Air quality forecasting method and device, storage medium and electronic equipment

Abstract

The invention provides an air quality forecasting method and device, a storage medium and electronic equipment, and the method comprises the steps: obtaining initial air quality forecasting data and site ensemble forecasting data, the initial air quality forecasting data comprises an initial pollutant concentration forecasting value of each target grid in a target area, and the site ensemble forecasting data comprises an initial pollutant concentration forecasting value of each target grid in the target area; each target grid is a grid division result of the target area under high resolution; based on the initial air quality forecast data, determining a first area concentration forecast result of each sub-area in the target area, and based on the site ensemble forecast data, determining a second area concentration forecast result of each sub-area; and determining a forecast deviation coefficient of each sub-region based on the first region concentration forecast result and the second region concentration forecast result of each sub-region so as to correct the initial air quality forecast data to obtain target air quality forecast data. According to the embodiment of the invention, the accuracy of high-resolution air quality forecasting can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of air quality, and in particular, to an air quality forecasting method, device, storage medium, and electronic device. Background Art

[0002] Currently, the spatial resolution of numerical models for forecasting the spatial distribution of pollutant concentrations is mostly 27 km (kilometers) - 9 km or nested settings such as 27 km - 9 km - 3 km; however, in order to better support the work of "precision pollution control", it is necessary to carry out pollution prediction with higher spatial accuracy (i.e., higher resolution), and the process of obtaining a fine-resolution (i.e., high-resolution, small-scale) prediction result from a coarse-resolution (i.e., low-resolution, large-scale) prediction result is generally called downscaling. In this regard, related technologies usually nest high-resolution sub-regions in the coarse-resolution simulation area to refine the coarse-resolution results using numerical model technology. However, mature numerical models are usually based on research results with mesoscale resolution. Although increasing the spatial resolution can be run at the software level, the forecasting effect is poor and the forecasting accuracy is low. Based on this, there is currently no good solution to how to improve the accuracy of high-resolution air quality forecasting. Summary of the Invention

[0003] In view of this, embodiments of the present invention provide an air quality forecasting method, device, storage medium, and electronic device to solve problems such as the low accuracy of high-resolution air quality forecasting in related technologies. That is to say, embodiments of the present invention can determine the forecasting deviation coefficient of each sub-region through site set forecasting data, etc., so as to correct the initial air quality forecasting data and obtain target air quality forecasting data with higher accuracy, which can effectively improve the accuracy of high-resolution air quality forecasting, that is, the target air quality forecasting data can not only retain the characteristics of high-precision spatial distribution but also have a relatively accurate pollutant concentration forecasting level.

[0004] According to one aspect of the present invention, there is provided an air quality forecasting method, the method comprising: Obtaining initial air quality forecasting data, and obtaining site set forecasting data, where the initial air quality forecasting data includes initial pollutant concentration forecasting values of each target grid in a target region, and each target grid is a grid division result of the target region at high resolution; Based on the initial air quality forecasting data, determining a first regional concentration forecasting result of each sub-region in the target region, and based on the site set forecasting data, determining a second regional concentration forecasting result of each sub-region; Respectively based on the first regional concentration forecasting result and the second regional concentration forecasting result of each sub-region, determining the forecasting deviation coefficient of each sub-region; Based on the forecast deviation coefficients of the respective sub-regions, correct the initial air quality forecast data to obtain the target air quality forecast data.

[0005] According to another aspect of the present invention, there is provided an air quality forecasting device, the device comprising: An acquisition unit, configured to acquire initial air quality forecast data and acquire station set forecast data, where the initial air quality forecast data includes initial pollutant concentration forecast values of each target grid in a target region, and each of the target grids is a grid division result of the target region at a high resolution; A processing unit, configured to determine a first regional concentration forecast result of each sub-region in the target region based on the initial air quality forecast data, and determine a second regional concentration forecast result of each sub-region based on the station set forecast data; The processing unit is further configured to determine the forecast deviation coefficient of each sub-region respectively based on the first regional concentration forecast result and the second regional concentration forecast result of each sub-region; The processing unit is further configured to correct the initial air quality forecast data based on the forecast deviation coefficients of the respective sub-regions to obtain the target air quality forecast data.

[0006] According to another aspect of the present invention, there is provided an electronic device, the electronic device comprising a processor and a memory storing a program, wherein the program includes instructions that, when executed by the processor, cause the processor to execute the method mentioned above.

[0007] According to another aspect of the present invention, there is provided a non-transitory computer-readable storage medium storing computer instructions, the computer instructions being used to cause a computer to execute the method mentioned above.

