring and consideration of the achievement of carbon dioxide emission reduction targets relies on accurate carbon emission monitoring schemes, but due to the influence of weather and other factors, it is difficult to accurately monitor carbon emissions.

For the above problems, no effective solution has been proposed yet.

Contents of the invention

Embodiments of the present invention provide a method for processing air pollutant emission flux, a storage medium, and a computer terminal, so as to at least solve the technical problem of low monitoring accuracy of air pollutant concentration in the related art.

According to an aspect of an embodiment of the present invention, a method for processing air pollutant emission flux is provided, including: obtaining the observed value of the concentration of air pollutants and the initial emission flux of the concentration of air pollutants, wherein the observed value is used to represent The concentration of atmospheric pollutants collected at the target site; use the atmospheric transport model to process the observations and initial emission flux to obtain the target derivative of the atmospheric pollutant concentration, where the atmospheric transport model is constructed through a deep learning framework, and the atmospheric transport model It is used to characterize the dynamic transmission process of air pollutant concentration in the atmosphere, and the target derivative is used to characterize the derivative of the observed value relative to the emission flux; based on the target derivative, the initial emission flux is updated to obtain the target emission of air pollutant concentration flux.

According to another aspect of an embodiment of the present invention, a method for processing air pollutant emission flux is provided, including: obtaining the observed value of carbon concentration and the initial carbon emission flux in a carbon-neutral scenario, wherein the observed value is used to represent The value obtained by measuring the carbon concentration at the target site; the observed value and the initial carbon emission flux are processed by the atmospheric transfer model to obtain the target derivative. The dynamic transmission process of the gas with high concentration in the atmosphere, the target derivative is used to characterize the derivative of the observed value relative to the carbon emission flux; based on the target derivative, the initial carbon emission flux is updated to obtain the target carbon emission flux.

According to another aspect of the embodiments of the present invention, a method for processing air pollutant emission flux is provided, including: displaying the observed value of the concentration of air pollutants and the initial emission flux of the concentration of air pollutants in an interactive interface, wherein, The observed value is used to represent the value obtained by measuring the concentration of air pollutants at the target site; in response to the touch operation detected in the interactive interface, the observed value and the initial emission flux are processed using the atmospheric transfer model to obtain the air pollution The target derivative of the pollutant concentration, in which the atmospheric transport model is constructed through a deep learning framework, the atmospheric transport model is used to characterize the kinetic transport process of the atmospheric pollutant concentration in the atmosphere, and the target derivative is used to characterize the derivative of the observed value relative to the emission flux ;Display the target emission flux of atmospheric pollutant concentration in the interactive interface, where the target emission flux passes through the target Derivatives are obtained by updating the initial emission flux.

According to another aspect of the embodiments of the present invention, a method for processing air pollutant emission flux is provided, including: the cloud server receives the observed value of the air pollutant concentration and the initial emission flux of the air pollutant concentration uploaded by the client, Among them, the observed value is used to represent the value obtained by measuring the concentration of atmospheric pollutants at the target site; the cloud server uses the atmospheric transfer model to process the observed value and the initial emission flux to obtain the target derivative of the concentration of atmospheric pollutants, where, The atmospheric transport model is constructed through a deep learning framework. The atmospheric transport model is used to characterize the dynamic transport process of atmospheric pollutant concentrations in the atmosphere. The target derivative is used to represent the derivative of the observed value relative to the emission flux; the cloud server based on the target derivative The emission flux is updated to obtain the target emission flux of atmospheric pollutant concentration; the cloud server returns the target emission flux to the client.

According to another aspect of the embodiments of the present invention, a method for processing air pollutant emission flux is provided, including: the cloud server receives an air pollutant emission flux processing request uploaded by a client, wherein the air pollutant emission flux processing The request at least includes: a preset time period and the concentration of atmospheric pollutants; the cloud server obtains the observed value of the concentration of atmospheric pollutants within the preset time period and the initial emission flux of the concentration of atmospheric pollutants, wherein the observed value is used to represent the The value obtained by measuring the concentration of atmospheric pollutants; the cloud server uses the atmospheric transport model to process the observed value and the initial emission flux to obtain the target derivative of the atmospheric pollutant concentration. The atmospheric transport model is constructed through a deep learning framework, and the atmospheric The transport model is used to characterize the dynamic transport process of atmospheric pollutant concentration in the atmosphere, and the target derivative is used to characterize the derivative of the observed value relative to the emission flux; the cloud server updates the initial emission flux based on the target derivative to obtain the atmospheric pollutant concentration target emission flux; the cloud server returns the target emission flux to the client.

According to another aspect of the embodiments of the present application, there is also provided a storage medium, the storage medium includes a stored program, wherein, when the program is running, the device where the computer-readable storage medium is located is controlled to execute the air pollutant in any one of the above-mentioned embodiments. Emission flux treatment method.

According to another aspect of the embodiments of the present application, there is also provided a computer terminal, including: a processor and a memory, the processor is used to run the program stored in the memory, wherein, when the program is running, the air pollution in any one of the above embodiments is executed Treatment method of waste emission flux.

Through the above steps, firstly, the observed value of the concentration of atmospheric pollutants and the initial emission flux of the concentration of atmospheric pollutants are obtained, wherein the observed values are used to represent the concentration of the concentration of atmospheric pollutants collected at the target site; Value and initial emission flux are processed to obtain the target derivative of the concentration of atmospheric pollutants. The atmospheric transport model is constructed through a deep learning framework. The atmospheric transport model is used to characterize the dynamic transport process of the concentration of atmospheric pollutants in the atmosphere. The target derivative It is used to characterize the derivative of the observed value relative to the emission flux; based on the target derivative, the initial emission flux is updated to obtain the target emission flux of the concentration of atmospheric pollutants, which realizes the purpose of improving the monitoring of the target emission flux. It is easy to notice that the observed value of atmospheric pollutant concentration and the initial emission flux of atmospheric pollutant concentration can be obtained first, and then the observed value and initial emission flux can be compared according to the power transfer model of atmospheric pollutant concentration in the atmosphere of the atmospheric transfer model In order to determine the influence factors of environmental factors in the atmosphere on the prediction accuracy, and determine the observation error according to the influence factors, update the initial emission flux according to the target derivative determined by the observation error, so as to improve the initial emission flux The accuracy of the flux, and thus solve the technical problem of low accuracy in monitoring the concentration of air pollutants in related technologies.

Description of drawings

The accompanying drawings described here are used to provide a further understanding of the present invention and constitute a part of the application. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations to the present invention. In the attached picture:

Fig. 1 is a kind of hardware structural block diagram of the computer terminal (or mobile device) that is used to realize the air pollutant discharge flux processing method according to the embodiment of the present invention;

Fig. 2 is a flow chart of a method for processing air pollutant emission flux according to an embodiment of the present invention;

3 is a schematic diagram of surface carbon flux inversion based on an automatic gradient method according to an embodiment of the present invention;

Fig. 4 is a flowchart of another air pollutant emission flux processing method according to an embodiment of the present invention;

Fig. 5 is a flowchart of another air pollutant emission flux processing method according to an embodiment of the present invention;

Fig. 6 is a flowchart of another air pollutant emission flux processing method according to an embodiment of the present invention;

Fig. 7 is a flowchart of another air pollutant emission flux processing method according to an embodiment of the present invention;

8 is a schematic diagram of another air pollutant emission flux processing device according to an embodiment of the present invention;

9 is a schematic diagram of another air pollutant emission flux processing device according to an embodiment of the present invention;

Fig. 10 is a schematic diagram of another air pollutant emission flux processing device according to an embodiment of the present invention;

Fig. 11 is a schematic diagram of another air pollutant emission flux processing device according to an embodiment of the present invention;

Fig. 12 is a schematic diagram of another air pollutant emission flux processing device according to an embodiment of the present invention;

Fig. 13 is a structural block diagram of a computer terminal according to an embodiment of the present application.

Detailed ways

In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only It is an embodiment of a part of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present invention.

It should be noted that the terms "first" and "second" in the description and claims of the present invention and the above drawings are used to distinguish similar objects, but not necessarily used to describe a specific sequence or sequence. It is to be understood that the data so used are interchangeable under appropriate circumstances such that the embodiments of the invention described herein can be practiced in sequences other than those illustrated or described herein. Furthermore, the terms "comprising" and "having", as well as any variations thereof, are intended to cover a non-exclusive inclusion, for example, a process, method, system, product or device comprising a sequence of steps or elements is not necessarily limited to the expressly listed instead, may include other steps or elements not explicitly listed or inherent to the process, method, product or apparatus.

First of all, some nouns or terms that appear during the description of the embodiments of the present application are applicable to the following explanations:

Automatic gradient: The principle of automatic differentiation is to decompose the mathematical operation into some basic operations, and then use the chain rule to calculate the gradient of the loading function. Existing common deep learning frameworks, such as deep learning library (tensorflow), scientific computing library (pytorch), etc., have integrated automatic derivation functions. Usually, only the forward process is encoded, and the calculation graph is constructed, and the automatic derivation capability of the framework can be used to obtain the derivative of the output to the input.

Atmospheric carbon flux inversion: Through atmospheric transport models and corresponding inversion algorithms, the observed values of major greenhouse gases, such as carbon dioxide, methane, and other concentration distributions, can be used to obtain information on global carbon sources and carbon sinks.

At present, for the inversion of surface carbon emissions from carbon dioxide concentration monitoring, common solutions include variational adjoint methods and ensemble Kalman filtering schemes.

The variational adjoint method first linearizes the nonlinear dynamical model and then obtains its linear adjoint form to construct a linearized state transition. Then, using the variational method, the derivative of the error function with respect to the input parameters is calculated. Its disadvantage is that the tangent mode relies on the first-order Taylor expansion of the original numerical format, and there is a large numerical error.

The ensemble Kalman filtering method, through the ensemble forecasting method, completes the state transition of variables and the covariance matrix state transition. At the observation point, the state value and the corresponding covariance matrix are corrected by the Bayesian method. The disadvantage of ensemble Kalman filtering is that ensemble forecasting usually consumes large computing resources and storage resources, and usually can only be updated in a single direction in time series.

In order to solve the above problems, this application provides a method for processing the emission flux of air pollutants, which updates the initial emission flux of carbon dioxide by performing reverse deduction on the observed value of carbon dioxide concentration, so as to achieve the target emission with high accuracy flux.

Example 1

According to an embodiment of the present invention, an embodiment of a method for processing air pollutant emission flux is also provided. It should be noted that the steps shown in the flowcharts of the accompanying drawings can be implemented in a computer system such as a set of computer-executable instructions and, although a logical order is shown in the flowcharts, in some cases the steps shown or described may be performed in an order different from that shown or described herein.

