CN114357894A

CN114357894A - Atmospheric pollutant processing method, storage medium and computer terminal

Info

Publication number: CN114357894A
Application number: CN202210234949.2A
Authority: CN
Inventors: 陈国栋; 姚易辰; 李展; 仲晓辉; 陈磊; 胡媛; 杜飞; 王志斌; 李�昊
Original assignee: Alibaba Damo Institute Hangzhou Technology Co Ltd
Current assignee: Alibaba Damo Institute Hangzhou Technology Co Ltd
Priority date: 2022-03-11
Filing date: 2022-03-11
Publication date: 2022-04-15
Anticipated expiration: 2042-03-11
Also published as: CN114357894B

Abstract

The invention discloses an atmospheric pollutant processing method, a storage medium and a computer terminal. Wherein, the method comprises the following steps: acquiring target spectral data and meteorological data, wherein the target spectral data is used for representing radiation data of atmospheric pollutants; and processing the target spectrum data and the gas image data by using a target inversion model to obtain the target emission flux of the atmospheric pollutants, wherein the target inversion model comprises a first hidden layer and a second hidden layer, the first hidden layer is used for inverting to obtain target concentration, and the second hidden layer is used for inverting to obtain the target emission flux based on the target concentration. The invention solves the technical problem of lower efficiency of obtaining the emission flux of the atmospheric pollutants by inversion in the related technology.

Description

Atmospheric pollutant processing method, storage medium and computer terminal

Technical Field

The invention relates to the field of treatment of atmospheric pollutants, in particular to a method for treating the atmospheric pollutants, a storage medium and a computer terminal.

Background

The traditional carbon inversion task based on satellite remote sensing data is mainly divided into two parts: firstly, solving an inverse problem through a solar radiation transmission model according to radiation spectrum data of a satellite to obtain the concentration of carbon dioxide in the atmosphere; after obtaining the atmospheric carbon dioxide concentration spatiotemporal distribution, the spatiotemporal distribution of the carbon emission flux source sink can be obtained by solving an inverse problem about the carbon emission flux through an atmospheric transport model.

Whether the carbon dioxide concentration is inverted by spectral data or the carbon emission flux is inverted by the carbon dioxide concentration, the conventional inversion method relies on solving the inverse problem associated with the physical process, the former relying on a solar radiation transport model and the latter relying on a more complex atmospheric chemical transport model. Moreover, the traditional inversion method mostly adopts an inverse problem variational optimization solution scheme or an ensemble Kalman filtering scheme, and usually consumes more computing resources.

More importantly, for the carbon emission flux inversion task, the traditional approach relies on inversion data of atmospheric carbon dioxide concentration as input. However, in the process of inverting the carbon dioxide concentration by using the spectral data, a large error is mostly introduced by using the traditional method, so that the accuracy of inversion of the carbon emission flux is greatly influenced. Secondly, in the carbon emission flux inversion process, the carbon emission flux inversion tasks of the land and ocean ecosphere are usually processed separately, so that the calculation amount linearly increases as the number of the inversion tasks increases.

In view of the above problems, no effective solution has been proposed.

Disclosure of Invention

The embodiment of the invention provides an atmospheric pollutant processing method, a storage medium and a computer terminal, which at least solve the technical problem of low efficiency of obtaining the atmospheric pollutant emission flux by inversion in the related technology.

According to an aspect of an embodiment of the present invention, there is provided a method for treating atmospheric pollutants, including: acquiring target spectral data and meteorological data, wherein the target spectral data is used for representing radiation data of atmospheric pollutants; and processing the target spectrum data and the gas image data by using a target inversion model to obtain the target emission flux of the atmospheric pollutants, wherein the target inversion model comprises a first hidden layer and a second hidden layer, the first hidden layer is used for inverting to obtain target concentration, and the second hidden layer is used for inverting to obtain the target emission flux based on the target concentration.

According to another aspect of the embodiments of the present invention, there is also provided a method for treating atmospheric pollutants, including: acquiring target spectral data and meteorological data, wherein the target spectral data is used for representing radiation data of atmospheric pollutants; and processing the target spectrum data and the gas image data by using a target inversion model to obtain the target emission flux of the atmospheric pollutants, wherein the target inversion model comprises a first hidden layer and a second hidden layer, the first hidden layer is used for inverting to obtain target concentration, and the second hidden layer is used for inverting to obtain the target emission flux based on the target concentration.

According to another aspect of the embodiments of the present invention, there is also provided a method for treating atmospheric pollutants, including: acquiring carbon spectral data and meteorological data, wherein the carbon spectral data is used for representing radiation data of carbon; and processing the carbon spectrum data and the meteorological data by using a target inversion model to obtain land carbon emission flux and ocean carbon emission flux, wherein the target inversion model comprises a first hidden layer and a second hidden layer, the first hidden layer is used for inverting to obtain carbon concentration, and the second hidden layer is used for inverting to obtain the land carbon emission flux and the ocean carbon emission flux based on the carbon concentration.

According to another aspect of the embodiments of the present invention, there is also provided a method for treating atmospheric pollutants, including: the cloud server receives target spectrum data and meteorological data uploaded by the client, wherein the target spectrum data is used for representing radiation data of atmospheric pollutants; the cloud server processes the target spectrum data and the gas image data by using a target inversion model to obtain target emission flux of the atmospheric pollutants, wherein the target inversion model comprises a first hidden layer and a second hidden layer, the first hidden layer is used for inversion to obtain target concentration, and the second hidden layer is used for inversion to obtain the target emission flux based on the target concentration.

According to another aspect of the embodiments of the present invention, there is also provided a storage medium, where the storage medium includes a stored program, and when the program runs, the apparatus where the computer readable storage medium is located is controlled to execute the method for treating the atmospheric pollutants in any one of the above embodiments.

According to another aspect of the embodiments of the present invention, there is also provided a computer terminal, including: a processor and a memory, the processor being configured to execute the program stored in the memory, wherein the program is configured to execute the method for treating atmospheric pollutants in any of the above embodiments.

In the embodiment of the invention, target spectrum data and meteorological data are firstly acquired, wherein the target spectrum data is used for representing radiation data of atmospheric pollutants; the target inversion model is used for processing the target spectrum data and the gas image data to obtain the target emission flux of the atmospheric pollutants, wherein the target inversion model comprises a first hidden layer and a second hidden layer, the first hidden layer is used for obtaining target concentration through inversion, the second hidden layer is used for obtaining the target emission flux based on target concentration through inversion, and the efficiency of obtaining the atmospheric pollutants emission flux through inversion is improved. It is easy to notice that the inversion of the target concentration of the atmospheric pollutants and the inversion of the target emission flux of the atmospheric pollutants can be wholly performed in one model, the inversion efficiency can be greatly improved, and compared with a mode of obtaining the target emission flux through two-step inversion in the related technology, the inversion method has the advantages that the first hidden layer and the second hidden layer belong to the same target inversion model, so that the influence of the inversion error of the target concentration on the downstream target emission flux can be relieved, the inversion accuracy of the target emission flux can be improved, and the technical problem of low efficiency of obtaining the atmospheric pollutant emission flux through inversion in the related technology is solved.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. In the drawings:

fig. 1 shows a hardware configuration block diagram of a computer terminal (or mobile device) for implementing a method for treating atmospheric pollutants;

FIG. 2 is a flow chart of a method of treating atmospheric pollutants according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a network structure of a target inversion model according to an embodiment of the invention;

FIG. 4 is a schematic diagram of a conventional distributed inverse-mission carbon inversion scheme;

FIG. 5 is a schematic diagram of an integrated multi-task carbon inversion scheme provided in accordance with an embodiment of the present invention;

FIG. 6 is a flow chart of another method of treating atmospheric pollutants according to an embodiment of the present invention;

FIG. 7 is a flow chart of yet another method for treating atmospheric pollutants in accordance with an embodiment of the present invention;

FIG. 8 is a schematic view of an apparatus for treating atmospheric pollutants according to an embodiment of the present invention;

FIG. 9 is a schematic view of another atmospheric pollutant treating device according to an embodiment of the present invention;

FIG. 10 is a schematic view of a further apparatus for treating atmospheric pollutants in accordance with an embodiment of the present invention;

fig. 11 is a block diagram of a computer terminal according to an embodiment of the present invention.

Detailed Description

In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

First, some terms or terms appearing in the description of the embodiments of the present invention are applicable to the following explanations:

satellite radiation spectrum: the radiation received by the satellite mainly comprises thermal radiation emitted by the ground and solar radiation reflected by the ground, the former being in the infrared region and the latter being in the near-infrared region. Due to the absorption effect of carbon dioxide in the atmosphere, the spectral data received by the satellite has obvious absorption peaks, so that the concentration of the carbon dioxide in the atmosphere can be inverted through the spectral data.