[0008] Embodiments of the present invention can obtain initial air quality forecast data and obtain site ensemble forecast data. The initial air quality forecast data includes initial pollutant concentration forecast values for each target grid in the target area, and each target grid is the result of grid division of the target area at a high resolution. Then, based on the initial air quality forecast data, the first regional concentration forecast results for each sub-region in the target area can be determined, and based on the site ensemble forecast data, the second regional concentration forecast results for each sub-region can be determined. Based on this, the forecast deviation coefficients for each sub-region can be determined respectively based on the first regional concentration forecast results and the second regional concentration forecast results for each sub-region; thus, based on the forecast deviation coefficients for each sub-region, the initial air quality forecast data can be corrected to obtain the target air quality forecast data. It can be seen that the embodiments of the present invention can determine the forecast deviation coefficients for each sub-region through site ensemble forecast data and the like, so as to correct the initial air quality forecast data to obtain target air quality forecast data with higher accuracy, which can effectively improve the accuracy of high-resolution air quality forecasts, that is, the target air quality forecast data can not only retain the characteristics of high-precision spatial distribution but also have a relatively accurate pollutant concentration forecast level. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In the following description of exemplary embodiments with reference to the accompanying drawings, more details, features, and advantages of the present invention are disclosed. In the drawings: Figure 1 The flowchart of an air quality forecasting method according to an exemplary embodiment of the present invention is shown; Figure 2 The flowchart of another air quality forecasting method according to an exemplary embodiment of the present invention is shown; Figure 3 The flowchart of yet another air quality forecasting method according to an exemplary embodiment of the present invention is shown; Figure 4 The schematic block diagram of an air quality forecasting device according to an exemplary embodiment of the present invention is shown; Figure 5 The structural block diagram of an exemplary electronic device capable of implementing the embodiments of the present invention is shown. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0010] Embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to more thoroughly and completely understand the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not used to limit the protection scope of the present invention.

[0011] It should be understood that the various steps described in the method embodiments of the present invention may be executed in different orders and / or executed in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this regard.

[0012] As used herein, the term "comprising" and its variations are open-ended, that is, "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description. It should be noted that the concepts such as "first" and "second" mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order of the functions executed by these devices, modules or units or their interdependent relationships.

[0013] It should be noted that the modifications of "one" and "a plurality" mentioned in the present invention are illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly stated in the context, it should be understood as "one or more".

[0014] The names of the messages or information exchanged between multiple devices in the embodiments of the present invention are only for illustrative purposes and are not used to limit the scope of these messages or information.

[0015] It should be noted that the execution subject of the air quality prediction method provided in the embodiments of the present invention may be one or more electronic devices, and the present invention does not make any limitations in this regard; among them, the electronic device may be a terminal (i.e., a client) or a server. Then, when the execution subject includes multiple electronic devices, and at least one terminal and at least one server are included in the multiple electronic devices, the air quality prediction method provided in the embodiments of the present invention may be jointly executed by the terminal and the server. Correspondingly, the terminals mentioned herein may include, but are not limited to: smart phones, tablet computers, laptop computers, desktop computers, smart watches, intelligent voice interaction devices, smart home appliances, vehicle terminals, aircraft, and so on. The server mentioned herein may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms, and so on.

[0016] Based on the above description, an embodiment of the present invention provides an air quality forecasting method, which can be executed by the aforementioned electronic device (terminal or server); alternatively, the air quality forecasting method can be jointly executed by the terminal and the server. For the sake of convenience of explanation, hereinafter, the case where the electronic device executes the air quality forecasting method will be taken as an example for illustration; as Figure 1 shown, the air quality forecasting method may include the following steps S101-S104: S101, obtain initial air quality forecasting data and obtain site set forecasting data. The initial air quality forecasting data includes the initial pollutant concentration forecasting values of each target grid in the target area, and each target grid is the grid division result of the target area at a high resolution.

[0017] Optionally, the resolution (i.e., spatial resolution) corresponding to each target grid (i.e., grid point) can be 1 km, or 2 km, etc.; the embodiments of the present invention do not limit this. Optionally, a pollutant concentration forecasting value (such as the initial pollutant concentration forecasting value, etc.) can be the concentration forecasting value of the target pollutant, that is, the embodiments of the present invention can all be for forecasting or observing the concentration of the target pollutant; optionally, the target pollutant can be PM2.5 (fine particulate matter), or CO (carbon monoxide), or NOx (nitrogen oxides), etc., and the embodiments of the present invention do not limit this. Optionally, the target area can be any area, and the embodiments of the present invention do not limit this.

[0018] In the embodiments of the present invention, the obtaining method of the initial air quality forecasting data may include but is not limited to the following several types: The first obtaining method: The initial air quality forecasting data is stored in the own storage space of the electronic device. In this case, the initial air quality forecasting data can be obtained from the own storage space.

[0019] The second obtaining method: The electronic device can obtain the download link of the initial air quality forecasting data, and use the forecasting data downloaded based on the download link of the initial air quality forecasting data as the initial air quality forecasting data.

[0020] The third acquisition method: The electronic device can acquire low-resolution grid concentration prediction data and downscaling indication data. The low-resolution grid concentration prediction data can include the concentration prediction results of each initial grid in the target area, and the resolution corresponding to each initial grid is lower than the resolution corresponding to each target grid (that is, the scale corresponding to each initial grid is larger than the scale corresponding to each target grid, and the larger the scale, the lower the resolution). The downscaling indication data can include the grid indication data of each target grid; and call the target statistical downscaling model to perform downscaling prediction based on the low-resolution grid concentration prediction data and the downscaling indication data to obtain the initial air quality forecast data, so as to realize the acquisition of the initial air quality forecast data, etc. Optionally, each initial grid can be the grid division result of the target area at a low resolution (such as 27 km or 9 km, etc.). That is to say, the resolution corresponding to each initial grid can be 27 km or 9 km, etc., and the embodiments of the present invention do not limit this. Optionally, the grid indication data of a target grid can include, but is not limited to, at least one of the following: the geographical information of the corresponding target grid (such as terrain height, land use type, etc.), emission information (such as population number, traffic flow information, emission rate, etc.), and meteorological information (such as temperature, humidity, wind speed, etc.), etc.; the embodiments of the present invention do not limit this. Based on this, the downscaling indication data can include, but is not limited to, at least one of the following: high-precision emissions (including the emission information of each target grid), high-precision geographical information (including the geographical information of each target grid), and high-precision meteorology (including the meteorological information of each target grid), etc., and the embodiments of the present invention do not limit this.