The method embodiment provided in Embodiment 1 of the present application may be executed in a mobile terminal, a computer terminal, or a similar computing device. Fig. 1 shows a block diagram of a hardware structure of a computer terminal (or mobile device) for implementing a method for processing air pollutant emission flux. As shown in Figure 1, the computer terminal 10 (or mobile device 10) may include one or more (shown by 102a, 102b, ..., 102n in the figure) processors (processors may include but not limited to microprocessors) MCU or a processing device such as a programmable logic device FPGA), a memory 104 for storing data, and a transmission module 106 for communication functions. In addition, it can also include: a display, an input/output interface (I/O interface), a universal serial bus (USB) port (which can be included as one of the ports of the BUS bus), a network interface, a power supply, and/or or camera. Those of ordinary skill in the art can understand that the structure shown in FIG. 1 is only a schematic diagram, and it does not limit the structure of the above-mentioned electronic device. For example, computer terminal 10 may also include more or fewer components than shown in FIG. 1 , or have a different configuration than that shown in FIG. 1 .

It should be noted that the one or more processors and/or other data processing circuits described above may generally be referred to herein as "data processing circuits". The data processing circuit may be implemented in whole or in part as software, hardware, firmware or other arbitrary combinations. In addition, the data processing circuit can be a single independent processing module, or be fully or partially integrated into any of the other elements in the computer terminal 10 (or mobile device). As mentioned in the embodiment of the present application, the data processing circuit is used as a processor control (for example, the selection of the terminal path of the variable resistor connected to the interface).

The memory 104 can be used to store software programs and modules of application software, such as the program instruction/data storage device corresponding to the air pollutant emission flux processing method in the embodiment of the present invention, the processor runs the software program stored in the memory 104 and module, so as to execute various functional applications and data processing, that is, to realize the above-mentioned air pollutant emission flux processing method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include a memory that is remotely located relative to the processor, and these remote memories may be connected to the computer terminal 10 through a network. Examples of the aforementioned networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The transmission device 106 is used to receive or transmit data via a network. The specific example of the above-mentioned network may include a wireless network provided by the communication provider of the computer terminal 10 . In one example, the transmission device 106 includes a network interface controller (NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (Radio Frequency, RF) module, which is used to communicate with the Internet in a wireless manner.

The display may be, for example, a touchscreen liquid crystal display (LCD), which may enable a user to interact with the user interface of the computer terminal 10 (or mobile device).

It should be noted here that, in some optional embodiments, the computer device (or mobile device) shown in FIG. 1 may include hardware components (including circuits), software components (including computer code), or a combination of both hardware and software elements. It should be noted that FIG. 1 is only one example of a particular embodiment, and is intended to illustrate the types of components that may be present in a computer device (or mobile device) as described above.

Under the above operating environment, the present application provides a treatment method for air pollutant emission flux as shown in FIG. 2 . Fig. 2 is a flowchart of a method for processing air pollutant emission flux according to an embodiment of the present invention.

Step S202, obtaining the observed value of the air pollutant concentration and the initial emission flux of the air pollutant concentration.

Among them, the observed value is used to represent the concentration of air pollutants collected by the target site.

The air pollutant concentration mentioned above may be carbon dioxide, PM2.5, sulfide, suspended particulate matter (such as dust, smoke) and the like.

The aforementioned initial emission flux may be an estimated emission flux of atmospheric pollutant concentration, or alternatively, may be an estimated emission flux of atmospheric pollutant concentration based on historical emission conditions. The aforementioned initial discharge flux can also be set to any value.

The above-mentioned target stations may be one or more preset stations, which are used to collect concentrations of atmospheric pollutants. Wherein, the target site may be a site set on the ground, and the target site may also be a satellite remote sensing device.

In an optional embodiment, the observed value of the air pollutant concentration can be the observed value of the air pollutant concentration within a period of time, and the concentration of the air pollutant concentration can be determined according to the observed value of the air pollutant concentration in a period of time Because the concentration of atmospheric pollutants will be affected by natural meteorology and other aspects during the emission process, the concentration of atmospheric pollutants will change over a period of time, for example, the carbon neutrality of plants, wind, Influenced by rain etc. Therefore, the emission fluxes of atmospheric pollutant concentrations obtained from observations are less accurate. At this time, the factors affecting the emission flux can be obtained by analyzing the observed values of the concentration of atmospheric pollutants over a period of time, and the initial emission flux can be updated according to this factor, so that the obtained target emission flux is more accurate. Accuracy.

Step S204, using the atmospheric transport model to process the observed value and the initial emission flux to obtain the target derivative of the atmospheric pollutant concentration.

Among them, the atmospheric transport model is constructed through a deep learning framework. The atmospheric transport model is used to characterize the dynamic transport process of the concentration of atmospheric pollutants in the atmosphere, and the target derivative is used to characterize the derivative of the observed value relative to the emission flux.

In an optional embodiment, the atmospheric transmission model is constructed through a deep learning framework, which can represent the dynamic transmission process of the concentration of atmospheric pollutants in the atmosphere. By analyzing the observations over a period of time, the atmospheric The operation law of pollutant concentration in the atmosphere, and based on the operation law, the initial emission flux is adjusted to make the data of the initial emission flux more accurate.

In another optional embodiment, the prediction can be made based on the observed value at the beginning of the observation and the initial emission flux, and the observed value at the end of the observation can be obtained from the prediction. Observed values are compared to determine the error between the two, and the error of the initial emission flux can be determined according to the error. Optionally, the difference between the error between the observed values and the initial emission flux can be expressed in the form of a derivative Furthermore, the initial emission flux can be updated based on the target derivative, so as to obtain a more accurate target emission flux.

Step S206, based on the target derivative, the initial emission flux is updated to obtain the target emission flux of the air pollutant concentration.

The above target derivative is used to convert the error in the observation value into the error in the emission flux.

In an optional embodiment, since the target derivative is the data obtained according to the observation error, the initial emission flux can be updated according to the target derivative to reduce the observation error of the initial emission flux, so that the accuracy can be obtained Higher target emission flux.

In an optional embodiment, in a carbon trading scenario, carbon trading can be realized by obtaining the target emission flux of atmospheric pollutant concentration, wherein carbon trading refers to using carbon dioxide emission rights as a commodity, and the buyer By paying a certain amount to the seller, a certain amount of carbon dioxide emission rights can be obtained, thereby forming a transaction of carbon dioxide emission rights. The observed value of carbon dioxide and the initial emission flux of carbon dioxide can be obtained, and the observed value and initial emission flux can be processed through the atmospheric transport model to obtain the target derivative of carbon dioxide, and the initial emission flux can be updated according to the target derivative to obtain the air pollutant Concentration target emission flux, because it is obtained by deducing the initial emission flux from the actual observation value, so the accuracy of the target emission flux obtained is relatively high. Therefore, in the carbon trading scenario, it can be Carry out carbon trading based on highly accurate target emission fluxes, thereby reducing related economic losses.

In an optional embodiment, the carbon emission flux can be estimated in a new energy scenario, such as a new energy vehicle scenario, and the carbon emission flux that can be reduced by using a new energy vehicle can be determined. Optional The observed value of carbon dioxide emitted in the new energy vehicle operating area and the initial emission flux of carbon dioxide can be obtained, and then the observed value and the initial emission flux of the new energy vehicle operating area are processed using the atmospheric transport model to obtain the carbon dioxide emission flux The target derivative, the initial emission flux can be updated according to the target derivative, and the target emission flux of carbon dioxide in the new energy vehicle operating area can be obtained, and the emission flux of the historical time period in which no new energy vehicles are used in the area can be obtained. By comparing The emission flux and target emission flux in the historical time period can determine the emission reduction effect of new energy vehicles. In another optional embodiment, the carbon emission flux can be estimated in the autonomous driving scenario. Since the vehicles in the autonomous driving scenario are driven by new energy sources, it has an important role in reducing the carbon emission flux. certain effect, in order to obtain the observed value of carbon dioxide emitted in the operating area of the autonomous vehicle and the initial emission flux of carbon dioxide, and then use the atmospheric transport model to process the observed value and the initial emission flux in the operating area of the autonomous vehicle to obtain The target derivative of carbon dioxide, the initial emission flux can be updated according to the target derivative, and the target emission flux of carbon dioxide in the operating area of the automatic driving vehicle can be obtained, and the emission flux of the historical period when the automatic driving vehicle is not used in the area can be obtained, By comparing the emission flux of the historical time period with the target emission flux, the emission reduction effect of the autonomous vehicle can be determined.

In the above-mentioned embodiments of the present application, the observed value includes: the concentration of air pollutant concentration collected within a preset time period, Using the atmospheric transfer model to process the observed values and initial emission fluxes to obtain the target derivatives of atmospheric pollutant concentrations includes: using the atmospheric transfer model to perform reverse deduction on the initial observed values and initial emission fluxes to obtain the prediction of atmospheric pollutant concentrations value, where the initial observation value is used to represent the concentration of air pollutants collected at the beginning of the preset time period, and the predicted value is used to represent the concentration of air pollutants deduced within the preset time period; based on the prediction Value and observation value, determine the observation error of the concentration of atmospheric pollutants; obtain the derivative of the observation error relative to the emission flux, and obtain the target derivative.

The aforementioned preset time period can be set by itself.

In an optional embodiment, the atmospheric transport model can be used to conduct reverse deduction on the initial observation value and initial emission flux to obtain the predicted value of the concentration of atmospheric pollutants. Optionally, the reverse deduction can be carried out During the deduction process, weather is combined with deduction, so that the factors that affect the emission flux data can be considered during the deduction process, and the predicted value of the concentration of atmospheric pollutants within the preset time period can be deduced, wherein the predicted value can be the preset time The value of the concentration of atmospheric pollutants obtained from the prediction of a certain point within the range, the observation error of the concentration of atmospheric pollutants can be determined according to the predicted value obtained by reverse deduction and the value at this point obtained by actual observation, so as to determine the concentration of atmospheric pollutants according to the observation error Determine the error of the initial emission flux of air pollutant concentration. Optionally, the target derivative for updating the initial emission flux can be determined by obtaining the derivative of the observation error relative to the emission flux.

In the above embodiments of the present application, the atmospheric transport model includes: a derivative calculation layer, a differential equation construction layer and a time integration layer.

The above-mentioned derivative calculation layer is used to determine the spatial derivative according to the relevant data of the atmospheric pollutant concentration and the velocity field of the atmosphere.

The above-mentioned differential equation construction layer is used to construct the target differential equation according to the spatial derivative and the initial emission flux.

The above-mentioned time integration layer is used to perform an integral operation on the target differential equation to obtain a predicted value.

In the above-mentioned embodiments of the present application, the atmospheric transport model is used to reversely deduce the initial observation value and the initial emission flux, and the predicted value of the air pollutant concentration includes: using the derivative calculation layer to calculate the density of the air pollutant concentration and the velocity of the atmosphere The differential operation of the field and the initial observation value is performed to obtain the spatial derivative; the differential equation construction layer is used to construct the equation of the spatial derivative and the initial emission flux, and the target differential equation is obtained, which has nothing to do with the time function; the time integration layer is used to obtain the target differential equation The equation is integrated to obtain the predicted value.

In an optional embodiment, the predicted value can be obtained by the following formula:

Among them, ρ is the density of air pollutant concentration, is the velocity field of the atmosphere, is the spatial derivative, S _c is the initial emission flux, is the predicted value, is the symbol for the difference operation.