Carbon inversion: satellite data carbon inversion generally involves two tasks: the method comprises the steps of inverting the concentration of carbon dioxide in the atmosphere according to satellite spectrum data, and inverting the space-time distribution of carbon emission flux according to the concentration data of the carbon dioxide in the atmosphere.

At present, global warming not only causes the rise of sea level, but also increases the frequency of extreme disastrous weather, so reducing carbon emissions is becoming an increasingly common concern of society. The monitoring of the atmospheric carbon dioxide concentration and the analysis of global carbon emission flux sources and sinks play an important guiding role in formulating carbon emission regulations of various countries. China in 2020 puts forward a carbon dioxide emission reduction target, and aims to reach a carbon emission peak value before 2030 years and realize carbon neutralization before 2060 years. The method has the advantages that the carbon dioxide emission monitoring is efficient and accurate, the implementation of a double-carbon target can be effectively supported and guided, more scientific research support and fact basis can be provided for the negotiation of weather meetings in China, and more speaking rights can be obtained in the field of environment negotiation. The proposal adopts a multi-task machine learning scheme, establishes a model for directly inverting the space-time distribution of carbon emission flux based on satellite radiation spectrum data, and realizes multi-task carbon emission flux prediction of marine and land ecosphere.

Gases in the atmosphere, such as carbon dioxide and methane, can transmit short-wave radiation incident to the ground from the sun, but can absorb long-wave radiation from the ground, so that the heat dissipation of the earth is slowed down, and global warming is caused. The carbon dioxide is the most important greenhouse gas at present due to the characteristics of high content, long residence time and the like. A great deal of research has shown that carbon dioxide emitted by human production and life, especially carbon emissions that have increased dramatically after the industrial revolution, are the leading causes of greenhouse effect and global warming.

Global warming not only causes the rise of sea level but also increases the frequency of extreme disastrous weather, so reducing carbon emissions is becoming an increasing concern in international conferences. The monitoring of atmospheric carbon dioxide concentration and the analysis of global carbon emission flux sources and sinks play an important guiding role in the formulation of carbon emission policies of various countries. China in 2020 puts forward a carbon dioxide emission reduction target, and aims to reach a carbon emission peak value before 2030 years and realize carbon neutralization before 2060 years. The method has the advantages that the carbon dioxide emission monitoring is efficient and accurate, the implementation of the national 'double carbon' target can be effectively supported and guided, more scientific research support and fact basis can be provided for the international climate conference negotiation of China, and more speaking rights can be obtained in the field of environmental socialization.

The traditional base monitoring station causes great difficulty in the inversion of carbon concentration and carbon emission flux in the global range due to the reasons of less stations, sparse spatial distribution and the like. The greenhouse gas monitoring method based on satellite remote sensing is a main observation mode for monitoring global greenhouse gas change due to the fact that stable, long-time sequence and continuous and uninterrupted observation data of a global large area range can be provided. The proposal adopts a multi-task machine learning scheme, establishes a model for directly inverting the space-time distribution of carbon emission flux based on satellite radiation spectrum data, and realizes multi-task carbon emission flux prediction of marine and land ecosphere.

For current carbon emission flux inversion tasks, the traditional approach relies on inversion data of atmospheric carbon dioxide concentration as input. However, in the process of inverting the carbon dioxide concentration by using the spectral data, a large error is mostly introduced by using the traditional method, so that the accuracy of inversion of the carbon emission flux is greatly influenced. Secondly, in the carbon emission flux inversion process, the carbon emission flux inversion tasks of the land and ocean ecosphere are usually processed separately, so that the calculation amount linearly increases as the number of the inversion tasks increases.

In order to solve the problems, the application provides a processing method of the atmospheric pollutants, and compared with the traditional inversion algorithm (a variation method and a collective Kalman filtering method) based on solving an inverse problem, by integrating the tasks of carbon concentration inversion and carbon emission flux inversion into a model, the inversion speed of satellite data can be greatly improved through the quick online prediction capability of a machine learning model. Meanwhile, compared with the traditional two-step inversion method, the method can relieve the influence of carbon concentration inversion errors on the inversion accuracy of the downstream carbon emission flux, so that the accuracy of carbon source convergence prediction is improved.

Example 1

There is also provided, in accordance with an embodiment of the present invention, an embodiment of a method for treating atmospheric pollutants, it being noted that the steps illustrated in the flowchart of the accompanying drawings may be implemented in a computer system, such as a set of computer-executable instructions, and that, although a logical order is illustrated in the flowchart, in some cases, the steps illustrated or described may be performed in an order different than that illustrated herein.

The method provided by the first embodiment of the present invention may be executed in a mobile terminal, a computer terminal, or a similar computing device. Fig. 1 shows a hardware configuration block diagram of a computer terminal (or mobile device) for implementing the atmospheric pollutant treatment method. As shown in fig. 1, the computer terminal 10 (or mobile device 10) may include one or more processors (shown as 102a, 102b, … …, 102n in the figures) which may include, but are not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA, a memory 104 for storing data, and a transmission module 106 for communication functions. Besides, the method can also comprise the following steps: a display, an input/output interface (I/O interface), a Universal Serial BUS (USB) port (which may be included as one of the ports of the BUS), a network interface, a power source, and/or a camera. It will be understood by those skilled in the art that the structure shown in fig. 1 is only an illustration and is not intended to limit the structure of the electronic device. For example, the computer terminal 10 may also include more or fewer components than shown in FIG. 1, or have a different configuration than shown in FIG. 1.

It should be noted that the one or more processors and/or other data processing circuitry described above may be referred to generally herein as "data processing circuitry". The data processing circuitry may be embodied in whole or in part in software, hardware, firmware, or any combination thereof. Further, the data processing circuit may be a single stand-alone processing module, or incorporated in whole or in part into any of the other elements in the computer terminal 10 (or mobile device). As referred to in the embodiments of the invention, the data processing circuit acts as a processor control (e.g. selection of a variable resistance termination path connected to the interface).

The memory 104 may be used to store software programs and modules of application software, such as program instructions/data storage devices corresponding to the method for treating the atmospheric pollutants in the embodiment of the present invention, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, implementing the above-mentioned method for treating the atmospheric pollutants. 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 examples, the memory 104 may further include memory located remotely from the processor, which may be connected to the computer terminal 10 via a network. Examples of such 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 for receiving or transmitting data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a Network adapter (NIC) that can be connected to other Network devices through a base station to communicate with the internet. In one example, the transmission device 106 can be a Radio Frequency (RF) module, which is used to communicate with the internet in a wireless manner.

The display may be, for example, a touch screen type Liquid Crystal Display (LCD) that may enable a user to interact with a user interface of the computer terminal 10 (or mobile device).

It should be noted here that in some alternative embodiments, the computer device (or mobile device) shown in fig. 1 described above may include hardware elements (including circuitry), software elements (including computer code stored on a computer-readable medium), or a combination of both hardware and software elements. It should be noted that fig. 1 is only one example of a particular specific example and is intended to illustrate the types of components that may be present in the computer device (or mobile device) described above.

Under the operating environment, the application provides a method for treating the atmospheric pollutants as shown in fig. 2. Fig. 2 is a flowchart of a method for treating atmospheric pollutants according to a first embodiment of the present invention.

Step S202, target spectrum data and meteorological data are acquired.

Wherein the target spectral data is used to represent radiation data of the atmospheric pollutant.

The above-mentioned atmospheric pollutant concentrations may be carbon dioxide, PM2.5, sulfides, suspended particulate matter (e.g., dust, smoke), and the like.

The target spectrum data may be acquired by a satellite. The radiation received by the satellite mainly comprises ground released thermal radiation and ground reflected solar radiation, the ground released thermal radiation is in an infrared light region, the ground reflected solar radiation is in a near-infrared light region, and due to the absorption effect of carbon dioxide in the atmosphere, the spectrum data received by the satellite has an obvious absorption peak, so that the concentration of the carbon dioxide in the atmosphere can be obtained through inversion of target spectrum data.

The meteorological data can be temperature, humidity, air pressure, cloud, aerosol and the like.

In an alternative embodiment, the target spectral data may be a time series of satellite spectra over a preset time period. The meteorological data may be a time series of meteorological elements over a preset time period.

In an alternative embodiment, the target spectral data may be acquired by a correlated carbon monitoring satellite (GOSAT), and the meteorological data may be acquired from meteorological prediction or reanalysis data.

And step S204, processing the target spectrum data and the meteorological data by using the target inversion model to obtain the target emission flux of the atmospheric pollutants.