[0021] Optionally, when obtaining the low-resolution grid concentration prediction data, the low-resolution grid concentration prediction data may be stored in the storage space of the electronic device itself, so as to obtain the low-resolution grid concentration prediction data from the storage space of the electronic device itself; or, a download link for the low-resolution grid concentration prediction data may be obtained, so as to use the prediction data downloaded based on the download link for the low-resolution grid concentration prediction data as the low-resolution grid concentration prediction data; or, the electronic device may call a target air quality numerical model, and based on the data to be predicted, determine the low-resolution grid concentration prediction data, that is, the data to be predicted may be input into the target air quality numerical model to output the low-resolution grid concentration prediction data, so as to obtain the low-resolution grid concentration prediction data, and so on; the embodiments of the present invention do not limit this. Optionally, the target air quality numerical model may be NAQPMS (Nested Air Quality Prediction Modeling System), or CMAQ (an air quality prediction and evaluation system), or CAMx (an atmospheric pollutant calculation model based on atmospheric chemistry for ozone, particulate matter, etc.), or WRF-Chem (an online coupled model of atmospheric chemistry), and so on; the embodiments of the present invention do not limit this. Optionally, the data to be predicted may include, but is not limited to, at least one of the following: emission information, geographical information, meteorological information, and initial pollutant concentrations of each initial grid within a target time range (such as the next 7 days or 15 days, etc.), the embodiments of the present invention do not limit this; optionally, the target time range may include a target prediction time, and the target prediction time may be any prediction time (such as the first day in the future or a certain hour on the second day in the future, etc.), the embodiments of the present invention do not limit this. In the embodiments of the present invention, the low-resolution grid concentration prediction data, the initial air quality prediction data, the target air quality prediction data, the site ensemble prediction data, etc. may all include the corresponding values (such as concentration prediction results, initial pollutant concentration prediction values, target pollutant concentration prediction values, or site pollutant concentration prediction values, etc.) of the corresponding prediction objects (such as initial grids, target grids, or observation sites, etc.) at the target prediction time, that is to say, one data may correspond to one prediction time, and so on.

[0022] Correspondingly, when obtaining the downscaling indication data, the downscaling indication data may be stored in the storage space of the electronic device itself, and at this time, the downscaling indication data may be obtained from the storage space of the electronic device itself; or, a download link for the downscaling indication data may be obtained, so as to obtain the downscaling indication data based on the download link for the downscaling indication data, and so on. Optionally, the downscaling indication data may include the grid indication data of each target grid at the target prediction time.

[0023] Further, when invoking the target statistical downscaling model to perform downscaling prediction based on the low-resolution grid concentration prediction data and the downscaling indication data to obtain the initial air quality forecast data, the electronic device may input the low-resolution grid concentration prediction data and the downscaling indication data into the target statistical downscaling model to output the initial air quality forecast data; based on this, the initial air quality forecast data can be obtained through the statistical downscaling method. Optionally, the target statistical downscaling model can be a logistic regression model, a linear regression model, or any neural network model, etc.; the embodiments of the present invention do not limit this.

[0024] Optionally, the acquisition method of the site ensemble forecast data may include, but is not limited to, at least one of the following: The first acquisition method: The site ensemble forecast data is stored in the own storage space of the electronic device. In this case, the electronic device can obtain the site ensemble forecast data from its own storage space.

[0025] The second acquisition method: The electronic device can obtain the download link of the site ensemble forecast data to download the site ensemble forecast data based on the download link of the site ensemble forecast data.

[0026] The third acquisition method: The electronic device can obtain the data to be forecasted and respectively invoke each air quality numerical model in the air quality numerical model ensemble. Based on the data to be forecasted, determine the air quality forecast data under each air quality numerical model, that is, the data to be forecasted can be respectively input into each air quality numerical model to output the air quality forecast data under each air quality numerical model; then, the forecast data of the sites to be corrected under each air quality numerical model can be respectively determined from the air quality forecast data under each air quality numerical model; further, the deviation correction parameter set under each air quality numerical model can be determined, and based on the deviation correction parameter set under each air quality numerical model, the forecast data of the sites to be corrected under each air quality numerical model are respectively corrected to obtain the corrected site forecast data under each air quality numerical model; based on this, the site ensemble forecast data can be determined based on the corrected site forecast data under each air quality numerical model to achieve the acquisition of the site ensemble forecast data, etc. Optionally, the air quality numerical model ensemble may include, but is not limited to, at least one of the following: NAQPMS, CMAQ, CAMx, and WRF-Chem, etc.; the embodiments of the present invention do not limit this.