In another optional embodiment, the derivative calculation layer can be used to perform a differential operation on the density of atmospheric pollutant concentration, the velocity field of the atmosphere, and the initial observation value to obtain the spatial derivative, so as to determine the space of the initial observation value in the atmosphere distribution, and then use the differential equation to construct the equation of the spatial derivative and the initial emission flux to obtain the target differential method, so as to determine the movement of the initial emission flux in space according to the spatial derivative, and use the time integration layer to carry out the target differential equation Integral operation, so as to determine the predicted value of the concentration of atmospheric pollutants obtained by reverse deduction within the preset time period. By comparing the predicted value with the actual observed value, the observation error can be determined.

In the above-mentioned embodiments of the present application, after updating the initial emission flux based on the target derivative, the air pollutant concentration After the target emission flux of 1 degree, the method also includes: using the atmospheric transport model to process the initial observation value and the target emission flux to obtain the target predicted value of the concentration of atmospheric pollutants; based on the target predicted value and the target emission flux, determine Whether the target emission flux satisfies the preset condition; in response to the target emission flux meeting the preset condition, based on the target predicted value, determine the first derivative of the atmospheric pollutant concentration, wherein the first derivative is used to characterize the observation error relative to the target emission The derivative of the amount; the target emission flux is updated based on the first derivative.

The above-mentioned target predicted value may be a predicted value at a certain moment in the predicted value. It should be noted that the prediction process can be continuous, and the initial observation value and initial emission flux can be reversely deduced through the atmospheric transport model to obtain the predicted values at all moments within the preset time period.

The aforementioned preset condition may be a condition that the iteration does not converge. That is, the target discharge flux does not meet the required accuracy.

In an optional embodiment, after the target discharge flux is obtained, the target discharge flux can be used as the initial discharge flux for reverse deduction, so as to verify whether the target discharge flux reaches the required accuracy , Optionally, the atmospheric transport model can be used to deduce the initial observation value and the target emission flux inversely to obtain the target prediction value, and the observation error can be determined according to the real initial observation value and the predicted target prediction value, if If the observation error is large, it means that the target emission flux satisfies the condition that the iteration does not converge. At this time, the target emission flux needs to be updated again. Further, under the condition that the target emission flux satisfies the unconverged iteration, the first derivative of the atmospheric pollutant concentration can be determined according to the derivative of the observation error between the target predicted value and the real observed value relative to the emission flux, And the target emission flux is updated according to the first derivative, so as to further improve the accuracy of the target emission flux.

Further, if the target discharge flux does not satisfy the preset condition, it means that the accuracy of the target discharge flux has reached the required accuracy, and at this time, there is no need to update the target discharge flux.

In the above embodiments of the present application, based on the target predicted value and the target emission flux, determining whether the target emission flux meets the preset conditions includes: constructing a first error function based on the target predicted value and the observed value; based on the target emission flux and the initial The emission flux is used to construct a second error function; the weighted sum of the first error function and the second error function is obtained to obtain a loss function of the concentration of atmospheric pollutants; in response to the loss function being greater than a preset threshold, it is determined that the target emission flux satisfies the preset A condition; in response to the loss function being smaller than a preset threshold, it is determined that the target emission flux does not satisfy the preset condition.

The above-mentioned first error function is used to represent the error between the target predicted value and the real observed value.

The above-mentioned second error function is used to represent the error between the target emission flux and the initial emission flux, that is, the error between the prior emission flux and the finally obtained emission flux.

The aforementioned preset threshold can be set by itself.

In an optional embodiment, there are corresponding error covariance matrices for the above-mentioned observations and emission fluxes, that is, the above-mentioned first error function and second error function, by obtaining the first error function and the second The weighted sum of the two error functions can obtain the loss function of the concentration of atmospheric pollutants. When the loss function is greater than the preset threshold, it indicates that the error is large. At this time, the target emission flux can also be iteratively updated until the loss function is less than or equal to Preset threshold, if the loss function is less than the preset threshold, it can be determined that the target emission flux does not meet the preset conditions, at this time, the iterative update of the target emission flux can be stopped, and the target emission flux can be determined as the final emission flux .

In the above embodiments of the present application, updating the initial emission flux based on the target derivative to obtain the target emission flux of the air pollutant concentration includes: obtaining the weighted sum of the initial emission flux and the target derivative to obtain the target emission flux.

In an optional embodiment, the target emission flux of atmospheric pollutant concentration can be obtained by the following formula:

Among them, S _n can be the initial emission flux of the n-th iteration, S _n+1 can be the target emission flux of the n+1-th iteration, can be the target derivative, and L can be the observation error.

In the above embodiments of the present application, after updating the initial emission flux based on the target derivative to obtain the target emission flux of the air pollutant concentration, the method further includes: outputting the target emission flux; receiving the first emission flux corresponding to the target emission flux A feedback information, wherein the first feedback information is used to represent whether to update the target emission flux; in response to the first feedback information is to update the target emission flux, the observation value and the target emission flux are calculated using the atmospheric transfer model processing to obtain the second derivative of the concentration of air pollutants; based on the second derivative, the target emission flux is updated.

The above-mentioned first feedback information is used to indicate whether the atmospheric transport model needs to be used to update the target emission flux.

The above-mentioned second derivative is the derivative of the second iteration step.

In an optional embodiment, the target discharge flux can be output to the client, and the staff of the client can judge whether iterative update is needed according to the first feedback information, and if necessary, feedback can continue to update the target discharge flux To update, the observed value and the target emission flux can be processed by the atmospheric transport model to obtain the second derivative of the concentration of atmospheric pollutants, so as to continue to update the target emission flux based on the second derivative, and obtain the updated target emission flux quantity.

In the above-mentioned embodiments of the present application, after obtaining the observed value of the concentration of atmospheric pollutants and the initial emission flux of the concentration of atmospheric pollutants, the method further includes: outputting the observed value and the initial emission flux; receiving second feedback information, wherein, The second feedback information is obtained by modifying the observed value and the initial emission flux; the atmospheric transport model is used to process the second feedback information to obtain the third derivative of the concentration of atmospheric pollutants; the initial emission flux is updated based on the third derivative , to get the target emission flux.

In an optional embodiment, after obtaining the observed value of the atmospheric pollutant concentration and the initial emission flux of the atmospheric pollutant concentration, the observed value and the initial emission flux can be output to the client of the staff for working The personnel check the data accuracy of the observation value and the initial emission flux. If the data of the observation value or the initial emission flux are considered to be inaccurate, the observation value and the initial emission flux can be modified to obtain the second feedback information, which can be used The atmospheric transport model processes the second feedback information to obtain the third derivative of the concentration of air pollutants. The initial emission flux can be updated according to the third derivative to obtain the target emission flux.

In the above embodiments of the present application, after updating the initial emission flux based on the target derivative to obtain the target emission flux of the atmospheric pollutant concentration, the method further includes: based on the target emission flux, determining each grid point on the target map Grid emission fluxes for ; display grid emission fluxes on the target map.

The above-mentioned target map may be a map corresponding to a region where emission flux needs to be detected, and the target map may also be a global map.

The above-mentioned target emission flux may be the emission flux of air pollutant concentrations of multiple factories.

The above-mentioned grid points may be the locations on the target map of the factories that emit the concentration of air pollutants. The above-mentioned grid point emission flux may be the emission flux of the air pollutant concentration emitted by the factory.

In an optional embodiment, the grid point emission flux of the grid point where the factory is located on the target map may be determined according to the target emission flux, and the grid point emission flux is displayed on the target map. Optionally, it can display the grid point emission flux data on the grid point corresponding to the target map, and can also display the grid point emission flux on the grid point corresponding to the target map in the form of air mass, wherein the more air mass Larger means more emission flux, and smaller air mass means smaller emission flux; Other vector diagrams can also be used to represent the grid point emission flux corresponding to the grid point of the target map. The larger the vector diagram, the larger the emission flux, and the smaller the vector diagram, the smaller the emission flux.

In the above embodiments of the present application, after displaying the grid point emission fluxes on the target map, the method further includes: receiving the selected target area on the target map; summarizing the grid point emission fluxes of all grid points contained in the target area, Obtain the regional emission flux corresponding to the target area; output the regional emission flux.

In an optional embodiment, the user can also select the sum of emission fluxes corresponding to the target area from the target map. Optionally, the user can determine the target area on the target map, and can click on the corresponding area in the target area. The target area can also be determined by means of the method, and the target area can also be determined by selecting the area of the target map. After the target area is determined, the grid point emission fluxes of all grid points contained in the target area can be summarized to obtain the corresponding Regional emission flux, so that the regional emission flux of the target area can be output, so that the user can analyze the regional emission flux of the target area.

Figure 3 is a schematic diagram of surface carbon flux inversion based on the automatic gradient method. Among them, the carbon transport equation can be coded through the deep learning framework, and its automatic differentiation ability can be used to obtain the derivative term of the observation for the surface carbon source and sink, and then update the surface carbon flux. The entire inversion process mainly involves three processes, namely, the carbon dynamics transport process, the error calculation process, and the derivative update process.

For the carbon dynamics transport process, it may contain a dynamics module, which includes a derivative calculation layer, a differential equation construction layer and a time integration layer, wherein the derivative calculation layer includes density, The velocity field of the atmosphere and the initial observation value determine the spatial derivative, and the right-hand term of the partial differential equation can be constructed on the spatial derivative and the initial emission flux in the differential equation construction layer to obtain the target differential equation. After obtaining the target differential equation, the target differential equation can be The differential equation is time-integrated. Optionally, a single-step time integration can be performed on the target differential equation first, and then multi-step time integration can be performed to obtain a predicted value of the carbon dioxide concentration. It should be noted that the dynamics module is coded using a deep learning framework to build an overall calculation graph. The deep learning framework provides an automatic differentiation function for all tensor operations. It only needs to build a forward calculation graph to automatically obtain the gradient relationship between any two groups of tensors in the calculation graph. During the carbon dynamics transportation process, it realizes The initial carbon concentration and surface carbon sink flux, and the power transfer process under the transport of atmospheric wind speed, obtained the predicted value of the temporal and spatial distribution of carbon dioxide concentration.

For the error calculation process, it can determine the observation error function after obtaining the predicted value of carbon dioxide concentration and the observed value of carbon dioxide concentration, that is, the above-mentioned first error function, and can determine the background error according to the initial emission flux and target emission flux function, that is, the above-mentioned second error function, and the loss function of the concentration of air pollutants is obtained according to the weighted sum of the first error function and the second error function. Among them, in the process of carbon inversion, the carbon flux prior is generally obtained from the surface inventory model. Among them, the carbon flux prior is the above-mentioned initial emission flux.

For the derivation update process, after obtaining the loss function, it can determine whether the target carbon flux satisfies the condition of unconverged iteration. If so, it needs to continue to update the carbon flux. Optionally, it can be based on the carbon dynamics The automatic gradient calculation in the transport module is used to obtain the derivative of the loss function for any learnable parameter. When the learning parameter is carbon flux, the derivative of the loss function with respect to the carbon flux can be obtained, and then the carbon flux can be updated to obtain the carbon flux a posteriori. Among them, the carbon flux posterior is the above-mentioned target emission flux. The processing steps of the above modules can be iterated repeatedly, that is, the correction of the carbon flux according to the observed value can be realized. After the iteration converges, the carbon flux posterior can be used as the final result.