The target inversion model comprises a first hidden layer and a second hidden layer, the first hidden layer is used for obtaining target concentration through inversion, and the second hidden layer is used for obtaining target emission flux based on target concentration through inversion.

The target concentration may be an atmospheric carbon dioxide concentration, wherein the atmospheric carbon dioxide concentration may be a carbon concentration time series within a preset time period, and the target emission flux may be an average terrestrial carbon emission flux and/or an average marine carbon emission flux.

The first hidden layer and the second hidden layer can be efficient representations of the original high-dimensional input compressed into a low-dimensional hidden space (latency). The hidden layer representation is dependent on the downstream prediction task. The first hidden layer may be a shared hidden layer, and the shared hidden layer representation indicates that multiple tasks are considered when compressing or finding a high-efficiency hidden layer representation.

In an optional embodiment, if a plurality of tasks need to be simultaneously inverted through a target inversion model, a first hidden layer of the target inversion model can consider the plurality of tasks, so that inversion accuracy of the plurality of tasks can be improved, for example, if three tasks of atmospheric carbon dioxide concentration, terrestrial carbon emission flux and ocean carbon emission flux need to be simultaneously processed through the target inversion model, the three prediction tasks share a shallow hidden layer representation, that is, the first hidden layer can be shared, the carbon dioxide concentration can be obtained through inversion of the first hidden layer, the latter two tasks can share a hidden layer representation after carbon dioxide concentration prediction, that is, the second hidden layer can be shared, wherein the second hidden layer comprises two parallel sub-hidden layers which are respectively directed at the terrestrial carbon emission flux task and the ocean carbon emission flux task.

In another alternative embodiment, when the target inversion model is obtained through training, three tasks of atmospheric carbon dioxide concentration, terrestrial carbon emission flux and marine carbon emission flux prediction can be used for supervision at the same time, and the main structure is realized through an encoding-decoding process.

In another alternative embodiment, the carbon transaction may be implemented by acquiring the target emission flux of the atmospheric pollutants in a carbon transaction scenario, wherein the carbon transaction refers to a transaction in which the carbon dioxide emission right is taken as a commodity, and the buyer can acquire a certain amount of carbon dioxide emission right by paying a certain amount to the seller, so as to form the carbon dioxide emission right. The target emission flux is directly obtained through inversion of the target inversion model, and the target emission flux can share the first hidden layer of other tasks in the process of obtaining the target emission flux, so that the accuracy of the target emission flux obtained through the target inversion model is high. Therefore, in a carbon trading scenario, carbon trading may be performed according to a target emission flux with higher accuracy, thereby reducing associated economic losses.

In another optional embodiment, the carbon emission flux may be estimated in a new energy resource scene, for example, a new energy resource vehicle scene, a target emission flux of carbon dioxide that can be reduced by using the new energy resource vehicle may be determined, optionally, target spectrum data and meteorological data of carbon dioxide emitted in an operation area of the new energy resource vehicle may be obtained, the target spectrum data and the meteorological data are processed by using a target inversion model to obtain the target emission flux of carbon dioxide, an emission flux of a historical time period in which the new energy resource vehicle is not used in the area may be obtained, and an emission reduction effect of the new energy resource vehicle may be determined by comparing the emission flux of the historical time period with the target emission flux.

In yet another alternative embodiment, the target emission flux of carbon dioxide may be estimated in an automatic driving scenario, and since the vehicle in the automatic driving scenario is driven by a new energy source, the target emission flux of carbon dioxide may have a certain effect on reducing the emission flux of carbon dioxide, the target spectrum data and the meteorological data of carbon dioxide in the operation area of the automatic driving vehicle may be obtained, so that the target spectrum data and the meteorological data are processed by using a target inversion model to obtain the target emission flux of atmospheric pollutants, the emission flux of a historical time period in which the automatic driving vehicle is not used in the area may be obtained, and the emission reduction effect of the automatic driving vehicle may be determined by comparing the emission flux of the historical time period with the target emission flux.

Through the steps, target spectrum data and meteorological data are obtained firstly, wherein the target spectrum data are used for representing radiation data of atmospheric pollutants; the target inversion model is used for processing the target spectrum data and the gas image data to obtain the target emission flux of the atmospheric pollutants, wherein the target inversion model comprises a first hidden layer and a second hidden layer, the first hidden layer is used for obtaining target concentration through inversion, the second hidden layer is used for obtaining the target emission flux based on target concentration through inversion, and the efficiency of obtaining the atmospheric pollutants emission flux through inversion is improved. It is easy to notice that the inversion of the target concentration of the atmospheric pollutants and the inversion of the target emission flux of the atmospheric pollutants can be wholly performed in one model, the inversion efficiency can be greatly improved, and compared with a mode of obtaining the target emission flux through two-step inversion in the related technology, the inversion method has the advantages that the first hidden layer and the second hidden layer belong to the same target inversion model, so that the influence of the inversion error of the target concentration on the downstream target emission flux can be relieved, the inversion accuracy of the target emission flux can be improved, and the technical problem of low efficiency of obtaining the atmospheric pollutant emission flux through inversion in the related technology is solved.

In the above embodiment of the present application, processing the target spectrum data and the meteorological data by using the target inversion model to obtain the target emission flux of the atmospheric pollutants includes: inverting the target spectrum data and the gas image data by using the first hidden layer to obtain the target concentration of the atmospheric pollutants; and inverting the target concentration by using the second hidden layer to obtain the target emission flux.

The target spectral data is a satellite spectral time series. The meteorological data is a meteorological element time series. The target concentration of the atmospheric pollutant may be a carbon dioxide concentration. The target emission flux described above may be an average ocean carbon emission flux and an average land carbon emission flux.

Fig. 3 is a schematic diagram of a network structure of an object inversion model according to an embodiment of the present invention, where 1 represents a satellite spectrum time series, 2 represents a meteorological element time series, 3 represents a carbon dioxide concentration, 4 represents an average terrestrial carbon emission flux, 5 represents an average marine carbon emission flux, 6 represents a first hidden layer, and 7 represents a second hidden layer. As shown in fig. 3, the satellite spectrum time series and the meteorological element time series may be input to the first hidden layer by means of encoding and decoding, then the first hidden layer may obtain the carbon dioxide concentration by inversion according to the satellite spectrum time series and the meteorological element time series, and the carbon dioxide concentration may be input to the second hidden layer by means of encoding and decoding, so that the second hidden layer may obtain the average ocean carbon emission flux and the average terrestrial carbon emission flux by means of inversion according to the carbon dioxide concentration.

In the above embodiments of the present application, the target emission flux includes a land emission flux and an ocean emission flux, where the target concentration is inverted by using the second hidden layer to obtain the target emission flux, and the method includes: inverting the target concentration by using a first sub hidden layer contained in the second hidden layer to obtain land emission flux; inverting the target concentration by using a second sub hidden layer contained in the second hidden layer to obtain ocean emission flux; the first sub hidden layer and the second sub hidden layer run in parallel.

The land discharge flux may be an average land discharge flux, that is, the average land discharge flux may be calculated by averaging a time series of land discharge fluxes obtained within a preset time period. The marine emission flux may be an average marine emission flux, that is, the average marine emission flux may be obtained by averaging a time series of marine emission fluxes obtained within a preset time period.

In an optional embodiment, the land emission flux and the marine emission flux may be obtained simultaneously through a target inversion model, and optionally, after the target concentration is obtained through the first hidden layer, the target concentration may be inverted through a first sub hidden layer in the second hidden layer to obtain the land emission flux, and at the same time, the target concentration may be inverted through a second sub hidden layer in the second hidden layer to obtain the marine emission flux.

In the above embodiment of the present application, the output layer of the first hidden layer is connected to the input layer of the second hidden layer in a plug-in and pull-out manner, and the method further includes: receiving a target inversion task; determining a plugging state between an output layer of the first hidden layer and an input layer of the second hidden layer based on a target inversion task; and based on the plugging state, processing the target spectrum data and the meteorological data by using the target inversion model to obtain an inversion result corresponding to the plugging state.

The target inversion task described above may require a target inversion model to obtain the results required by the user. The target inversion task can be to obtain only the target concentration, and the target inversion task can also be to obtain only the target emission flux. The target inversion task may also be to obtain only the average marine emission flux, and the target inversion task may also be to obtain only the average land emission flux.

The plugging state is used for controlling and realizing different inversion tasks.