[0027] Among them, the forecast data of the station to be corrected under an air quality numerical model may include the forecast concentration values of the stations to be corrected of each observation station in the target area under the corresponding air quality numerical model. The air quality forecast data under an air quality numerical model may include the pollutant concentration forecast values (i.e., the pollutant concentration forecast values of the target pollutant) of each grid (such as the initial grid or the grid divided according to other resolutions) in the target area under the corresponding air quality numerical model. Optionally, when determining the forecast data of the stations to be corrected under each air quality numerical model from the air quality forecast data under each air quality numerical model, for any air quality numerical model in the set of air quality numerical models and any observation station in the target area, the electronic device may determine, from the air quality forecast data under any air quality numerical model, the pollutant concentration forecast value of the grid closest to any observation station as the forecast concentration value of the station to be corrected of any observation station under any air quality numerical model, so as to determine the forecast data of the stations to be corrected under any air quality numerical model from the air quality forecast data under any air quality numerical model; or, the pollutant concentration forecast values of the first P grids closest to any observation station may be determined for weighted summation, so as to obtain the forecast concentration value of the station to be corrected of any observation station under any air quality numerical model, where P is a positive integer, and so on; the embodiments of the present invention do not limit this. Optionally, an observation station may be a national control station, and so on.

[0028] Optionally, the set of bias correction parameters under each air quality numerical model can be set according to experience or actual requirements, or determined by historical forecast data and historical observation data under each air quality numerical model. The embodiments of the present invention do not limit this. Optionally, the historical forecast data under an air quality numerical model may include at least one historical concentration forecast value (i.e., the historical concentration forecast value of the target pollutant) of each observation site under the corresponding air quality numerical model, that is, it may include at least one historical concentration forecast value of each observation site under the corresponding air quality numerical model and at least one historical forecast time. Here, the time forecast indicators corresponding to each historical forecast time may be the same as the time forecast indicator corresponding to the target forecast time (a time forecast indicator can be used to indicate the interval duration of the forecast time. For example, the time forecast indicators can all be the second day in the future. At this time, a historical concentration forecast value can be the historical concentration forecast value obtained by forecasting the second day after any historical time at any past historical time, etc.). The historical observation data may include at least one historical concentration observation value of each observation site at at least one historical forecast time (that is, it may include at least one historical concentration observation value (i.e., the historical concentration observation value of the target pollutant) of each observation site). A historical concentration forecast value of an observation site corresponds to a historical concentration observation value of the corresponding observation site, that is, a historical concentration forecast value of an observation site at a historical forecast time corresponds to a historical concentration observation value of the corresponding observation site at the corresponding historical forecast time.

[0029] In one implementation, a set of bias correction parameters under an air quality numerical model may include the bias correction parameters of each observation station under the corresponding air quality numerical model, that is, may include the bias correction parameters of each observation station under the corresponding air quality numerical model and the time prediction index corresponding to the target prediction time. Then, when determining the set of bias correction parameters under each air quality numerical model based on the historical prediction data and historical observation data under each air quality numerical model, for any observation station in the target area and any air quality numerical model in the set of air quality numerical models, the electronic device may calculate the root mean square error of at least one historical concentration prediction value of any observation station under any air quality numerical model and at least one historical concentration observation value of any observation station, obtain the root mean square error of any observation station under any air quality numerical model, and determine the sign of the root mean square error of any observation station under any air quality numerical model (that is, the positive or negative sign, such as the sign is positive if the mean operation result is positive, and the sign is negative if the mean operation result is negative, etc.) based on the mean operation result between the differences between at least one historical concentration observation value of any observation station and the corresponding historical concentration prediction value of any observation station under any air quality numerical model, so as to determine the bias correction parameter of any observation station under any air quality numerical model (such as a positive root mean square error or a negative root mean square error, etc.) based on the root mean square error of any observation station under any air quality numerical model and the sign of the root mean square error of any observation station under any air quality numerical model; alternatively, the mean operation result between the differences between at least one historical concentration observation value of any observation station and the corresponding historical concentration prediction value of any observation station under any air quality numerical model may be used as the bias correction parameter of any observation station under any air quality numerical model; alternatively, the mean operation result between the ratios of at least one historical concentration observation value of any observation station and the corresponding historical concentration prediction value of any observation station under any air quality numerical model may be used as the bias correction parameter of any observation station under any air quality numerical model, etc.; the embodiments of the present invention do not limit this. In this case, the embodiments of the present invention can more accurately describe the bias correction parameters of different observation stations under different air quality numerical models.

[0030] In another embodiment, a set of bias correction parameters under an air quality numerical model may include bias correction parameters under the corresponding air quality numerical model, that is, it may include bias correction parameters under the corresponding air quality numerical model and the time prediction index corresponding to the target prediction time. At this time, the bias correction parameters of different observation stations under the same air quality numerical model are the same. Then, when determining the set of bias correction parameters under each air quality numerical model based on the historical prediction data and historical observation data under each air quality numerical model, for any air quality numerical model in the set of air quality numerical models, the electronic device may determine the bias correction parameters under any air quality numerical model based on at least one historical concentration prediction value of each observation station under any air quality numerical model and at least one historical concentration observation value of each observation station, so as to obtain the set of bias correction parameters under any air quality numerical model. For example, the mean operation result between the differences between at least one historical concentration observation value of each observation station and the corresponding historical concentration prediction value of the corresponding observation station under any air quality numerical model is used as the bias correction parameter under any air quality numerical model, and so on. The embodiments of the present invention do not limit this. In this case, the bias correction parameters of different observation stations under any air quality numerical model may all be the bias correction parameters under any air quality numerical model, that is, at this time, the bias correction parameters of different observation stations under any air quality numerical model may be the same.