It should be noted that for the foregoing method embodiments, for the sake of simple description, they are expressed as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the described action sequence. Because of the present invention, certain steps may be performed in other orders or simultaneously. Secondly, those skilled in the art should also know It should be noted that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

Through the description of the above embodiments, those skilled in the art can clearly understand that the air pollutant emission flux processing method according to the above embodiments can be implemented by means of software plus a necessary general-purpose hardware platform, and of course it can also be realized by hardware , but in many cases the former is a better implementation. Based on such an understanding, the essence of the technical solution of the present invention or the part that contributes to the prior art can be embodied in the form of software products, and the computer software products are stored in a storage medium (such as ROM/RAM, disk, CD) contains several instructions to make a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) execute the method of each embodiment of the present invention.

Example 2

According to an embodiment of the present application, an embodiment of a method for processing air pollutant emission flux is also provided. 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, although a logical order is shown in the flowcharts, in some cases the steps shown or described may be performed in an order different from that shown or described herein.

Fig. 4 is a flow chart of a method for processing air pollutant emission flux according to an embodiment of the present invention. As shown in Fig. 4, the method may include the following steps:

Step S402, obtaining the observed value of carbon concentration and the initial carbon emission flux under the carbon neutral scenario.

Among them, the observed value is used to represent the value obtained by measuring the carbon concentration at the target site.

The above-mentioned carbon neutral scenario may be a scenario in which carbon dioxide emission flux is neutralized through afforestation, energy conservation and emission reduction.

The observed value of carbon concentration in the above-mentioned carbon neutral scenario may be the observed value of carbon dioxide concentration that has undergone carbon neutralization.

Step S404, using the atmospheric transport model to process the observed value and the initial carbon emission flux to obtain the target derivative of carbon dioxide.

Among them, the atmospheric transport model is constructed through a deep learning framework. The atmospheric transport model is used to characterize the dynamic transport process of gases that affect carbon concentration in the atmosphere, and the target derivative is used to characterize the derivative of the observed value relative to the carbon emission flux.

Step S406, based on the target derivative, the initial carbon emission flux is updated to obtain the target carbon emission flux of carbon concentration.

In the above embodiments of the present application, the observed value includes: the carbon concentration collected within a preset time period, and the observed value and the initial carbon emission flux are processed using the atmospheric transfer model, and the target derivative of the carbon concentration obtained includes: using the atmospheric transfer model The initial observation value and the initial carbon emission flux are reversely deduced to obtain the predicted value of carbon concentration. The initial observation value is used to represent the carbon concentration collected at the beginning of the preset time period, and the predicted value is used to represent the predicted value. Set the deduced carbon concentration in the time period; determine the observation error of carbon concentration based on the predicted value and observation value; obtain the derivative of the observation error relative to the emission flux to obtain the target derivative.

In the above-mentioned embodiments of the present application, the initial observation value and the initial carbon emission flux are reversely deduced using the atmospheric transport model, and the predicted value of the carbon concentration obtained includes: using the derivative calculation layer to calculate the density of the carbon concentration, the velocity field of the atmosphere, and the initial The difference operation is performed on the observed value to obtain the spatial derivative; the differential equation construction layer is used to construct the equation of the spatial derivative and the initial carbon emission flux to obtain the target differential equation, which has nothing to do with the time function; Integrate the operation to get the predicted value.

In the above embodiments of the present application, after updating the initial carbon emission flux based on the target derivative to obtain the target carbon emission flux of carbon concentration, the method further includes: using the atmospheric transport model to compare the initial observation value and the target carbon emission flux process to obtain the target predicted value of carbon concentration; based on the target predicted value and the target carbon emission flux, determine whether the target carbon emission flux meets the preset condition; in response to the target carbon emission flux meeting the preset condition, based on the target predicted value , to determine the first derivative of the carbon concentration, wherein the first derivative is used to characterize the derivative of the observation error relative to the target emission; based on the first derivative, the target carbon emission flux is updated.

In the above embodiments of the present application, based on the target predicted value and the target carbon emission flux, determining whether the target carbon emission flux meets the preset conditions includes: constructing a first error function based on the target predicted value and the observed value; amount and the initial carbon emission flux, construct the second error function; obtain the weighted sum of the first error function and the second error function, and obtain the loss function of carbon concentration; in response to the loss function being greater than the preset threshold, determine the target carbon emission flux A preset condition is met; in response to the loss function being smaller than a preset threshold, it is determined that the target carbon emission flux does not satisfy the preset condition.

In the above embodiments of the present application, after updating the initial carbon emission flux based on the target derivative to obtain the target carbon emission flux of carbon concentration, the method further includes: outputting the target carbon emission flux; receiving the target carbon emission flux corresponding to The first feedback information, wherein, the first feedback information is used to represent whether to update the target carbon emission flux; in response to the first feedback information is to update the target carbon emission flux, use the atmospheric transport model to compare the observed value and the target The carbon emission flux is processed to obtain the second derivative of carbon concentration; based on the second derivative, the target carbon emission flux is updated.

In the above embodiments of the present application, after obtaining the observed value of carbon concentration and the initial carbon emission flux, the method further includes: outputting the observed value and the initial carbon emission flux; receiving second feedback information, wherein the second feedback information is passed through Obtained by modifying the observed value and the initial carbon emission flux; using the atmospheric transport model to process the second feedback information to obtain the third derivative of carbon concentration; based on the third derivative, the initial carbon emission flux is updated to obtain the target carbon emission flux.

It should be noted that the preferred implementations involved in the above examples of the present application are the same as those provided in Example 1, as well as the application scenarios and implementation process, but are not limited to the solutions provided in Example 1.

Example 3

According to an embodiment of the present application, an embodiment of a method for processing air pollutant emission flux is also provided. It should be noted that the steps shown in the flowcharts of the accompanying drawings can be implemented in a computer system such as a set of computer-executable instructions and, although a logical order is shown in the flowcharts, in some cases the steps shown or described may be performed in an order different from that shown or described herein.

Fig. 5 is a flow chart of a method for processing air pollutant emission flux according to an embodiment of the present invention. As shown in Fig. 5, the method may include the following steps:

Step S502, displaying the observed value of the air pollutant concentration and the initial emission flux of the air pollutant concentration in the interactive interface.

Among them, the observed value is used to represent the value obtained by measuring the concentration of the air pollutant concentration at the target station.

Step S504, in response to the touch operation detected in the interactive interface, use the atmospheric transport model to process the observed value and the initial emission flux to obtain the target derivative of the atmospheric pollutant concentration.

Step S506, displaying the target emission flux of air pollutant concentration in the interactive interface.

Among them, the target emission flux is obtained by updating the initial emission flux by the target derivative.

In the above-mentioned embodiments of the present application, the observed value includes: the concentration of the air pollutant concentration collected within a preset time period, and the observed value and the initial emission flux are processed using the atmospheric transport model to obtain the target derivative of the air pollutant concentration including : The atmospheric transport model is used to deduce the initial observation value and the initial emission flux in reverse to obtain the predicted value of the concentration of air pollutants, where the initial observation value is used to represent the air pollutants collected at the beginning of the preset time period The concentration of the concentration, the predicted value is used to represent the concentration of the atmospheric pollutant concentration deduced within the preset time period; based on the predicted value and the observed value, the observation error of the concentration of the atmospheric pollutant is determined; the derivative of the observation error relative to the emission flux is obtained , to get the target derivative.

In the above-mentioned embodiments of the present application, after updating the initial emission flux based on the target derivative to obtain the target emission flux of the atmospheric pollutant concentration, the method further includes: using the atmospheric transport model to perform an initial observation value and the target emission flux processing to obtain the target predicted value of the air pollutant concentration; based on the target predicted value and the target emission flux, determine whether the target emission flux meets the preset condition; in response to the target emission flux meeting the preset condition, based on the target predicted value, determine The first derivative of the air pollutant concentration; the target emission flux is updated based on the first derivative.

In the above embodiments of the present application, after displaying the target emission flux of atmospheric pollutant concentration in the interactive interface, the method further includes: receiving first feedback information corresponding to the target emission flux, wherein the first feedback information is used to represent whether Update the target emission flux; respond to the first feedback information to update the target emission flux, use the atmospheric transfer model to process the observed value and the target emission flux, and obtain the second derivative of the concentration of atmospheric pollutants; based on the first The second derivative updates the target emission flux.

In the above-mentioned embodiment of the present application, the observed value of the air pollutant concentration and the initial emission flux of the air pollutant concentration are displayed in the interactive interface, and the method also includes: outputting the observed value and the initial emission flux; receiving second feedback information, Among them, the second feedback information is obtained by modifying the observed value and the initial emission flux; the second feedback information is processed by the atmospheric transport model to obtain the third derivative of the concentration of atmospheric pollutants; the initial emission flux is calculated based on the third derivative Update to get the target emission flux.

Example 4

Fig. 6 is a flowchart of a method for processing air pollutant emission flux according to an embodiment of the present invention. As shown in Fig. 6, the method may include the following steps:

Step S602, the cloud server receives the observed value of the air pollutant concentration and the initial emission flux of the air pollutant concentration uploaded by the client;

Among them, the observed value is used to represent the value obtained by measuring the concentration of atmospheric pollutants at the target site;

Step S604, the cloud server uses the atmospheric transport model to process the observed value and the initial emission flux to obtain the target derivative of the atmospheric pollutant concentration;

Among them, the atmospheric transport model is constructed through a deep learning framework, the atmospheric transport model is used to characterize the kinetic transport process of the concentration of atmospheric pollutants in the atmosphere, and the target derivative is used to characterize the derivative of the observed value relative to the emission flux;

Step S606, the cloud server updates the initial emission flux based on the target derivative to obtain the target emission flux of the air pollutant concentration;

Step S608, the cloud server returns the target emission flux to the client.

In the above-mentioned embodiments of the present application, the observed value includes: the concentration of the air pollutant concentration collected within a preset time period, and the cloud server uses the atmospheric transfer model to process the observed value and the initial emission flux to obtain the target concentration of the air pollutant Derivatives include: the cloud server uses the atmospheric transport model to reversely deduce the initial observation value and initial emission flux to obtain the predicted value of the concentration of atmospheric pollutants. The concentration of the air pollutant concentration, the predicted value is used to represent the concentration of the air pollutant concentration deduced within the preset time period; the cloud server determines the observation error of the air pollutant concentration based on the predicted value and the observed value; the cloud server obtains the observed The derivative of the error with respect to the emission flux yields the target derivative.

In the above-mentioned embodiments of the present application, the cloud server uses the atmospheric transport model to reversely deduce the initial observation value and the initial emission flux, and obtains the predicted value of the concentration of atmospheric pollutants, including: the cloud server uses the derivative calculation layer to calculate the density of the concentration of atmospheric pollutants , the velocity field of the atmosphere and the initial observation value, and obtain the spatial derivative; the cloud server uses the differential equation construction layer to construct the equation for the spatial derivative and the initial emission flux, and obtains the target differential equation, which has nothing to do with the time function; The server uses the time integration layer to perform an integral operation on the target differential equation to obtain a predicted value.