In an optional embodiment, a target inversion task sent by a user may be received, and according to the target inversion task, the plugging/unplugging state between the output layer of the first hidden layer and the input layer of the second hidden layer may be controlled, so that the target spectrum data and the meteorological data are processed through the target inversion model, and an inversion result corresponding to the plugging/unplugging state is obtained. Optionally, when the target inversion task only needs to obtain the target concentration, the input layer of the second hidden layer may be pulled out of the output layer of the first hidden layer, so that the target inversion model may be directly output on the output layer of the first hidden layer to obtain the target concentration without performing subsequent operations.

When the target inversion task needs to acquire the target emission flux, the input layer of the second hidden layer can be inserted into the output layer of the first hidden layer, so that the target inversion model can acquire the target emission flux on the output layer of the second hidden layer.

When the target inversion task needs to acquire only the average marine emission flux, the input layer of the first sub hidden layer in the second hidden layer can be pulled out of the output layer of the first hidden layer, so that the target inversion task can directly output the average marine emission flux at the output layer of the second sub hidden layer in the second hidden layer. The manner in which only the average land emission flux is obtained is the same as the above method and will not be described further herein.

In another optional embodiment, if the target concentration and the target emission flux are to be acquired simultaneously, the target concentration may be copied when the target inversion model performs inversion to obtain the target concentration, and the copied target concentration and the copied target emission flux may be output.

In the above embodiments of the present application, the target inversion task includes at least one of the following: the system comprises a first inversion task and a second inversion task, wherein the first inversion task is used for representing inversion to obtain target concentration, the plugging state corresponding to the first inversion task is that an output layer of a first hidden layer is not connected with an input layer of a second hidden layer in a plugging mode, the second inversion task is used for representing inversion to obtain target emission flux, and the plugging state corresponding to the second inversion task is that the output layer of the first hidden layer is connected with the input layer of the second hidden layer in a plugging mode.

The non-connection in the plugging and unplugging manner means that the input layer of the second hidden layer is unplugged from the output layer of the first hidden layer, so that the second hidden layer and the first hidden layer are not connected. And then only the target concentration can be obtained by inversion in the first inversion task.

The above connection by plugging means that the input layer of the second hidden layer is inserted into the output layer of the first hidden layer, so that the second hidden layer and the first hidden layer are connected. And further, inversion in a second inversion task can be completed to obtain target emission flux.

In the above embodiment of the present application, the method further includes: acquiring training data, wherein the training data comprises: sample spectrum data, sample meteorological data, a sample concentration observation value and a sample emission flux observation value; carrying out inversion on the sample spectrum data and the sample meteorological data by using the first hidden layer to obtain the predicted sample concentration of the atmospheric pollutants; inverting the predicted sample concentration and the sample meteorological data by using the second hidden layer to obtain predicted emission flux; determining a target loss function based on the sample concentration observed value, the predicted sample concentration, the sample emission flux observed value and the predicted emission flux; and training the target inversion model by using the target loss function.

The sample concentration observations described above may be the actual observed concentration of atmospheric pollutants. The sample emission flux observation described above may be an actually observed emission flux of atmospheric pollutants.

In an optional embodiment, after the sample spectrum data and the sample meteorological data are inverted by using the first hidden layer, the predicted sample concentration of the atmospheric pollutants can be obtained, and the accuracy of the target inversion model can be improved according to the difference between the predicted sample concentration and the sample concentration observed value. The predicted sample concentration and the sample meteorological data can be inverted according to the second hidden layer to obtain the predicted emission flux, and the precision of the target inversion model can be improved according to the difference value between the predicted emission flux and the sample emission flux.

In another alternative embodiment, in order to avoid the overfitting of the target inverse model, the generalization capability of the inverse model can be improved by using the deviation between the predicted emission flux and the prior emission flux as a regularization function. The regularization refers to the last term of a loss function, and a general regularization function can make certain punishment on values of model parameters to improve the generalization effect of the model, because for an inversion problem, the inversion model is required to predict to approximate observation data as much as possible, but meanwhile, the prediction is expected not to have too large deviation with the prior (empirical) estimation. The deviation from the observed data will be as a first loss function and the deviation from the a priori estimate will be as part of the regularization, as part of the last term.

In the above embodiments of the present application, determining a target loss function based on the sample concentration observed value, the predicted sample concentration, the sample emission flux observed value, and the predicted emission flux includes: constructing a first loss function based on the sample concentration observed value and the predicted sample concentration; constructing a second loss function based on the sample emission flux observed value and the predicted emission flux; and obtaining the weighted sum of the first loss function and the second loss function to obtain a target loss function.

In an alternative embodiment, a first loss function may be constructed from the difference between the sample concentration observations and the predicted sample concentrations, and a second loss function may be constructed from the difference between the sample emission flux observations and the predicted emission flux.

The weight value of the weighted sum of the first loss function and the second loss function can be set by a user according to requirements.

In the above embodiment of the present application, the method further includes: determining the grid point concentration of each grid point on the target map based on the target concentration; and displaying the grid point concentration on the target map.

The target map may be a map corresponding to an area where the concentration of the atmospheric pollutant needs to be detected, and the target map may also be a global map.

The target concentration may be a plurality of plant atmospheric contaminant concentrations.

The grid point may be a location on the target map where a factory that emits atmospheric pollutants is located. The grid point concentration may be a plant-emitted atmospheric pollutant concentration.

In an alternative embodiment, the grid point concentration of the grid point where the factory is located on the target map can be determined according to the target concentration, and the grid point concentration is displayed on the target map. Optionally, the data of the grid point concentration may be displayed on the grid points corresponding to the target map, and the grid point emission flux may be displayed on the grid points corresponding to the target map in an air mass manner, where the larger the air mass, the more the concentration is indicated, and the smaller the air mass, the less the concentration is indicated; the grid point discharge flux corresponding to the grid points of the target map can be represented by other vector diagrams, wherein the larger the vector diagram is, the larger the density is, and the smaller the vector diagram is, the smaller the density is.

In the above embodiment of the present application, after the grid point concentration is displayed on the target map, the method further includes: receiving a selected target area on a target map; summarizing the grid point concentrations of all grid points contained in the target area to obtain the area concentration corresponding to the target area; and outputting the regional density.

In an optional embodiment, the user may further select a sum of concentrations corresponding to the target area from the target map, optionally, the user may determine the target area on the target map, may determine the target area by clicking the corresponding area in the target area, may also determine the target area by framing the area of the target map, and after the target area is determined, may sum the grid point concentrations of all grid points included in the target area to obtain an area concentration corresponding to the target area, so that the area concentration of the target area may be output, so that the user may analyze the area concentration of the target area.

In the above embodiment of the present application, the method further includes: determining a grid point emission flux of each grid point on the target map based on the target emission flux; and displaying the grid point emission flux on the target map.

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

The target emission flux may be an emission flux of a plurality of plant atmospheric pollutants.

The grid point may be a location on the target map where a factory that emits atmospheric pollutants is located. The grid point emission flux may be an emission flux of atmospheric pollutants emitted from a plant.

In the above embodiment of the present application, after the grid point discharge flux is displayed on the target map, the method further includes: receiving a selected target area on a target map; summarizing the grid point discharge fluxes of all grid points contained in the target area to obtain an area discharge flux corresponding to the target area; the output area discharges flux.

In an optional embodiment, the user may further select a total emission flux corresponding to the target area from the target map, optionally, the user may determine the target area on the target map, may determine the target area by clicking the corresponding area in the target area, may also determine the target area by selecting the area of the target map in a frame, and after the target area is determined, may collect the grid point emission fluxes of all grid points included in the target area to obtain an area emission flux corresponding to the target area, so that the area emission flux of the target area may be output, so that the user may analyze the area emission flux of the target area.

FIG. 4 is a schematic diagram of a conventional distributed inverse task carbon inversion scheme, which is processed separately for different inversion tasks, i.e., the tasks of carbon concentration time series, average terrestrial carbon emission flux, and average marine carbon emission flux; when processing the task of the carbon concentration time sequence, performing carbon inversion through a satellite spectrum time sequence and a meteorological element time sequence to obtain the carbon concentration time sequence; when the task of processing the average terrestrial carbon emission flux is processed, carbon inversion is carried out through a carbon concentration time sequence and a meteorological element time sequence to obtain the average terrestrial carbon emission flux; when processing the task of the carbon concentration time series, the average terrestrial carbon emission flux is obtained by performing carbon inversion through the carbon concentration time series and the meteorological element time series. For the carbon emission flux inversion task, the traditional approach relies on inversion data of atmospheric carbon dioxide concentration as input. However, in the process of inverting the carbon dioxide concentration by using the spectral data, a large error is mostly introduced by using the traditional method, so that the accuracy of inversion of the carbon emission flux is greatly influenced. Secondly, in the carbon emission flux inversion process, the carbon emission flux inversion tasks of the land and ocean ecosphere are usually processed separately, so that the calculation amount linearly increases as the number of the inversion tasks increases.