[0031] Optionally, when correcting the forecast data of the stations to be corrected under each air quality numerical model by using the set of bias correction parameters under each air quality numerical model to obtain the corrected forecast data of the stations under each air quality numerical model, for any observation station in the target area and any air quality numerical model in the set of air quality numerical models, the electronic device may use the bias correction parameter of any observation station under any air quality numerical model to correct the concentration value of the forecast of the station to be corrected of any observation station under any air quality numerical model, so as to obtain the corrected concentration value of the forecast of the station to be corrected of any observation station under any air quality numerical model. The corrected forecast data of the stations under an air quality numerical model includes the corrected concentration values of the forecast of the stations of each observation station under the corresponding air quality numerical model. Optionally, when a bias correction parameter is used to indicate the difference between the historical concentration observation value and the historical concentration forecast value of an observation station (such as the bias correction parameter determined by the difference between the historical concentration observation value and the historical concentration forecast value), the concentration value of the forecast of the station to be corrected of any observation station under any air quality numerical model may be added with the bias correction parameter of any observation station under any air quality numerical model, so as to correct the concentration value of the forecast of the station to be corrected of any observation station under any air quality numerical model; or, when a bias correction parameter is used to indicate the difference between the historical concentration forecast value and the historical concentration observation value of an observation station (such as the bias correction parameter determined by the difference between the historical concentration forecast value and the historical concentration observation value), the concentration value of the forecast of the station to be corrected of any observation station under any air quality numerical model may be subtracted by the bias correction parameter of any observation station under any air quality numerical model, so as to correct the concentration value of the forecast of the station to be corrected of any observation station under any air quality numerical model; or, when a bias correction parameter is used to indicate the ratio between the historical concentration forecast value and the historical concentration observation value of an observation station (such as the bias correction parameter determined by the ratio between the historical concentration forecast value and the historical concentration observation value), the concentration value of the forecast of the station to be corrected of any observation station under any air quality numerical model may be divided by the bias correction parameter of any observation station under any air quality numerical model, so as to correct the concentration value of the forecast of the station to be corrected of any observation station under any air quality numerical model; or, when a bias correction parameter is used to indicate the ratio between the historical concentration observation value and the historical concentration forecast value of an observation station, the concentration value of the forecast of the station to be corrected of any observation station under any air quality numerical model may be multiplied by the bias correction parameter of any observation station under any air quality numerical model, so as to correct the concentration value of the forecast of the station to be corrected of any observation station under any air quality numerical model, and so on; the embodiments of the present invention do not limit this.

[0032] Optionally, when determining the site ensemble forecast data based on the corrected site forecast data under each air quality numerical model, for any observation site in the target area, the electronic device can determine the weights of each air quality numerical model, and perform weighted summation on the corrected site forecast concentration values of any observation site under each air quality numerical model according to the weights of each air quality numerical model, so as to obtain the site pollutant concentration forecast value of any observation site. Optionally, the weights of each air quality numerical model can be set according to experience or actual requirements, or can be determined by the historical forecast data and historical observation data under each air quality numerical model (the higher the accuracy of the historical forecast data under an air quality numerical model (such as the smaller the difference between the historical forecast data and the historical observation data under an air quality numerical model, or the higher the correlation with the historical observation data, etc.), the greater the weight of the corresponding air quality numerical model. At this time, the weight ratio of the better numerical model can be highlighted), etc.; the embodiments of the present invention do not limit this. Optionally, in other embodiments, the corrected site forecast concentration values of any observation site under each air quality numerical model can also be averaged to obtain the site pollutant concentration forecast value of any observation site, etc.; the present invention does not limit this. Based on this, the embodiments of the present invention can obtain site ensemble forecast data with relatively high accuracy, that is, the accuracy of the site pollutant concentration forecast value of each observation site can be effectively improved, so as to correct the initial air quality forecast data through the site pollutant concentration forecast value with relatively high accuracy.

[0033] Optionally, the number of initial air quality forecast data and site ensemble forecast data can both be one or more. One initial air quality forecast data corresponds to one site ensemble forecast data. That is to say, the number of target areas and target forecast times can both be one or more. The embodiments of the present invention do not limit this; for example, when the number of target forecast times is multiple, the target air quality forecast data of the target area at each target forecast time can be determined respectively, etc. For the convenience of description, in the following, one initial air quality forecast data and site ensemble forecast data, that is, one target area and one target forecast time are used as examples for description.

[0034] S102, based on the initial air quality forecast data, determine the first regional concentration forecast results of each sub-region in the target area, and based on the site ensemble forecast data, determine the second regional concentration forecast results of each sub-region.