In the above embodiments of the present application, after the cloud server updates the initial emission flux based on the target derivative to obtain the target emission flux of the atmospheric pollutant concentration, the method further includes: the cloud server uses the atmospheric transfer model to update the initial observation value and the target emission flux. The emission flux is processed to obtain the target prediction value of the atmospheric pollutant concentration; the cloud server determines whether the target emission flux meets the preset condition based on the target prediction value and the target emission flux; in response to the target emission flux meeting the preset condition, The cloud server determines the first derivative of the atmospheric pollutant concentration based on the target prediction value, wherein the first derivative is used to represent the derivative of the observation error relative to the target emission; the cloud server updates the target emission flux based on the first derivative.

In the above embodiments of the present application, the cloud server determines whether the target emission flux satisfies the preset condition based on the target predicted value and the target emission flux includes: the cloud server constructs a first error function based on the target predicted value and the observed value; the cloud service The controller constructs a second error function based on the target emission flux and the initial emission flux; the cloud server obtains the weighted sum of the first error function and the second error function to obtain the loss function of the concentration of atmospheric pollutants; in response to the loss function being greater than the preset threshold, the cloud server determines that the target emission flux satisfies the preset condition; in response to the loss function being smaller than the preset threshold, the cloud server determines that the target emission flux does not satisfy the preset condition.

In the above embodiments of the present application, after the cloud server updates the initial emission flux based on the target derivative to obtain the target emission flux of the air pollutant concentration, the method further includes: the cloud server outputs the target emission flux; receives the target emission flux The first feedback information corresponding to the amount, wherein the first feedback information is used to represent whether to update the target emission flux; in response to the first feedback information is to update the target emission flux, the cloud server uses the atmospheric transfer model to update the observed value and the target emission flux to obtain the second derivative of the concentration of air pollutants; the cloud server updates the target emission flux based on the second derivative.

In the above embodiments of the present application, after the cloud server obtains the observed value of the air pollutant concentration and the initial emission flux of the air pollutant concentration, the method further includes: the cloud server outputs the observed value and the initial emission flux; the cloud server receives the first Two feedback information, wherein, the second feedback information is obtained by modifying the observed value and the initial emission flux; the cloud server uses the atmospheric transport model to process the second feedback information to obtain the third derivative of the concentration of atmospheric pollutants; the cloud server is based on The third derivative updates the initial emission flux to obtain the target emission flux.

Example 5

Fig. 7 is a flowchart of a method for processing air pollutant emission flux according to an embodiment of the present invention. As shown in Fig. 7, the method may include the following steps:

Step S702, the cloud server receives the air pollutant emission flux processing request uploaded by the client;

Wherein, the air pollutant discharge flux processing request at least includes: a preset time period and air pollutant concentration.

Step S704, the cloud server obtains the observed value of the air pollutant concentration and the initial emission flux of the air pollutant concentration within a preset time period;

Step S706, the cloud server uses the atmospheric transport model to process the observed value and the initial emission flux to obtain the target derivative of the atmospheric pollutant concentration;

Step S708, the cloud server updates the initial emission flux based on the target derivative to obtain the target emission flux of the air pollutant concentration;

Step S710, the cloud server returns the target emission flux to the client.

In the above-mentioned embodiments of the present application, the observed value includes: the concentration of the air pollutant concentration collected within a preset time period, and the cloud server uses the atmospheric transfer model to process the observed value and the initial emission flux to obtain the target concentration of the air pollutant Derivatives include: the cloud server uses the atmospheric transport model to perform reverse deduction on the initial observations and initial emission fluxes, and obtains The predicted value of the air pollutant concentration, where the initial observation value is used to represent the concentration of the air pollutant concentration collected at the beginning of the preset time period, and the predicted value is used to represent the air pollution deduced within the preset time period The concentration of the pollutant concentration; the cloud server determines the observation error of the atmospheric pollutant concentration based on the predicted value and the observed value; obtains the derivative of the observation error relative to the emission flux, and obtains the target derivative.

In the above embodiments of the present application, after the cloud server updates the initial emission flux based on the target derivative to obtain the target emission flux of the atmospheric pollutant concentration, the method further includes: the cloud server uses the atmospheric transfer model to update the initial observation value and the target emission flux. The emission flux is processed to obtain the target prediction value of the atmospheric pollutant concentration; the cloud server determines whether the target emission flux meets the preset condition based on the target prediction value and the target emission flux; in response to the target emission flux meeting the preset condition, The cloud server determines the first derivative of the atmospheric pollutant concentration based on the target prediction value, wherein the first derivative is used to characterize the derivative of the observation error relative to the target emission; based on the first derivative, the target emission flux is updated.

In the above embodiments of the present application, the cloud server determines whether the target emission flux meets the preset condition based on the target predicted value and the target emission flux includes: the cloud server constructs a first error function based on the target predicted value and the observed value; The target emission flux and the initial emission flux construct a second error function; the cloud server obtains the weighted sum of the first error function and the second error function to obtain a loss function of atmospheric pollutant concentration; in response to the loss function being greater than a preset threshold, The cloud server determines that the target emission flux satisfies the preset condition; in response to the loss function being smaller than the preset threshold, the cloud server determines that the target emission flux does not satisfy the preset condition.

In the above embodiments of the present application, after the cloud server updates the initial emission flux based on the target derivative to obtain the target emission flux of atmospheric pollutant concentration, the method further includes: the cloud server outputs the target emission flux; the cloud server receives the target emission flux; The first feedback information corresponding to the emission flux, wherein the first feedback information is used to represent whether to update the target emission flux; in response to the first feedback information is to update the target emission flux, the cloud server uses the atmospheric transport model to The observed value and the target emission flux are processed to obtain the second derivative of the concentration of atmospheric pollutants; the cloud server updates the target emission flux based on the second derivative.

Example 6

According to an embodiment of the present application, there is also provided an atmospheric air pollutant for implementing the above-mentioned air pollutant emission flux processing method. The pollutant discharge flux processing device, as shown in FIG. 8 , the device 800 includes: an acquisition module 802 , a processing module 804 , and an update module 806 .

Among them, the acquisition module is used to obtain the observed value of the air pollutant concentration and the initial emission flux of the air pollutant concentration, wherein the observed value is used to represent the concentration of the air pollutant concentration collected by the target site; the processing module is used to use The atmospheric transport model processes the observed values and the initial emission flux to obtain the target derivative of the atmospheric pollutant concentration. The atmospheric transport model is constructed through a deep learning framework, and the atmospheric transport model is used to characterize the dynamics of the atmospheric pollutant concentration in the atmosphere In the transmission process, the target derivative is used to represent the derivative of the observed value relative to the emission flux; the update module is used to update the initial emission flux based on the target derivative to obtain the target emission flux of the concentration of atmospheric pollutants.

It should be noted here that the acquisition module 802, processing module 804, and update module 806 correspond to steps S202 to S206 in Embodiment 1. The examples and application scenarios implemented by the three modules are the same as those of the corresponding steps, but not It is limited to the content disclosed in the above-mentioned embodiment 1. It should be noted that, as a part of the device, the above modules can run in the computer terminal 10 provided in Embodiment 1.

In the above embodiments of the present application, the observed value includes: the concentration of air pollutants collected within a preset time period, and the processing module includes: a deduction unit, a first determination unit, and an acquisition unit.

Among them, the deduction unit is used to reversely deduce the initial observation value and the initial emission flux by using the atmospheric transport model to obtain the predicted value of the concentration of atmospheric pollutants. The concentration of the air pollutant concentration obtained, the predicted value is used to represent the concentration of the air pollutant concentration deduced within the preset time period; the first determination unit is used to determine the observation error of the air pollutant concentration based on the predicted value and the observed value ; The acquisition unit is used to obtain the derivative of the observation error relative to the emission flux, and obtain the target derivative.

In the above embodiments of the present application, the derivation unit includes: a differential operation subunit, a construction subunit, and an integral operation subunit.

Among them, the differential operation subunit is used to use the derivative calculation layer to perform differential operations on the density of atmospheric pollutant concentration, the velocity field of the atmosphere and the initial observation value to obtain the spatial derivative; the construction subunit is used to use the differential equation to construct the layer pair spatial derivative and The equation is constructed for the initial emission flux to obtain the target differential equation, which has nothing to do with the time function; the integral operation subunit is used to perform an integral operation on the target differential equation by using the time integral layer to obtain the predicted value.

In the above embodiments of the present application, the device further includes: a determination module.

Among them, the processing module is also used to use the atmospheric transport model to process the initial observation value and the target emission flux to obtain the target predicted value of the concentration of atmospheric pollutants; the determination module is used to determine the target emission based on the target predicted value and the target emission flux Whether the flux meets the preset condition; the determination module is also used to determine the first derivative of the atmospheric pollutant concentration based on the target predicted value in response to the target emission flux meeting the preset condition, wherein the first derivative is used to characterize the relative observation error Based on the derivative of the target emission; the update module is also used to update the target emission flux based on the first derivative.

In the above embodiments of the present application, the determination module includes: a construction unit, a weighting unit, and a second determination unit.

Among them, the construction unit is used to construct the first error function based on the target predicted value and the observed value; the construction unit is also used to construct the second error function based on the target emission flux and the initial emission flux; the weighting unit is used to obtain the first error function function and the weighted sum of the second error function to obtain the loss function of the atmospheric pollutant concentration; the second determination unit is used to determine that the target emission flux meets the preset condition in response to the loss function being greater than the preset threshold; the second determination unit also uses In response to the loss function being smaller than a preset threshold, it is determined that the target emission flux does not satisfy a preset condition.

In the above embodiments of the present application, the device further includes: an output module and a receiving module.

Wherein, the output module is used to output the target emission flux; the receiving module is used to receive the first feedback information corresponding to the target emission flux, wherein the first feedback information is used to indicate whether to update the target emission flux; the update module is used to In order to update the target emission flux in response to the first feedback information, use the atmospheric transport model to process the observed value and the target emission flux to obtain the second derivative of the concentration of atmospheric pollutants; the update module is also used to update the target emission flux based on the second derivative The target emission flux is updated.

In the above embodiments of the present application, the output module is also used to output the observed value and the initial emission flux; the receiving module is also used to receive the second feedback information, wherein the second feedback information is obtained by modifying the observed value and the initial emission flux ; The processing module is also used to process the second feedback information using the atmospheric transport model to obtain the third derivative of the concentration of air pollutants; the update module is also used to update the initial emission flux based on the third derivative to obtain the target emission flux .

Example 7

According to an embodiment of the present application, an air pollutant emission flux processing device for implementing the above air pollutant emission flux processing method is also provided, as shown in FIG. 9 , the device includes: an acquisition module 902 and a processing module 904 , Updating the module 906.