Fig. 5 is a schematic diagram of an integrated multi-task carbon inversion scheme provided in an embodiment of the present invention, in a process of training a target inversion model, three tasks of atmospheric carbon dioxide concentration, terrestrial carbon flux, and marine carbon flux prediction may be simultaneously supervised, the three prediction tasks may share a shallow hidden layer representation, and after a carbon concentration time series is obtained by inversion according to a satellite spectrum time series and a meteorological element time series, features between different tasks may be shared in a multi-task learning manner, so that an average terrestrial emission flux and an average marine emission flux with higher accuracy may be obtained.

As can be seen from fig. 4 and 5, compared with the conventional satellite data carbon inversion scheme, the carbon concentration inversion and carbon emission flux inversion tasks are integrated into one model, and compared with the conventional inversion algorithm (variation method and ensemble kalman filtering method) based on solving the inverse problem, the rapid online prediction capability of the machine learning model can greatly improve the inversion speed of the satellite data. Meanwhile, compared with the traditional two-step inversion method, the method can relieve the influence of carbon concentration inversion errors on the inversion accuracy of the downstream carbon emission flux, and improve the accuracy of carbon source sink prediction. Finally, the scheme adopts a method for simultaneously training a plurality of carbon emission flux prediction tasks, so that a large amount of resource consumption caused by independent processing of a plurality of tasks is avoided.

It should be noted that, for simplicity of description, the above-mentioned method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present invention is not limited by the order of acts, as some steps may occur in other orders or concurrently in accordance with the invention. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and modules referred to are not necessarily required by the invention.

Through the above description of the embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but the former is a better implementation mode in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (such as a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.

Example 2

Fig. 6 is a flow chart of a method for treating atmospheric pollutants according to an embodiment of the present invention, and as shown in fig. 6, the method may include the following steps:

step S602, acquiring carbon spectrum data and meteorological data.

Wherein the carbon spectral data is used to represent radiation data of the carbon.

And step S604, processing the carbon spectrum data and the meteorological data by using the target inversion model to obtain the terrestrial carbon emission flux and the marine carbon emission flux.

The target inversion model comprises a first hidden layer and a second hidden layer, the first hidden layer is used for obtaining carbon concentration through inversion, and the second hidden layer is used for obtaining terrestrial carbon emission flux and ocean carbon emission flux based on carbon concentration through inversion.

In the above embodiments of the present application, processing the target spectrum data and the meteorological data by using the target inversion model to obtain the terrestrial carbon emission flux and the marine carbon emission flux includes: performing inversion on the target spectrum data and the gas image data by using the first hidden layer to obtain the carbon concentration; and inverting the carbon concentration by using the second hidden layer to obtain the terrestrial carbon emission flux and the marine carbon emission flux.

In the above embodiment of the present application, the obtaining of the terrestrial carbon emission flux and the marine carbon emission flux by inverting the carbon concentration with the second hidden layer includes: inverting the carbon concentration by utilizing a first sub hidden layer contained in the second hidden layer to obtain the land carbon emission flux; inverting the carbon concentration by using a second sub hidden layer contained in the second hidden layer to obtain ocean carbon emission flux; the first sub hidden layer and the second sub hidden layer run in parallel.

In the above embodiments of the present application, the target inversion task includes at least one of the following: the system comprises a first inversion task and a second inversion task, wherein the first inversion task is used for representing inversion to obtain carbon concentration, the plugging state corresponding to the first inversion task is that an output layer of a first hidden layer is not connected with an input layer of a second hidden layer in a plugging mode, the second inversion task is used for representing inversion to obtain terrestrial carbon emission flux and ocean carbon emission flux, and the plugging state corresponding to the second inversion task is that the output layer of the first hidden layer is connected with the input layer of the second hidden layer in a plugging mode.

In the above embodiment of the present application, the method further includes: acquiring training data, wherein the training data comprises: sample carbon spectrum data, sample meteorological data, a sample carbon concentration observation value and a sample carbon emission flux observation value; inverting the sample carbon spectrum data and the sample meteorological data by using the first hidden layer to obtain a predicted sample carbon concentration; inverting the predicted sample carbon concentration and the sample meteorological data by using the second hidden layer to obtain predicted carbon emission flux; determining a target loss function based on the sample carbon concentration observed value, the predicted sample carbon concentration, the sample carbon emission flux observed value and the predicted carbon emission flux; and training the target inversion model by using the target loss function.

In the above embodiments of the present application, determining a target loss function based on the sample carbon concentration observed value, the predicted sample carbon concentration, the sample carbon emission flux observed value, and the predicted carbon emission flux includes: constructing a first loss function based on the sample carbon concentration observed value and the predicted carbon sample concentration; constructing a second loss function based on the sample carbon emission flux observed value and the predicted carbon emission flux; and obtaining the weighted sum of the first loss function and the second loss function to obtain a target loss function.

In the above embodiment of the present application, the method further includes: determining a grid point concentration of each grid point on the target map based on the carbon concentration; and displaying the grid point concentration on the target map.

In the above embodiment of the present application, the method further includes: determining grid point emission flux of each grid point on the target map based on the land carbon emission flux and the ocean carbon emission flux; and displaying the grid point emission flux on the target map.

It should be noted that the preferred embodiments described in the above examples of the present application are the same as the schemes, application scenarios, and implementation procedures provided in example 1, but are not limited to the schemes provided in example 1.

Example 3

There is also provided, in accordance with an embodiment of the present application, an embodiment of a method for treating atmospheric pollutants, it being noted that the steps illustrated in the flowchart of the accompanying drawings may be implemented in a computer system, such as a set of computer-executable instructions, and that, although a logical order is illustrated in the flowchart, in some cases, the steps illustrated or described may be performed in an order different than that illustrated herein.

Fig. 7 is a flow chart of a method for treating atmospheric pollutants according to an embodiment of the present invention, as shown in fig. 7, the method includes the following steps:

step 702, the cloud server receives target spectrum data and meteorological data uploaded by the client.

Step S704, the cloud server processes the target spectrum data and the meteorological data by using the target inversion model, so as to obtain a target emission flux of the atmospheric pollutants.

In the above embodiment of the present application, the cloud server processes the target spectrum data and the meteorological data by using the target inversion model to obtain the target emission flux of the atmospheric pollutants, including: the cloud server utilizes the first hidden layer to invert the target spectrum data and the gas image data to obtain the target concentration of the atmospheric pollutants; and the cloud server utilizes the second hidden layer to invert the target concentration to obtain the target emission flux.

In the above embodiments of the present application, the target emission flux includes a land emission flux and an ocean emission flux, where the target concentration is inverted by using the second hidden layer to obtain the target emission flux, and the method includes: the cloud server utilizes the first sub hidden layer contained in the second hidden layer to invert the target concentration to obtain the land emission flux; the cloud server utilizes a second sub hidden layer contained in the second hidden layer to invert the target concentration to obtain ocean emission flux; the first sub hidden layer and the second sub hidden layer run in parallel.

In the above embodiment of the present application, the output layer of the first hidden layer is connected to the input layer of the second hidden layer in a plug-in and pull-out manner, and the method further includes: the cloud server receives a target inversion task; the cloud server determines a plugging state between an output layer of the first hidden layer and an input layer of the second hidden layer based on the target inversion task; and the cloud server processes the target spectrum data and the meteorological data by using the target inversion model based on the plugging state to obtain an inversion result corresponding to the plugging state.

In the above embodiment of the present application, the method further includes: the cloud server acquires training data, wherein the training data comprises: sample spectrum data, sample meteorological data, a sample concentration observation value and a sample emission flux observation value; carrying out inversion on the sample spectrum data and the sample meteorological data by using the first hidden layer to obtain the predicted sample concentration of the atmospheric pollutants; the cloud server utilizes the second hidden layer to invert the predicted sample concentration and the sample meteorological data to obtain the predicted emission flux; the cloud server determines a target loss function based on the sample concentration observed value, the predicted sample concentration, the sample emission flux observed value and the predicted emission flux; and the cloud server trains the target inversion model by using the target loss function.

In the above embodiments of the present application, determining a target loss function based on the sample concentration observed value, the predicted sample concentration, the sample emission flux observed value, and the predicted emission flux includes: the cloud server constructs a first loss function based on the sample concentration observation value and the predicted sample concentration; the cloud server constructs a second loss function based on the sample emission flux observed value and the predicted emission flux; the cloud server obtains a weighted sum of the first loss function and the second loss function to obtain a target loss function.