[0035] Among them, the target area may include at least one sub - area. Optionally, a sub - area can be a city (such as a county or a city, etc.), or a region after dividing the target area according to a specified area division method, or the entire target area, etc.; the embodiments of the present invention do not limit this. Optionally, the specified area division method can be set according to experience or set according to actual needs, and the embodiments of the present invention do not limit this.

[0036] S103. Based on the first - area concentration prediction results and the second - area concentration prediction results of each sub - area respectively, determine the prediction deviation coefficient of each sub - area.

[0037] S104. Based on the prediction deviation coefficients of each sub - area, revise the initial air quality prediction data to obtain the target air quality prediction data.

[0038] In summary, the initial air quality prediction data can be determined by the target statistical downscaling model (that is, it can be determined by the statistical downscaling method). In this case, the embodiments of the present invention can implement a method for optimizing the downscaling results based on the statistical downscaling prediction results (that is, the initial air quality prediction data) and the station - ensemble prediction data. That is to say, the station - ensemble prediction data of the station type can be used to revise and optimize the initial air quality prediction data of the grid type, so that the optimized downscaling prediction results (that is, the following target air quality prediction data) not only retain the characteristics of high - precision spatial distribution but also have a relatively accurate pollutant concentration prediction level. And the core of the dynamic downscaling method (that is, by nesting a high - resolution sub - area in a coarse - resolution simulation area to refine the coarse - resolution results using numerical model technology) is to perform downscaling based on the numerical model method. Since the complexity of the atmospheric evolution equation characterized by the numerical model is high and the numerical solution process involves a large number of iterative calculations, this method consumes a large amount of computing resources. Based on this, the embodiments of the present invention can also obtain the initial air quality prediction data through the statistical downscaling method to effectively reduce the consumption of computing resources. However, although the initial air quality prediction data obtained by the statistical downscaling method is a grid - based result, its accuracy is low, and the station - ensemble prediction data is a result of the station type and cannot express the pollutant spatial distribution condition more precisely and comprehensively, but the accuracy of the station - ensemble prediction data is high. Therefore, the embodiments of the present invention can use the station - ensemble prediction data of the station type to revise and optimize the grid - based initial air quality prediction data obtained by downscaling to obtain an accurate and comprehensive grid - based downscaling result (that is, the target air quality prediction data), so as to obtain an accurate and comprehensive target air quality prediction data while ensuring low consumption of computing resources, that is, effectively improve the pollutant concentration prediction accuracy of the statistical downscaling prediction results.

[0039] Embodiments of the present invention can obtain initial air quality forecast data and obtain site ensemble forecast data. The initial air quality forecast data includes initial pollutant concentration forecast values for each target grid in the target area, and each target grid is the grid division result of the target area at a high resolution. Then, based on the initial air quality forecast data, the first regional concentration forecast results for each sub-region in the target area can be determined, and based on the site ensemble forecast data, the second regional concentration forecast results for each sub-region can be determined. Based on this, the forecast deviation coefficients for each sub-region can be determined respectively based on the first regional concentration forecast results and the second regional concentration forecast results of each sub-region; thus, based on the forecast deviation coefficients of each sub-region, the initial air quality forecast data can be corrected to obtain the target air quality forecast data. It can be seen that the embodiments of the present invention can determine the forecast deviation coefficients of each sub-region through site ensemble forecast data and other means, thereby correcting the initial air quality forecast data to obtain target air quality forecast data with higher accuracy, which can effectively improve the accuracy of high-resolution air quality forecasts, that is, the target air quality forecast data can not only retain the characteristics of high-precision spatial distribution but also have a relatively accurate pollutant concentration forecast level.

[0040] Based on the above description, embodiments of the present invention also propose a more specific air quality forecasting method. Correspondingly, this air quality forecasting method can be executed by the above-mentioned electronic device (terminal or server); or, this air quality forecasting method can be jointly executed by the terminal and the server. For the convenience of description, in the following, it will be described by taking the electronic device executing this air quality forecasting method as an example; please refer to Figure 2 , this air quality forecasting method may include the following steps S201 - S205: S201, obtain initial air quality forecast data and obtain site ensemble forecast data. The initial air quality forecast data includes initial pollutant concentration forecast values for each target grid in the target area, and each target grid is the grid division result of the target area at a high resolution.

[0041] S202, based on the initial air quality forecast data, determine the first regional concentration forecast results for each sub-region in the target area, and based on the site ensemble forecast data, determine the second regional concentration forecast results for each sub-region.

[0042] Optionally, when determining the first regional concentration prediction result of each sub-region in the target area based on the initial air quality prediction data, for any sub-region in the target area, the electronic device may determine the site prediction result of each observation site within any sub-region based on the initial air quality prediction data. Optionally, for any observation site within any sub-region, the initial pollutant concentration prediction value of a target grid closest to any observation site may be determined from the initial air quality prediction data, and the initial pollutant concentration prediction value of the target grid closest to any observation site may be used as the site prediction result of any observation site; alternatively, the initial pollutant concentration prediction values of the first Q target grids closest to any observation site may be determined from the initial air quality prediction data, and the weighted sum of the initial pollutant concentration prediction values of the first Q target grids closest to any observation site may be calculated, so that the weighted sum result is used as the site prediction result of any observation site. Q is a positive integer, and the weight of the initial pollutant concentration prediction value of a target grid may be negatively correlated with the distance between the corresponding target grid and any observation site, etc.; the embodiments of the present invention do not limit this.