Among them, the acquisition module is used to obtain the observed value of carbon concentration and the initial carbon emission flux in the carbon neutral scenario, wherein the observed value is used to represent the value obtained by measuring the carbon concentration at the target site; the processing module is used to use the atmospheric transport model The observed value and the initial carbon emission flux are processed to obtain the target derivative. The atmospheric transport model is constructed through a deep learning framework. The atmospheric transport model is used to characterize the dynamic transport process of gases affecting carbon concentration in the atmosphere. The target derivative is used is used to represent the derivative of the observed value relative to the carbon emission flux; the update module is used to update the initial carbon emission flux based on the target derivative to obtain the target carbon emission flux.

It should be noted here that the acquisition module 902, processing module 904, and update module 906 correspond to steps S402 to S406 in Embodiment 2, and the examples and application scenarios implemented by the three modules are the same as those of the corresponding steps, but not It is limited to the content disclosed in the above-mentioned embodiment 2. It should be noted that, as a part of the device, the above modules can run in the computer terminal 10 provided in Embodiment 1.

Example 8

According to an embodiment of the present application, an air pollutant emission flux processing device for implementing the above air pollutant emission flux processing method is also provided, as shown in FIG. 10 , the device includes: a first display module 1002, a processing Module 1004, a second display module 1006.

Among them, the first display module is used to display the observed value of the concentration of air pollutants and the initial emission flux of the concentration of air pollutants in the interactive interface, wherein the observed value is used to represent the concentration of the concentration of air pollutants measured by the target station. The value of ; the processing module is used to respond to the touch operation detected in the interactive interface, and use the atmospheric transfer model to process the observed value and the initial emission flux to obtain the target derivative of the concentration of atmospheric pollutants, wherein the atmospheric transfer model is passed through The deep learning framework is constructed, the atmospheric transport model is used to characterize the dynamic transport process of atmospheric pollutant concentration in the atmosphere, and the target derivative is used to characterize the derivative of the observed value relative to the emission flux; the second display module is used to display in the interactive interface The target emission flux of the air pollutant concentration, wherein the target emission flux is obtained by updating the initial emission flux through the target derivative.

It should be noted here that the above-mentioned first display module 1002, processing module 1004, and second display module 1006 correspond to steps S502 to S506 in Embodiment 3, examples and application scenarios realized by the three modules and corresponding steps The same, but not limited to the content disclosed in Embodiment 3 above. It should be noted that, as a part of the device, the above modules can run in the computer terminal 10 provided in Embodiment 1.

Example 9

According to an embodiment of the present application, an air pollutant emission flux processing device for implementing the above air pollutant emission flux processing method is also provided, as shown in FIG. 11 , the device includes: a receiving module 1102 and a processing module 1104 , an update module 1106 , and an acquisition module 1108 .

Among them, the receiving module is used to receive the observation value of air pollutant concentration and the initial emission flux of air pollutant concentration uploaded by the client, wherein the observation value is used to represent the value obtained by measuring the concentration of air pollutant concentration at the target site ; The processing module is used to use the atmospheric transport model to process the observed value and the initial emission flux to obtain the target derivative of the atmospheric pollutant concentration, wherein the atmospheric transport model is constructed through a deep learning framework, and the atmospheric transport model is used to represent the atmospheric pollutant concentration In the dynamic transmission process in the atmosphere, the target derivative is used to represent the derivative of the observed value relative to the emission flux; the update module is used to update the initial emission flux based on the target derivative to obtain the target emission flux of the concentration of atmospheric pollutants; The acquisition module is used to return the target emission flux to the client.

It should be noted here that the above receiving module 1102, processing module 1104, updating module 1106, and obtaining module 1108 correspond to steps S602 to S608 in Embodiment 4, examples and application scenarios realized by the four modules and corresponding steps The same, but not limited to the content disclosed in Embodiment 4 above. It should be noted that, as a part of the device, the above modules can run in the computer terminal 10 provided in Embodiment 1.

Example 10

According to an embodiment of the present application, an air pollutant emission flux processing device for implementing the above air pollutant emission flux processing method is also provided, as shown in FIG. 12 , the device includes: a receiving module 1202 and an acquisition module 1204 , a processing module 1206 , an updating module 1208 , and a feedback module 1210 .

Wherein, the receiving module is used to receive the air pollutant emission flux processing request uploaded by the client, wherein the air pollutant emission flux processing request at least includes: a preset time period and the concentration of air pollutants; the obtaining module is used to obtain the preset The observed value of the concentration of air pollutants and the initial emission flux of the concentration of air pollutants in the time period, where the observed value is used to represent the value obtained by measuring the concentration of the air pollutant concentration at the target site; the processing module is used to use the atmospheric transmission The model processes the observed value and the initial emission flux to obtain the target derivative of the concentration of atmospheric pollutants. Among them, the atmospheric transport model is constructed through a deep learning framework, and the atmospheric transport model is used to characterize the dynamics of the concentration of atmospheric pollutants in the atmosphere In the transmission process, the target derivative is used to represent the derivative of the observed value relative to the emission flux; the update module is used to update the initial emission flux based on the target derivative to obtain the target emission flux of the concentration of atmospheric pollutants; the feedback module is used to return the target Emit flux to the client.

It should be noted here that the receiving module 1202, acquiring module 1204, processing module 1206, updating module 1208, and feedback module 1210 correspond to steps S702 to S710 in Embodiment 5, and the five modules and corresponding steps realize The examples and application scenarios are the same, but are not limited to the content disclosed in Embodiment 4 above. It should be noted that, as a part of the device, the above modules can run in the computer terminal 10 provided in Embodiment 1.

Example 11

Embodiments of the present invention may provide a computer terminal, and the computer terminal may be any computer terminal device in a group of computer terminals. Optionally, in this embodiment, the foregoing computer terminal may also be replaced with a terminal device such as a mobile terminal.

Optionally, in this embodiment, the foregoing computer terminal may be located in at least one network device among multiple network devices of the computer network.

In this embodiment, the above-mentioned computer terminal can execute the program code of the following steps in the air pollutant emission flux processing method: obtain the observed value of the atmospheric pollutant concentration and the initial emission flux of the atmospheric pollutant concentration, wherein the observed value is used To characterize the concentration of atmospheric pollutants collected at the target site; use the atmospheric transport model to process the observed value and initial emission flux to obtain the target derivative of the atmospheric pollutant concentration, wherein the atmospheric transport model is constructed through a deep learning framework, and the atmospheric transport model The transport model is used to characterize the dynamic transport process of air pollutant concentration in the atmosphere, and the target derivative is used to characterize the derivative of the observed value relative to the emission flux; based on the target derivative, the initial emission flux is updated to obtain the air pollutant concentration Target emission flux.

Optionally, FIG. 13 is a structural block diagram of a computer terminal according to an embodiment of the present invention. As shown in FIG. 13 , the computer terminal A may include: one or more (only one is shown in the figure) processors and memory.

Wherein, the memory can be used to store software programs and modules, such as program instructions/modules corresponding to the air pollutant emission flux processing method and device in the embodiment of the present invention, and the processor runs the software programs and modules stored in the memory, thereby Execute various functional applications and data processing, that is, realize the above-mentioned air pollutant emission flux processing method. The memory may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory may further include a memory remotely located relative to the processor, and these remote memories may be connected to the terminal A through a network. Examples of the aforementioned networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The processor can call the information stored in the memory and the application program through the transmission device to perform the following steps: Obtain the observed value of the concentration of atmospheric pollutants and the initial emission flux of the concentration of atmospheric pollutants, wherein the observed values are used to represent the collection of the target site The concentration of the atmospheric pollutant concentration obtained; the atmospheric transport model is used to process the observed value and the initial emission flux to obtain the target derivative of the atmospheric pollutant concentration. The atmospheric transport model is constructed through a deep learning framework, and the atmospheric transport model is used to represent The dynamic transmission process of air pollutant concentration in the atmosphere, the target derivative is used to characterize the derivative of the observed value relative to the emission flux; based on the target derivative, the initial emission flux is updated to obtain the target emission flux of the air pollutant concentration.

Optionally, the above-mentioned processor can also execute the program code of the following steps: use the atmospheric transport model to reversely deduce the initial observation value and the initial emission flux to obtain the predicted value of the concentration of atmospheric pollutants, wherein the initial observation value is used for Characterize the concentration of the air pollutant concentration collected at the beginning of the preset time period, and the predicted value is used to represent the concentration of the air pollutant concentration deduced within the preset time period; based on the predicted value and the observed value, determine the concentration of the air pollutant The observation error of the concentration; obtain the derivative of the observation error relative to the emission flux, and obtain the target derivative.

Optionally, the above-mentioned processor can also execute the program code of the following steps: the atmospheric transport model includes: a derivative calculation layer, a differential equation construction layer and a time integration layer.

Optionally, the above-mentioned processor can also execute the program code of the following steps: use the derivative calculation layer to perform differential operations on the density of atmospheric pollutant concentration, the velocity field of the atmosphere, and the initial observation value to obtain spatial derivatives; use differential equations to construct layer pairs The spatial derivative and the initial emission flux are used to construct the equation to obtain the target differential equation, which has nothing to do with the time function; the time integral layer is used to integrate the target differential equation to obtain the predicted value.

Optionally, the above-mentioned processor can also execute the program code of the following steps: use the atmospheric transport model to process the initial observation value and the target emission flux to obtain the target predicted value of the concentration of atmospheric pollutants; Determine whether the target emission flux meets the preset condition; in response to the target emission flux meeting the preset condition, based on the target predicted value, determine the first derivative of the concentration of atmospheric pollutants, where the first derivative is used to characterize the relative observation error Based on the derivative of the target emission; based on the first derivative, the target emission flux is updated.

Optionally, the above-mentioned processor can also execute the program code of the following steps: constructing a first error function based on the target predicted value and observed value; constructing a second error function based on the target emission flux and the initial emission flux; obtaining the first The weighted sum of the error function and the second error function is used to obtain the loss function of the concentration of atmospheric pollutants; in response to the loss function being greater than the preset threshold, it is determined that the target emission flux meets the preset condition; in response to the loss function being less than the preset threshold, it is determined that the target The emission flux does not meet the preset conditions.

Optionally, the above-mentioned processor may also execute the program code for the following steps: outputting the target emission flux; receiving first feedback information corresponding to the target emission flux, wherein the first feedback information is used to represent whether the target emission flux is Update; In order to update the target emission flux in response to the first feedback information, the atmospheric transport model is used to process the observed value and the target emission flux to obtain the second derivative of the concentration of atmospheric pollutants; based on the second derivative, the target emission flux volume is updated.

Optionally, the above-mentioned processor may also execute the program code of the following steps: outputting the observed value and the initial emission flux; receiving second feedback information, wherein the second feedback information is obtained by modifying the observed value and the initial emission flux; The atmospheric transport model is used to process the second feedback information to obtain the third derivative of the concentration of air pollutants; based on the third derivative, the initial emission flux is updated to obtain the target emission flux.

The processor can call the information stored in the memory and the application program through the transmission device to perform the following steps: Obtain the observed value of carbon concentration and the initial carbon emission flux in the carbon neutral scenario, wherein the observed value is used to represent the impact of the target site on carbon The value obtained by measuring the concentration; use the atmospheric transport model to process the observed value and the initial carbon emission flux to obtain the target derivative. In the dynamic transport process in the atmosphere, the target derivative is used to represent the derivative of the observed value relative to the carbon emission flux; based on the target derivative, the initial carbon emission flux is updated to obtain the target carbon emission flux.