In the above embodiment of the present application, the method further includes: the cloud server determines the grid point concentration of each grid point on the target map based on the target concentration; and the cloud server displays the grid point concentration on the target map.

In the above embodiment of the present application, after the cloud server displays the grid point concentration on the target map, the method further includes: the cloud server receives a selected target area on a target map; the cloud server collects the grid point concentrations of all grid points contained in the target area to obtain the area concentration corresponding to the target area; and the cloud server outputs the regional concentration.

In the above embodiment of the present application, the method further includes: the cloud server determines grid point emission flux of each grid point on the target map based on the target emission flux; the cloud server displays the grid point emission flux on the target map.

In the above embodiment of the present application, after the cloud server displays the grid point emission flux on the target map, the method further includes: the cloud server receives a selected target area on a target map; the cloud server collects grid point emission fluxes of all grid points contained in the target area to obtain an area emission flux corresponding to the target area; the cloud server outputs the regional emission flux.

Example 4

According to an embodiment of the present application, there is also provided an apparatus for treating an atmospheric pollutant, which is used for implementing the method for treating an atmospheric pollutant, as shown in fig. 8, the apparatus 800 includes: an obtaining module 802 and a processing module 804.

The acquisition module is used for acquiring target spectrum data and meteorological data, wherein the target spectrum data is used for representing radiation data of atmospheric pollutants; the processing module is used for processing the target spectrum data and the gas image data by using a target inversion model to obtain target emission flux of the atmospheric pollutants, wherein the target inversion model comprises a first hidden layer and a second hidden layer, the first hidden layer is used for inversion to obtain target concentration, and the second hidden layer is used for inversion to obtain the target emission flux based on the target concentration.

It should be noted here that the acquiring module 802 and the processing module 804 correspond to steps S202 to S204 in embodiment 1, and the two modules are the same as the corresponding steps in the implementation example and application scenario, but are not limited to the disclosure of embodiment 1. It should be noted that the above modules may be operated in the computer terminal 10 provided in embodiment 1 as a part of the apparatus.

In the above embodiments of the present application, the processing module includes: the device comprises a first inversion unit and a second inversion unit.

The first inversion unit is used for inverting target spectrum data and meteorological data by using the first hidden layer to obtain the target concentration of the atmospheric pollutants; and the second inversion unit is used for inverting the target concentration by using the second hidden layer to obtain the target emission flux.

In the above embodiments of the present application, the target discharge flux includes a land discharge flux and an ocean discharge flux, and the second inversion unit includes: the first inversion subunit and the second inversion subunit.

The first inversion subunit is used for inverting the target concentration by using a first sub hidden layer contained in the second hidden layer to obtain the land emission flux; the second inversion subunit is used for inverting the target concentration by using a second sub-hidden layer contained in the second hidden layer to obtain ocean emission flux; the first sub hidden layer and the second sub hidden layer run in parallel.

In the above embodiment of the present application, the output layer of the first hidden layer is connected to the input layer of the second hidden layer in a plug-in and pull-out manner, and the apparatus further includes: the device comprises a receiving module and a determining module.

The receiving module is used for receiving a target inversion task; the determining module is used for determining the plugging state between the output layer of the first hidden layer and the input layer of the second hidden layer based on the target inversion task; the processing module is further used for processing the target spectrum data and the meteorological data by using the target inversion model based on the plugging state to obtain an inversion result corresponding to the plugging state.

In the above embodiment of the present application, the apparatus further includes: an inversion module and a training module.

Wherein, the acquisition module is also used for acquiring training data, wherein, training data includes: sample spectrum data, sample meteorological data, a sample concentration observation value and a sample emission flux observation value; the inversion module is also used for inverting the sample spectrum data and the sample meteorological data by utilizing the first hidden layer to obtain the predicted sample concentration of the atmospheric pollutants; the inversion module is also used for inverting the predicted sample concentration and the sample meteorological data by utilizing the second hidden layer to obtain the predicted emission flux; the determination module is further configured to determine a target loss function based on the sample concentration observation, the predicted sample concentration, the sample emission flux observation, and the predicted emission flux; the training module is further used for training the target inversion model by using the target loss function.

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

The constructing unit is used for constructing a first loss function based on the sample concentration observation value and the predicted sample concentration; the construction unit is further used for constructing a second loss function based on the sample emission flux observed value and the predicted emission flux; the weighting unit is used for obtaining the weighted sum of the first loss function and the second loss function to obtain a target loss function.

In the above embodiment of the present application, the apparatus further includes: a first display module.

The determining module is further used for determining the grid point concentration of each grid point on the target map based on the target concentration; the first display module is used for displaying the grid point concentration on the target map.

In the above embodiment of the present application, the apparatus further includes: the device comprises a first collecting module and a first output module.

The receiving module is also used for receiving a target area selected on the target map; the first summarizing module is further used for summarizing the grid point concentrations of all grid points contained in the target area to obtain the area concentration corresponding to the target area; the first output module is also used for outputting the regional concentration.

In the above embodiment of the present application, the apparatus further includes: and a second display module.

The determining module is further used for determining the grid point emission flux of each grid point on the target map based on the target emission flux; the second display module is also used for displaying the grid point emission flux on the target map.

In the above embodiment of the present application, the apparatus further includes: the second collecting module and the second output module.

The receiving module is also used for receiving a target area selected on the target map; the second summarizing module is used for summarizing the grid point emission fluxes of all grid points contained in the target area to obtain the area emission flux corresponding to the target area; the second output module is used for outputting the regional discharge flux.

Example 5

According to an embodiment of the present application, there is also provided an atmospheric pollutant treatment apparatus for implementing the above-mentioned atmospheric pollutant treatment method, as shown in fig. 9, the apparatus includes: an acquisition module 902 and a processing module 904.

The acquisition module is used for acquiring carbon spectrum data and meteorological data, wherein the carbon spectrum data is used for representing radiation data of carbon; the processing module is used for processing the carbon spectrum data and the meteorological data by using a target inversion model to obtain terrestrial carbon emission flux and marine carbon emission flux, wherein the target inversion model comprises a first hidden layer and a second hidden layer, the first hidden layer is used for obtaining carbon concentration through inversion, and the second hidden layer is used for obtaining the terrestrial carbon emission flux and the marine carbon emission flux based on carbon concentration through inversion.

It should be noted here that the above-mentioned acquiring module 902 and processing module 904 correspond to steps S602 to S604 in embodiment 2, and the two modules are the same as the examples and application scenarios realized by the corresponding steps, but are not limited to the disclosure of embodiment 2. It should be noted that the above modules may be operated in the computer terminal 10 provided in embodiment 1 as a part of the apparatus.

Example 6

According to an embodiment of the present application, there is also provided an atmospheric pollutant treatment apparatus for implementing the above-mentioned atmospheric pollutant treatment method, as shown in fig. 10, the apparatus including: a receiving module 1002 and a processing module 1004.

The receiving module is used for receiving target spectrum data and meteorological data uploaded by a client, wherein the target spectrum data is used for representing radiation data of atmospheric pollutants; the processing module is used for processing the target spectrum data and the gas image data by using a target inversion model to obtain target emission flux of the atmospheric pollutants, wherein the target inversion model comprises a first hidden layer and a second hidden layer, the first hidden layer is used for inversion to obtain target concentration, and the second hidden layer is used for inversion to obtain the target emission flux based on the target concentration.

It should be noted here that the receiving module 1002 and the processing module 1004 correspond to steps S602 to S604 in embodiment 3, and the two modules are the same as the corresponding steps in the implementation example and application scenario, but are not limited to the disclosure in embodiment 3. It should be noted that the above modules may be operated in the computer terminal 10 provided in embodiment 1 as a part of the apparatus.

Example 7

The embodiment of the invention can provide a computer terminal which can be any computer terminal device in a computer terminal group. Optionally, in this embodiment, the computer terminal may also be replaced with a terminal device such as a mobile terminal.

Optionally, in this embodiment, the computer terminal may be located in at least one network device of a plurality of network devices of a computer network.

In this embodiment, the computer terminal may execute the program code of the following steps in the method for treating the atmospheric pollutants: acquiring target spectral data and meteorological data, wherein the target spectral data is used for representing radiation data of atmospheric pollutants; and processing the target spectrum data and the gas image data by using a target inversion model to obtain the target emission flux of the atmospheric pollutants, wherein the target inversion model comprises a first hidden layer and a second hidden layer, the first hidden layer is used for inverting to obtain target concentration, and the second hidden layer is used for inverting to obtain the target emission flux based on the target concentration.

Alternatively, fig. 11 is a block diagram of a computer terminal according to an embodiment of the present invention. As shown in fig. 11, the computer terminal a may include: one or more processors (only one shown), memory.