[0043] Based on this, the electronic device may use the site prediction results of the above-mentioned each observation site (i.e., each observation site within any sub-region) to calculate the first regional concentration prediction result of any sub-region. Optionally, the electronic device may use the mean value between the site prediction results of each observation site within any sub-region as the first regional concentration prediction result of any sub-region; for example, taking the target pollutant as PM2.5 and any sub-region as a city as an example for illustration, assuming there are 10 observation sites within any sub-region (respectively represented as ST1, ST2,..., ST10), and the site prediction results of each observation site within any sub-region are respectively represented as PM2.5 SG1KM-ST1 、PM2.5 SG1KM-ST2 、…、PM2.5 SG1KM-ST10 , then the first regional concentration prediction result PM2.5 SG1KM-CITY = (PM2.5 SG1KM-ST1 + PM2.5 SG1KM-ST2 + … + PM2.5 SG1KM-ST10 ) / 10. Optionally, the electronic device may also determine the weights of each observation site within any sub-region, and perform a weighted sum of the site prediction results of each observation site within any sub-region according to the weights of each observation site within any sub-region to obtain the first regional concentration prediction result of any sub-region, etc.; optionally, the weights of each observation site within any sub-region may be set according to experience or according to time requirements, and the embodiments of the present invention do not limit this.

[0044] In an embodiment of the present invention, the site ensemble forecast data may include the site pollutant concentration forecast values of each observation site in the target area, that is, the site pollutant concentration forecast values of each observation site at the target forecast time; based on this, when determining the second area concentration forecast results of each sub-area based on the site ensemble forecast data, for any sub-area in the target area, the electronic device may determine the site pollutant concentration forecast values of each observation site within any sub-area from the site ensemble forecast data; and use the site pollutant concentration forecast values of each observation site to calculate the second area concentration forecast result of any sub-area. Optionally, the electronic device may use the mean value between the site pollutant concentration forecast values of each observation site within any sub-area as the second area concentration forecast result of any sub-area, that is, perform a mean operation on the site pollutant concentration forecast values of each observation site within any sub-area to obtain the second area concentration forecast result of any sub-area; by way of example, taking the target pollutant as PM2.5 for illustration, assuming that the site pollutant concentration forecast values of each observation site within any sub-area are PM2.5 OEF-ST1 , PM2.5 OEF-ST2 , …, PM2.5 OEF-ST10 , then the second area concentration forecast result PM2.5 OEF-CITY = (PM2.5 OEF-ST1 + PM2.5 OEF-ST2 + … + PM2.5 OEF-ST10 ) / 10. Optionally, the electronic device may also perform a weighted sum on the site pollutant concentration forecast values of each observation site within any sub-area according to the weights of each observation site within any sub-area to obtain the second area concentration forecast result of any sub-area, and so on. Among them, OEF may represent Optimal Ensemble Forecast, to mark the ensemble forecast; optionally, the weights of the above-mentioned various air quality numerical models may be determined based on the historical forecast data and historical observation data under each air quality numerical model through the OEF algorithm to highlight the weight ratio of the relatively better numerical models, and so on.

[0045] S203. Respectively determine the forecast deviation coefficients of each sub-area based on the first area concentration forecast results and the second area concentration forecast results of each sub-area.

[0046] In an embodiment of the present invention, for the mth sub-region in the target area, the electronic device may use the ratio between the second area concentration forecast result of the mth sub-region and the first area concentration forecast result of the mth sub-region as the forecast deviation coefficient of the mth sub-region, where m∈[1,M], and M is the number of sub-regions in the target area. For example, taking PM2.5 as an example, assuming that the second area concentration forecast result of the mth sub-region is PM2.5 OEF-CITY , the first area concentration forecast result of the mth sub-area is PM2.5 SG1KM-CITY , then the forecast deviation coefficient B of the mth sub-region is PM2.5-CITY =PM2.5 OEF-CITY / PM2.5 SG1KM-CITY .

[0047] Optionally, in other embodiments, the electronic device may also use the ratio between the first area concentration forecast result of the mth sub-area and the second area concentration forecast result of the mth sub-area as the forecast deviation coefficient of the mth sub-area, and so on; the present invention is not limited to this.

[0048] S204, for any target grid in the target area, determine the forecast deviation coefficient of the sub-area where any target grid is located from the forecast deviation coefficients of each sub-area.

[0049] S205, using the forecast deviation coefficient of the sub-region where any target grid is located, correct the initial pollutant concentration forecast value of any target grid to obtain the target pollutant concentration forecast value of any target grid, and the target air quality forecast data includes the target pollutant concentration forecast values of each target grid.

[0050] In the embodiment of the present invention, the electronic device may use the multiplication result between the forecast deviation coefficient of the sub-region where any target grid is located and the initial pollutant concentration forecast value of any target grid as the target pollutant concentration forecast value of any target grid, so as to correct the initial pollutant concentration forecast value of any target grid. For example, assuming that any target grid G A The sub-region is CITY1 (a sub-region is a city at this time), and the forecast deviation coefficient of this sub-region is B PM2.5-CITY1 , then any target grid G A The corrected result (i.e. the predicted value of the target pollutant concentration) G A-rev =G A ×B PM2.5-CITY1 In this case, the forecast deviation coefficient of the m-th sub-region may be: the ratio between the second-region concentration forecast result of the m-th sub-region and the first-region concentration forecast result of the m-th sub-region.