The processor can call the information stored in the memory and the application program through the transmission device to perform the following steps: display the observed value of the concentration of atmospheric pollutants and the initial emission flux of the concentration of atmospheric pollutants in the interactive interface, wherein the observed values are used for Characterize the value obtained by measuring the concentration of atmospheric pollutants at the target site; respond to the detection in the interactive interface The touch operation obtained by using the atmospheric transfer model is used to process the observed value and the initial emission flux to obtain the target derivative of the atmospheric pollutant concentration. The atmospheric transfer model is constructed through a deep learning framework, and the atmospheric transfer model is used to represent the atmospheric pollutants The kinetic transmission process of concentration in the atmosphere, the target derivative is used to characterize the derivative of the observed value relative to the emission flux; the target emission flux of the concentration of atmospheric pollutants is displayed in the interactive interface, where the target emission flux passes the target derivative to The initial emission flux is updated to get.

The processor can call the information stored in the memory and the application program through the transmission device to perform the following steps: the cloud server receives the observed value of the concentration of atmospheric pollutants and the initial emission flux of the concentration of atmospheric pollutants uploaded by the client, wherein the observed value It is used to represent the value obtained by measuring the concentration of atmospheric pollutants at the target site; the cloud server uses the atmospheric transport model to process the observed value and the initial emission flux to obtain the target derivative of the atmospheric pollutant concentration, where the atmospheric transport model passes The deep learning framework is constructed. The atmospheric transport model is used to characterize the dynamic transport process of atmospheric pollutant concentration in the atmosphere. The target derivative is used to characterize the derivative of the observed value relative to the emission flux; the cloud server calculates the initial emission flux based on the target derivative. Update to obtain the target emission flux of air pollutant concentration; the cloud server returns the target emission flux to the client.

The processor can call the information and application programs stored in the memory through the transmission device to perform the following steps: the cloud server receives the air pollutant emission flux processing request uploaded by the client, wherein the air pollutant emission flux processing request at least includes: Preset time period and air pollutant concentration; the cloud server obtains the observed value of the air pollutant concentration and the initial emission flux of the air pollutant concentration within the preset time period, where the observed value is used to represent the impact of the target site on the concentration of air pollutants The value obtained by measuring the concentration; the cloud server uses the atmospheric transport model to process the observed value and the initial emission flux to obtain the target derivative of the atmospheric pollutant concentration. The atmospheric transport model is constructed through a deep learning framework, and the atmospheric transport model is used for To characterize the dynamic transmission process of air pollutant concentration in the atmosphere, the target derivative is used to represent the derivative of the observed value relative to the emission flux; the cloud server updates the initial emission flux based on the target derivative to obtain the target emission of the air pollutant concentration flux; the cloud server returns the target emission flux to the client.

Using the embodiment of the present invention, first, obtain the observed value of the air pollutant concentration and the initial emission flux of the air pollutant concentration, wherein the observed value is used to represent the concentration of the air pollutant concentration collected by the target site; using the atmospheric transport model The observed value and the initial emission flux are processed to obtain the target derivative of the atmospheric pollutant concentration. The atmospheric transport model is constructed through a deep learning framework, and the atmospheric transport model is used to characterize the dynamic transport process of the atmospheric pollutant concentration in the atmosphere. The target derivative is used to represent the derivative of the observed value relative to the emission flux; based on the target derivative, the initial emission flux is updated to obtain the target emission flux of the concentration of atmospheric pollutants, which realizes the purpose of improving the monitoring of the target emission flux. It is easy to notice that the observed value of atmospheric pollutant concentration and the initial emission flux of atmospheric pollutant concentration can be obtained first, and then the observed value and initial emission flux can be compared according to the power transfer model of atmospheric pollutant concentration in the atmosphere of the atmospheric transfer model In order to determine the influence factors of environmental factors in the atmosphere on the prediction accuracy, and determine the observation error according to the influence factors, update the initial emission flux according to the target derivative determined by the observation error, so as to improve the initial emission flux The accuracy of the flux, and thus solve the technical problem of low accuracy in monitoring the concentration of air pollutants in related technologies.

Those of ordinary skill in the art can understand that the structure shown in Figure 13 is only schematic, and the computer terminal can also be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, an applause computer, and a mobile Internet device (Mobile Internet Devices, MID ), PAD and other terminal equipment. FIG. 13 does not limit the structure of the above-mentioned electronic device. For example, the computer terminal 10 may also include more or fewer components (eg, network interface, display device, etc.) than those shown in FIG. 13 , or have a configuration different from that shown in FIG. 13 .

Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing hardware related to the terminal device through a program, and the program can be stored in a computer-readable storage medium, and the storage medium can be Including: flash disk, read-only memory (Read-Only Memory, ROM), random access device (Random Access Memory, RAM), magnetic disk or optical disk, etc.

Example 4

The embodiment of the invention also provides a storage medium. Optionally, in this embodiment, the above-mentioned storage medium may be used to store the program code executed by the method for processing air pollutant emission flux provided in the first embodiment above.

Optionally, in this embodiment, the above-mentioned storage medium may be located in any computer terminal in the group of computer terminals in the computer network, or in any mobile terminal in the group of mobile terminals.

Optionally, in this embodiment, the storage medium is configured to store program codes for performing the following steps: Obtaining the observed value of the atmospheric pollutant concentration and the initial emission flux of the atmospheric pollutant concentration, wherein the observed value is used for Characterize the concentration of atmospheric pollutants collected at the target site; use the atmospheric transport model to process the observations and initial emission flux to obtain the target derivative of the atmospheric pollutant concentration, where the atmospheric transport model is constructed through a deep learning framework, and the atmospheric transport The model is used to characterize the dynamic transmission process of atmospheric pollutant concentration in the atmosphere, and the target derivative is used to characterize the derivative of the observed value relative to the emission flux; based on the target derivative, the initial emission flux is updated to obtain the target concentration of atmospheric pollutants emission flux.

Optionally, the above-mentioned storage medium is also configured to store program codes for performing the following steps: use the atmospheric transport model to reversely deduce the initial observation value and the initial emission flux, and obtain the predicted value of the concentration of atmospheric pollutants, wherein, The initial observation value is used to represent the concentration of air pollutants collected at the beginning of the preset time period, and the predicted value is used to represent the concentration of air pollutants deduced within the preset time period; based on the predicted value and the observed value , to determine the observation error of the atmospheric pollutant concentration; obtain the derivative of the observation error relative to the emission flux, and obtain the target derivative.

Optionally, the above-mentioned storage medium is also configured to store program codes for executing the following steps: the atmospheric transport model includes: a derivative calculation layer, a differential equation construction layer and a time integration layer.

Optionally, the above-mentioned storage medium is also configured to store program codes for performing the following steps: use the derivative calculation layer to perform differential operations on the density of atmospheric pollutant concentrations, the velocity field of the atmosphere, and the initial observation value to obtain spatial derivatives; use The differential equation construction layer constructs the equations for the spatial derivative and the initial emission flux to obtain the target differential equation, which has nothing to do with the time function; the time integration layer is used to integrate the target differential equation to obtain the predicted value.

Optionally, the above-mentioned storage medium is also configured to store program codes for performing the following steps: using the atmospheric transport model to process the initial observed value and the target emission flux to obtain the target predicted value of the concentration of atmospheric pollutants; based on the target prediction value and the target emission flux to determine whether the target emission flux meets the preset condition; in response to the target emission flux meeting the preset condition, based on the target predicted value, determine the first derivative of the atmospheric pollutant concentration; based on the first derivative to the target Emission fluxes are updated.

Optionally, the above-mentioned storage medium is also configured to store program codes for performing the following steps: constructing a first error function based on the target predicted value and observed value; constructing a second error function based on the target emission flux and the initial emission flux function; obtain the weighted sum of the first error function and the second error function, and obtain the loss function of the atmospheric pollutant concentration; in response to the loss function being greater than the preset threshold, determine that the target emission flux meets the preset condition; in response to the loss function being less than the preset A threshold is set to determine that the target emission flux does not meet the preset conditions.

Optionally, the above-mentioned storage medium is also configured to store program codes for performing the following steps: outputting the target emission flux; receiving first feedback information corresponding to the target emission flux, wherein the first feedback information is used to represent whether the target emission flux is Target The emission flux is updated; the target emission flux is updated in response to the first feedback information, and the observed value and the target emission flux are processed using the atmospheric transfer model to obtain the second derivative of the concentration of atmospheric pollutants; based on the second derivative Update the target emission flux.

Optionally, the above-mentioned storage medium is also configured to store program codes for performing the following steps: outputting the observed value and the initial emission flux; receiving second feedback information, wherein the second feedback information passes the observation value and the initial emission flux The second feedback information is processed by the atmospheric transport model to obtain the third derivative of the concentration of air pollutants; the initial emission flux is updated based on the third derivative to obtain the target emission flux.

Optionally, in this embodiment, the storage medium is configured to store program codes for performing the following steps: Obtaining the observed value of carbon concentration and the initial carbon emission flux in a carbon neutral scenario, wherein the observed value is used to represent The value obtained by measuring the carbon concentration at the target site; the observed value and the initial carbon emission flux are processed by the atmospheric transfer model to obtain the target derivative. The dynamic transmission process of the gas with high concentration in the atmosphere, the target derivative is used to characterize the derivative of the observed value relative to the carbon emission flux; based on the target derivative, the initial carbon emission flux is updated to obtain the target carbon emission flux.

Optionally, in this embodiment, the storage medium is configured to store program codes for performing the following steps: displaying the observed value of the concentration of atmospheric pollutants and the initial emission flux of the concentration of atmospheric pollutants in an interactive interface, wherein, The observed value is used to represent the value obtained by measuring the concentration of air pollutants at the target site; in response to the touch operation detected in the interactive interface, the observed value and the initial emission flux are processed using the atmospheric transfer model to obtain the air pollution The target derivative of the pollutant concentration, in which the atmospheric transport model is constructed through a deep learning framework, the atmospheric transport model is used to characterize the kinetic transport process of the atmospheric pollutant concentration in the atmosphere, and the target derivative is used to characterize the derivative of the observed value relative to the emission flux ; Display the target emission flux of atmospheric pollutant concentration in the interactive interface, where the target emission flux is obtained by updating the initial emission flux through the target derivative.

Optionally, in this embodiment, the storage medium is configured to store program codes for performing the following steps: the cloud server receives the observed value of the concentration of atmospheric pollutants uploaded by the client and the initial emission flux of the concentration of atmospheric pollutants, Among them, the observed value is used to represent the value obtained by measuring the concentration of atmospheric pollutants at the target site; the cloud server uses the atmospheric transfer model to process the observed value and the initial emission flux to obtain the target derivative of the concentration of atmospheric pollutants, where, The atmospheric transport model is constructed through a deep learning framework. The atmospheric transport model is used to characterize the dynamic transport process of atmospheric pollutant concentrations in the atmosphere. The target derivative is used to represent the derivative of the observed value relative to the emission flux; the cloud server based on the target derivative The emission flux is updated to obtain the target emission flux of atmospheric pollutant concentration; the cloud server returns the target emission flux to the client.