The memory may be configured to store software programs and modules, such as program instructions/modules corresponding to the method and apparatus for processing an atmospheric pollutant in the embodiment of the present invention, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, implements the above-described method for processing an atmospheric pollutant. 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 memory remotely located from the processor, and these remote memories may be connected to terminal a through a network. Examples of such 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 and application program stored in the memory through the transmission device to execute the following steps: acquiring target spectral data and meteorological data, wherein the target spectral data is used for representing radiation data of atmospheric pollutants; and processing the target spectrum data and the gas image data by using a target inversion model to obtain the target emission flux of the atmospheric pollutants, wherein the target inversion model comprises a first hidden layer and a second hidden layer, the first hidden layer is used for inverting to obtain target concentration, and the second hidden layer is used for inverting to obtain the target emission flux based on the target concentration.

Optionally, the processor may further execute the program code of the following steps: inverting the target spectrum data and the gas image data by using the first hidden layer to obtain the target concentration of the atmospheric pollutants; and inverting the target concentration by using the second hidden layer to obtain the target emission flux.

Optionally, the processor may further execute the program code of the following steps: inverting the target concentration by using a first sub hidden layer contained in the second hidden layer to obtain land emission flux; inverting the target concentration by using a second sub hidden layer contained in the second hidden layer to obtain ocean emission flux; the first sub hidden layer and the second sub hidden layer run in parallel.

Optionally, the processor may further execute the program code of the following steps: receiving a target inversion task; determining a plugging state between an output layer of the first hidden layer and an input layer of the second hidden layer based on a target inversion task; and based on the plugging state, processing the target spectrum data and the meteorological data by using the target inversion model to obtain an inversion result corresponding to the plugging state.

Optionally, the processor may further execute the program code of the following steps: the target inversion task includes at least one of: the system comprises a first inversion task and a second inversion task, wherein the first inversion task is used for representing inversion to obtain target concentration, the plugging state corresponding to the first inversion task is that an output layer of a first hidden layer is not connected with an input layer of a second hidden layer in a plugging mode, the second inversion task is used for representing inversion to obtain target emission flux, and the plugging state corresponding to the second inversion task is that the output layer of the first hidden layer is connected with the input layer of the second hidden layer in a plugging mode.

Optionally, the processor may further execute the program code of the following steps: acquiring training data, wherein the training data comprises: sample spectrum data, sample meteorological data, a sample concentration observation value and a sample emission flux observation value; carrying out inversion on the sample spectrum data and the sample meteorological data by using the first hidden layer to obtain the predicted sample concentration of the atmospheric pollutants; inverting the predicted sample concentration and the sample meteorological data by using the second hidden layer to obtain predicted emission flux; determining a target loss function based on the sample concentration observed value, the predicted sample concentration, the sample emission flux observed value and the predicted emission flux; and training the target inversion model by using the target loss function.

Optionally, the processor may further execute the program code of the following steps: constructing a first loss function based on the sample concentration observed value and the predicted sample concentration; constructing a second loss function based on the sample emission flux observed value and the predicted emission flux; and obtaining the weighted sum of the first loss function and the second loss function to obtain a target loss function.

Optionally, the processor may further execute the program code of the following steps: determining the grid point concentration of each grid point on the target map based on the target concentration; and displaying the grid point concentration on the target map.

Optionally, the processor may further execute the program code of the following steps: receiving a selected target area on a target map; summarizing the grid point concentrations of all grid points contained in the target area to obtain the area concentration corresponding to the target area; and outputting the regional density.

Optionally, the processor may further execute the program code of the following steps: determining a grid point emission flux of each grid point on the target map based on the target emission flux; and displaying the grid point emission flux on the target map.

Optionally, the processor may further execute the program code of the following steps: receiving a selected target area on a target map; summarizing the grid point discharge fluxes of all grid points contained in the target area to obtain an area discharge flux corresponding to the target area; the output area discharges flux.

The processor can call the information and application program stored in the memory through the transmission device to execute the following steps: acquiring carbon spectral data and meteorological data, wherein the carbon spectral data is used for representing radiation data of carbon; and processing the carbon spectrum data and the meteorological data by using a target inversion model to obtain land carbon emission flux and ocean carbon emission flux, wherein the target inversion model comprises a first hidden layer and a second hidden layer, the first hidden layer is used for inverting to obtain carbon concentration, and the second hidden layer is used for inverting to obtain the land carbon emission flux and the ocean carbon emission flux based on the carbon concentration.

The processor can call the information and application program stored in the memory through the transmission device to execute the following steps: the cloud server receives target spectrum data and meteorological data uploaded by the client, wherein the target spectrum data is used for representing radiation data of atmospheric pollutants; the cloud server processes the target spectrum data and the gas image data by using a target inversion model to obtain target emission flux of the atmospheric pollutants, wherein the target inversion model comprises a first hidden layer and a second hidden layer, the first hidden layer is used for inversion to obtain target concentration, and the second hidden layer is used for inversion to obtain the target emission flux based on the target concentration.

By adopting the embodiment of the invention, target spectrum data and meteorological data are firstly acquired, wherein the target spectrum data is used for representing the radiation data of atmospheric pollutants; the target inversion model is used for processing the target spectrum data and the gas image data to obtain the target emission flux of the atmospheric pollutants, wherein the target inversion model comprises a first hidden layer and a second hidden layer, the first hidden layer is used for obtaining target concentration through inversion, the second hidden layer is used for obtaining the target emission flux based on target concentration through inversion, and the efficiency of obtaining the atmospheric pollutants emission flux through inversion is improved. It is easy to notice that the inversion of the target concentration of the atmospheric pollutants and the inversion of the target emission flux of the atmospheric pollutants can be wholly performed in one model, the inversion efficiency can be greatly improved, and compared with a mode of obtaining the target emission flux through two-step inversion in the related technology, the inversion method has the advantages that the first hidden layer and the second hidden layer belong to the same target inversion model, so that the influence of the inversion error of the target concentration on the downstream target emission flux can be relieved, the inversion accuracy of the target emission flux can be improved, and the technical problem of low efficiency of obtaining the atmospheric pollutant emission flux through inversion in the related technology is solved.

It can be understood by those skilled in the art that the structure shown in fig. 11 is only an illustration, and the computer terminal may also be a terminal device such as a smart phone (e.g., an Android phone, an iOS phone, etc.), a tablet computer, a palmtop computer, a Mobile Internet Device (MID), a PAD, and the like. Fig. 11 is a diagram illustrating a structure of the electronic device. For example, the computer terminal 10 may also include more or fewer components (e.g., network interfaces, display devices, etc.) than shown in FIG. 11, or have a different configuration than shown in FIG. 11.

Those skilled in the art will appreciate that all or part of the steps in the methods of the above embodiments may be implemented by a program instructing hardware associated with the terminal device, where the program may be stored in a computer-readable storage medium, and the storage medium may include: flash disks, Read-Only memories (ROMs), Random Access Memories (RAMs), magnetic or optical disks, and the like.

Example 8

The embodiment of the invention also provides a storage medium. Optionally, in this embodiment, the storage medium may be configured to store a program code executed by the method for treating an atmospheric pollutant provided in the first embodiment.

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

Optionally, in this embodiment, the storage medium is configured to store program code for performing the following steps: acquiring target spectral data and meteorological data, wherein the target spectral data is used for representing radiation data of atmospheric pollutants; and processing the target spectrum data and the gas image data by using a target inversion model to obtain the target emission flux of the atmospheric pollutants, wherein the target inversion model comprises a first hidden layer and a second hidden layer, the first hidden layer is used for inverting to obtain target concentration, and the second hidden layer is used for inverting to obtain the target emission flux based on the target concentration.

Optionally, the storage medium is further configured to store program code for performing the following steps: inverting the target spectrum data and the gas image data by using the first hidden layer to obtain the target concentration of the atmospheric pollutants; and inverting the target concentration by using the second hidden layer to obtain the target emission flux.

Optionally, the storage medium is further configured to store program code for performing the following steps: inverting the target concentration by using a first sub hidden layer contained in the second hidden layer to obtain land emission flux; inverting the target concentration by using a second sub hidden layer contained in the second hidden layer to obtain ocean emission flux; the first sub hidden layer and the second sub hidden layer run in parallel.

Optionally, the storage medium is further configured to store program code for performing the following steps: receiving a target inversion task; determining a plugging state between an output layer of the first hidden layer and an input layer of the second hidden layer based on a target inversion task; and based on the plugging state, processing the target spectrum data and the meteorological data by using the target inversion model to obtain an inversion result corresponding to the plugging state.