Optionally, in this embodiment, the storage medium is configured to store program codes for performing the following steps: the cloud server receives the air pollutant emission flux processing request uploaded by the client, wherein the air pollutant emission flux processing The request at least includes: a preset time period and the concentration of atmospheric pollutants; the cloud server obtains the observed value of the concentration of atmospheric pollutants within the preset time period and the initial emission flux of the concentration of atmospheric pollutants, wherein the observed value is used to represent the The value obtained by measuring the concentration of atmospheric pollutants; the cloud server uses the atmospheric transport model to process the observed value and the initial emission flux to obtain the target derivative of the atmospheric pollutant concentration. The atmospheric transport model is constructed through a deep learning framework, and the atmospheric The transport model is used to characterize the dynamic transport process of atmospheric pollutant concentration in the atmosphere, and the target derivative is used to characterize the derivative of the observed value relative to the emission flux; the cloud server updates the initial emission flux based on the target derivative to obtain the atmospheric pollutant concentration target emission flux; the cloud server returns the target emission flux to the client.

Using the embodiment of the present invention, first, obtain the observed value of the concentration of air pollutants and the initial discharge of the concentration of air pollutants The emission flux, where the observed value is used to characterize the concentration of the air pollutant concentration collected by the target site; the observed value and the initial emission flux are processed using the atmospheric transport model to obtain the target derivative of the atmospheric pollutant concentration, where the atmospheric The transport model is constructed through a deep learning framework. The atmospheric transport model is used to characterize the dynamic transport process of atmospheric pollutant concentration in the atmosphere. The target derivative is used to characterize the derivative of the observed value relative to the emission flux; based on the target derivative, the initial emission flux The update is carried out to obtain the target emission flux of the air pollutant concentration, and the purpose of improving the monitoring of the target emission flux is realized. It is easy to notice that the observed value of atmospheric pollutant concentration and the initial emission flux of atmospheric pollutant concentration can be obtained first, and then the observed value and initial emission flux can be compared according to the power transfer model of atmospheric pollutant concentration in the atmosphere of the atmospheric transfer model In order to determine the influence factors of environmental factors in the atmosphere on the prediction accuracy, and determine the observation error according to the influence factors, update the initial emission flux according to the target derivative determined by the observation error, so as to improve the initial emission flux The accuracy of the flux, and thus solve the technical problem of low accuracy in monitoring the concentration of air pollutants in related technologies.

The serial numbers of the above embodiments of the present invention are for description only, and do not represent the advantages and disadvantages of the embodiments.

In the above-mentioned embodiments of the present invention, the descriptions of each embodiment have their own emphases, and for parts not described in detail in a certain embodiment, reference may be made to relevant descriptions of other embodiments.

In the several embodiments provided in this application, it should be understood that the disclosed technical content can be realized in other ways. Wherein, the device embodiments described above are only illustrative, for example, the division of the units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units or components can be combined or can be Integrate into another system, or some features may be ignored, or not implemented. In another point, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection of units or modules may be in electrical or other forms.

The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

If the integrated unit is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the essence of the technical solution of the present invention or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage media include: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disc, etc., which can store program codes. .

The above is only a preferred embodiment of the present invention, it should be pointed out that, for those of ordinary skill in the art, without departing from the principle of the present invention, some improvements and modifications can also be made, and these improvements and modifications can also be made. It should be regarded as the protection scope of the present invention.

Claims

A method for processing air pollutant emission flux, characterized in that it comprises:

Obtaining the observed value of the air pollutant concentration and the initial discharge flux of the air pollutant concentration, wherein the observed value is used to characterize the concentration of the air pollutant concentration collected by the target site;

The observed value and the initial emission flux are processed by an atmospheric transport model to obtain the target derivative of the atmospheric pollutant concentration, wherein the atmospheric transport model is constructed by a deep learning framework, and the atmospheric transport model is used for Characterize the kinetic transport process of the atmospheric pollutant concentration in the atmosphere, and the target derivative is used to characterize the derivative of the observed value relative to the emission flux;

The initial emission flux is updated based on the target derivative to obtain the target emission flux of the atmospheric pollutant concentration.
The method according to claim 1, wherein the observed value comprises: the concentration of the atmospheric pollutant concentration collected within a preset period of time, and the atmospheric transport model is used to compare the observed value and the initial discharge The flux is processed to obtain the target derivative of the air pollutant concentration including:

The atmospheric transport model is used to reversely deduce the initial observation value and the initial emission flux to obtain the predicted value of the atmospheric pollutant concentration, wherein the initial observation value is used to characterize the The concentration of the atmospheric pollutant concentration collected at the initial moment, and the predicted value is used to represent the concentration of the atmospheric pollutant concentration deduced within the preset time period;

determining an observation error of the atmospheric pollutant concentration based on the predicted value and the observed value;

Obtaining the derivative of the observation error with respect to the emission flux to obtain the target derivative.
The method according to claim 2, wherein the atmospheric transport model comprises: a derivative calculation layer, a differential equation construction layer and a time integration layer.
The method according to claim 3, characterized in that, using the atmospheric transport model to perform reverse deduction on the initial observed value and the initial emission flux, and obtaining the predicted value of the atmospheric pollutant concentration includes:

Using the derivative calculation layer to perform a differential operation on the density of the atmospheric pollutant concentration, the velocity field of the atmosphere, and the initial observation value to obtain a spatial derivative;

Using the differential equation construction layer to perform equation construction on the spatial derivative and the initial discharge flux to obtain a target differential equation, the target differential equation has nothing to do with the time function;

The time integration layer is used to perform an integral operation on the target differential equation to obtain the predicted value.
The method according to claim 2, wherein, after updating the initial emission flux based on the target derivative to obtain the target emission flux of the atmospheric pollutant concentration, the method further comprises:

using the atmospheric transport model to process the initial observed value and the target emission flux to obtain the target predicted value of the atmospheric pollutant concentration;

determining whether the target emission flux satisfies a preset condition based on the target predicted value and the target emission flux;

In response to the target emission flux meeting the preset condition, based on the target predicted value, determining the The first derivative of the concentration of atmospheric pollutants, wherein the first derivative is used to characterize the derivative of the observation error relative to the target emission;

The target emission flux is updated based on the first derivative.
The method according to claim 5, wherein, based on the target predicted value and the target emission flux, determining whether the target emission flux satisfies a preset condition comprises:

constructing a first error function based on the target predicted value and the observed value;

constructing a second error function based on the target emission flux and the initial emission flux;

Obtaining the weighted sum of the first error function and the second error function to obtain the loss function of the atmospheric pollutant concentration;

determining that the target emission flux satisfies the preset condition in response to the loss function being greater than a preset threshold;

In response to the loss function being smaller than the preset threshold, it is determined that the target emission flux does not satisfy the preset condition.
The method according to any one of claims 1 to 6, characterized in that, after updating the initial emission flux based on the target derivative to obtain the target emission flux of the atmospheric pollutant concentration, the The method also includes:

Based on the target emission flux, determine the grid point emission flux of each grid point on the target map;

The gridpoint emission fluxes are displayed on the target map.
The method according to claim 7, wherein after displaying the grid point emission flux on the target map, the method further comprises:

receiving the selected target area on the target map;

Summarizing the grid point emission fluxes of all grid points included in the target area to obtain the regional emission flux corresponding to the target area;

Outputs the area emission flux.
A method for processing air pollutant emission flux, characterized in that it comprises:

Obtain the observed value of carbon concentration and the initial carbon emission flux under the carbon neutral scenario, wherein the observed value is used to represent the value obtained by measuring the carbon concentration at the target site;

The observed value and the initial carbon emission flux are processed by using an atmospheric transport model to obtain a target derivative, wherein the atmospheric transport model is constructed by a deep learning framework, and the atmospheric transport model is used to characterize the impact on the carbon concentration The kinetic transport process of the gas in the atmosphere, the target derivative is used to characterize the derivative of the observed value relative to the carbon emission flux;

The initial carbon emission flux is updated based on the target derivative to obtain a target carbon emission flux.
The method according to claim 9, wherein the observed value includes: the carbon concentration collected within a preset time period, and the observed value and the initial carbon emission flux are carried out using an atmospheric transport model Processing to obtain the target derivative includes:

Using the atmospheric transport model to reverse the initial observations and the initial carbon emission flux, we get The predicted value of the carbon concentration, wherein the initial observed value is used to characterize the carbon concentration collected at the beginning of the preset time period, and the predicted value is used to characterize the preset time period The carbon concentration obtained by internal deduction;

determining an observation error for the carbon concentration based on the predicted value and the observed value;

Obtaining the derivative of the observation error relative to the carbon emission flux to obtain the target derivative.
A method for processing air pollutant emission flux, characterized in that it comprises:

The cloud server receives the observed value of the air pollutant concentration and the initial emission flux of the air pollutant concentration uploaded by the client, wherein the observed value is used to represent the measured value obtained by the target site for the concentration of the air pollutant concentration value;

The cloud server uses an atmospheric transport model to process the observed value and the initial emission flux to obtain the target derivative of the atmospheric pollutant concentration, wherein the atmospheric transport model is constructed by a deep learning framework, and the atmospheric transport model The transport model is used to characterize the kinetic transport process of the atmospheric pollutant concentration in the atmosphere, and the target derivative is used to characterize the derivative of the observed value relative to the emission flux;

The cloud server updates the initial emission flux based on the target derivative to obtain the target emission flux of the atmospheric pollutant concentration;

The cloud server returns the target emission flux to the client.
A method for processing air pollutant emission flux, characterized in that it comprises:

The cloud server receives the air pollutant emission flux processing request uploaded by the client, wherein the air pollutant emission flux processing request includes at least: a preset time period and air pollutant concentration;

The cloud server acquires the observed value of the air pollutant concentration and the initial emission flux of the air pollutant concentration within the preset time period, wherein the observed value is used to represent the target site's response to the air pollution The value obtained by measuring the concentration of the substance concentration;

The cloud server uses an atmospheric transport model to process the observed value and the initial emission flux to obtain the target derivative of the atmospheric pollutant concentration, wherein the atmospheric transport model is constructed by a deep learning framework, and the atmospheric transport model The transport model is used to characterize the kinetic transport process of the atmospheric pollutant concentration in the atmosphere, and the target derivative is used to characterize the derivative of the observed value relative to the emission flux;

The cloud server updates the initial emission flux based on the target derivative to obtain the target emission flux of the atmospheric pollutant concentration;

The cloud server returns the target emission flux to the client.
A storage medium, characterized in that the storage medium includes a stored program, wherein, when the program is running, the device where the storage medium is located is controlled to perform the air pollutant emission described in any one of claims 1 to 12 Flux processing method.
A computer terminal, characterized in that it includes: a memory and a processor, the processor is used to run a program stored in the memory, wherein, when the program runs, it executes the emission channel described in any one of claims 1 to 12 Volume processing method.