Optionally, the storage medium is further configured to store program code for performing the following steps: the target inversion task includes at least one of: the system comprises a first inversion task and a second inversion task, wherein the first inversion task is used for representing inversion to obtain target concentration, the plugging state corresponding to the first inversion task is that an output layer of a first hidden layer is not connected with an input layer of a second hidden layer in a plugging mode, the second inversion task is used for representing inversion to obtain target emission flux, and the plugging state corresponding to the second inversion task is that the output layer of the first hidden layer is connected with the input layer of the second hidden layer in a plugging mode.

Optionally, the storage medium is further configured to store program code for performing the following steps: acquiring training data, wherein the training data comprises: sample spectrum data, sample meteorological data, a sample concentration observation value and a sample emission flux observation value; carrying out inversion on the sample spectrum data and the sample meteorological data by using the first hidden layer to obtain the predicted sample concentration of the atmospheric pollutants; inverting the predicted sample concentration and the sample meteorological data by using the second hidden layer to obtain predicted emission flux; determining a target loss function based on the sample concentration observed value, the predicted sample concentration, the sample emission flux observed value and the predicted emission flux; and training the target inversion model by using the target loss function.

Optionally, the storage medium is further configured to store program code for performing the following steps: constructing a first loss function based on the sample concentration observed value and the predicted sample concentration; constructing a second loss function based on the sample emission flux observed value and the predicted emission flux; and obtaining the weighted sum of the first loss function and the second loss function to obtain a target loss function.

Optionally, the storage medium is further configured to store program code for performing the following steps: determining the grid point concentration of each grid point on the target map based on the target concentration; and displaying the grid point concentration on the target map.

Optionally, the storage medium is further configured to store program code for performing the following steps: receiving a selected target area on a target map; summarizing the grid point concentrations of all grid points contained in the target area to obtain the area concentration corresponding to the target area; and outputting the regional density.

Optionally, the storage medium is further configured to store program code for performing the following steps: determining a grid point emission flux of each grid point on the target map based on the target emission flux; and displaying the grid point emission flux on the target map.

Optionally, the storage medium is further configured to store program code for performing the following steps: receiving a selected target area on a target map; summarizing the grid point discharge fluxes of all grid points contained in the target area to obtain an area discharge flux corresponding to the target area; the output area discharges flux.

Optionally, in this embodiment, the storage medium is configured to store program code for performing the following steps: acquiring carbon spectral data and meteorological data, wherein the carbon spectral data is used for representing radiation data of carbon; and processing the carbon spectrum data and the meteorological data by using a target inversion model to obtain land carbon emission flux and ocean carbon emission flux, wherein the target inversion model comprises a first hidden layer and a second hidden layer, the first hidden layer is used for inverting to obtain carbon concentration, and the second hidden layer is used for inverting to obtain the land carbon emission flux and the ocean carbon emission flux based on the carbon concentration.

Optionally, in this embodiment, the storage medium is configured to store program code for performing the following steps: the cloud server receives target spectrum data and meteorological data uploaded by the client, wherein the target spectrum data is used for representing radiation data of atmospheric pollutants; the cloud server processes the target spectrum data and the gas image data by using a target inversion model to obtain target emission flux of the atmospheric pollutants, wherein the target inversion model comprises a first hidden layer and a second hidden layer, the first hidden layer is used for inversion to obtain target concentration, and the second hidden layer is used for inversion to obtain the target emission flux based on the target concentration.

The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.

In the above embodiments of the present invention, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.

In the embodiments provided in the present application, it should be understood that the disclosed technology can be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one type of division of logical functions, and there may be other divisions when actually implemented, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, units or modules, and may be in an electrical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.

The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.

The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.

Claims

1. A method for treating atmospheric pollutants, comprising:

acquiring target spectral data and meteorological data, wherein the target spectral data is used for representing radiation data of atmospheric pollutants;

and processing the target spectrum data and the meteorological data by using a target inversion model to obtain the target emission flux of the atmospheric pollutants, wherein the target inversion model comprises a first hidden layer and a second hidden layer, the first hidden layer is used for inverting to obtain target concentration, and the second hidden layer is used for inverting to obtain the target emission flux based on the target concentration.

2. The method of claim 1, wherein processing the target spectral data and the meteorological data using a target inversion model to obtain a target emission flux of the atmospheric pollutants comprises:

inverting the target spectrum data and the meteorological data by using the first hidden layer to obtain the target concentration of the atmospheric pollutants;

and utilizing the second hidden layer to invert the target concentration to obtain the target emission flux.

3. The method of claim 2, wherein the target emission flux comprises a terrestrial emission flux and a marine emission flux, and wherein inverting the target concentration using the second hidden layer to obtain the target emission flux comprises:

inverting the target concentration by utilizing a first sub hidden layer contained in the second hidden layer to obtain the land emission flux;

inverting the target concentration by using a second sub hidden layer contained in the second hidden layer to obtain the marine emission flux;

wherein the first sub hidden layer and the second sub hidden layer operate in parallel.

4. The method of claim 2, wherein the output layer of the first hidden layer is connected to the input layer of the second hidden layer by plugging, and the method further comprises:

receiving a target inversion task;

determining a plugging state between an output layer of the first hidden layer and an input layer of the second hidden layer based on the target inversion task;

and processing the target spectrum data and the meteorological data by using the target inversion model based on the plugging state to obtain an inversion result corresponding to the plugging state.

5. The method of claim 4, wherein the target inversion task comprises at least one of: the target concentration inversion method comprises a first inversion task and a second inversion task, wherein the first inversion task is used for representing inversion to obtain the target concentration, the plugging state corresponding to the first inversion task is that the output layer of the first hidden layer is not connected with the input layer of the second hidden layer in a plugging mode, the second inversion task is used for representing inversion to obtain the target emission flux, and the plugging state corresponding to the second inversion task is that the output layer of the first hidden layer is connected with the input layer of the second hidden layer in a plugging mode.

6. The method of claim 1, further comprising:

obtaining training data, wherein the training data comprises: sample spectrum data, sample meteorological data, a sample concentration observation value and a sample emission flux observation value;

inverting the sample spectrum data and the sample meteorological data by using the first hidden layer to obtain the predicted sample concentration of the atmospheric pollutants;

inverting the predicted sample concentration and the sample meteorological data by using the second hidden layer to obtain predicted emission flux;

determining a target loss function based on the sample concentration observations, the predicted sample concentrations, the sample emission flux observations, and the predicted emission flux;

and training the target inversion model by using the target loss function.

7. The method of claim 6, wherein determining a target loss function based on the sample concentration observations, the predicted sample concentrations, the sample emission flux observations, and the predicted emission flux comprises:

constructing a first loss function based on the sample concentration observations and the predicted sample concentrations;

constructing a second loss function based on the sample emission flux observations and the predicted emission flux;

and obtaining the weighted sum of the first loss function and the second loss function to obtain the target loss function.

8. The method according to any one of claims 1 to 7, further comprising:

determining the grid point concentration of each grid point on the target map based on the target concentration;

and displaying the grid point concentration on the target map.

9. The method of claim 8, wherein after displaying the grid point concentration on the target map, the method further comprises:

receiving a selected target area on the target map;

summarizing the grid point concentrations of all grid points contained in the target area to obtain the area concentration corresponding to the target area;

and outputting the region concentration.

10. The method according to any one of claims 1-7, further comprising:

determining a grid point emission flux of each grid point on a target map based on the target emission flux;

displaying the grid point emission flux on the target map.

11. The method of claim 10, wherein after displaying the grid point discharge flux on the target map, the method further comprises:

receiving a selected target area on the target map;

summarizing the grid point discharge fluxes of all grid points contained in the target area to obtain an area discharge flux corresponding to the target area;

outputting the zone discharge flux.

12. A method for treating atmospheric pollutants, comprising:

acquiring carbon spectral data and meteorological data, wherein the carbon spectral data is used for representing radiation data of carbon;

and processing the carbon spectrum data and the meteorological data by using a target inversion model to obtain a terrestrial carbon emission flux and an ocean carbon emission flux, wherein the target inversion model comprises a first hidden layer and a second hidden layer, the first hidden layer is used for inverting to obtain carbon concentration, and the second hidden layer is used for inverting to obtain the terrestrial carbon emission flux and the ocean carbon emission flux based on the carbon concentration.

13. A storage medium, characterized in that the storage medium comprises a stored program, wherein when the program runs, the storage medium is controlled to execute the method for treating the atmospheric pollutants according to any one of claims 1 to 12.

14. A computer terminal, comprising: a memory and a processor for executing a program stored in the memory, wherein the program is executed to perform the method for treating atmospheric pollutants according to any one of claims 1 to 12.