[0051] Optionally, in other embodiments, when the forecast deviation coefficient of the m-th sub-region is the ratio between the first regional concentration forecast result and the second regional concentration forecast result of the m-th sub-region, the division operation result between the initial pollutant concentration forecast value of any target grid and the forecast deviation coefficient of the sub-region where the target grid is located can be used as the target pollutant concentration forecast value of any target grid. That is, at this time, the target pollutant concentration forecast value of any target grid can be the result of dividing the initial pollutant concentration forecast value of any target grid by the forecast deviation coefficient of the sub-region where the target grid is located, and so on. The present invention does not limit this.

[0052] Optionally, the target air quality forecast data may include the target pollutant concentration forecast values of each target grid at the target forecast time. That is, the target pollutant concentration forecast value of the above-mentioned one target grid may refer to the target pollutant concentration forecast value of the corresponding target grid at the target forecast time.

[0053] Based on this, embodiments of the present invention can determine the inclusion relationship between each target grid and the sub-region according to the grid position, and then perform deviation correction on each target grid respectively, so as to correct the initial air quality forecast data to obtain the target pollutant concentration forecast values of each target grid, as Figure 3 shown. Correspondingly, when a sub-region is a city, embodiments of the present invention can finally obtain a grid-based high-precision downscaling forecast result that not only maintains the rationality of spatial distribution but also has a high overall forecast accuracy for the city.

[0054] Embodiments of the present invention can obtain initial air quality forecast data and obtain site ensemble forecast data. The initial air quality forecast data includes initial pollutant concentration forecast values for each target grid in the target area, and each target grid is the grid division result of the target area at a high resolution. Then, based on the initial air quality forecast data, the first regional concentration forecast results for each sub-region in the target area can be determined, and based on the site ensemble forecast data, the second regional concentration forecast results for each sub-region can be determined. Based on this, the forecast deviation coefficients for each sub-region can be determined respectively based on the first regional concentration forecast results and the second regional concentration forecast results of each sub-region. Further, for any target grid in the target area, the forecast deviation coefficient of the sub-region where any target grid is located can be determined from the forecast deviation coefficients of each sub-region; and the initial pollutant concentration forecast value of any target grid can be corrected using the forecast deviation coefficient of the sub-region where any target grid is located to obtain the target pollutant concentration forecast value of any target grid. The target air quality forecast data includes the target pollutant concentration forecast values for each target grid. It can be seen that the embodiments of the present invention can effectively improve the accuracy of the grid-based target air quality forecast data, and the target air quality forecast data with a higher spatial resolution (i.e., a finer spatial scale) can enable air pollution prevention and control personnel to have a more detailed understanding of the distribution of pollutants, and thus can support more precise prevention and control measures for high-pollution areas.

[0055] Based on the description of the related embodiments of the above air quality forecast method, embodiments of the present invention also propose an air quality forecast device. The air quality forecast device can be a computer program (including program code) running in an electronic device; as Figure 4 shown, the air quality forecast device can include an acquisition unit 401 and a processing unit 402. The air quality forecast device can execute Figure 1 or Figure 2 the air quality forecast method shown, that is, the air quality forecast device can run the above units: The acquisition unit 401 is used to acquire initial air quality forecast data and acquire site ensemble forecast data. The initial air quality forecast data includes initial pollutant concentration forecast values for each target grid in the target area, and each target grid is the grid division result of the target area at a high resolution; The processing unit 402 is used to determine the first regional concentration forecast results for each sub-region in the target area based on the initial air quality forecast data, and determine the second regional concentration forecast results for each sub-region based on the site ensemble forecast data; The processing unit 402 is further used to determine the forecast deviation coefficients for each sub-region respectively based on the first regional concentration forecast results and the second regional concentration forecast results of each sub-region; The processing unit 402 is further configured to correct the initial air quality forecast data based on the forecast deviation coefficients of the respective sub-regions to obtain target air quality forecast data.

[0056] In one implementation manner, when the processing unit 402 determines the forecast deviation coefficients of the respective sub-regions based on the first regional concentration forecast results and the second regional concentration forecast results of the respective sub-regions, it may specifically be configured to: For the m-th sub-region in the target region, the ratio between the second regional concentration forecast result of the m-th sub-region and the first regional concentration forecast result of the m-th sub-region is used as the forecast deviation coefficient of the m-th sub-region, where m ∈ [1, M], and M is the number of sub-regions in the target region.

[0057] In another implementation manner, when the processing unit 402 determines the first regional concentration forecast results of the respective sub-regions in the target region based on the initial air quality forecast data, it may specifically be configured to: For any sub-region in the target region, based on the initial air quality forecast data, determine the site prediction results of each observation site within the any sub-region; Using the site prediction results of the respective observation sites, calculate the first regional concentration forecast result of the any sub-region.

[0058] In another implementation manner, the site ensemble forecast data includes the site pollutant concentration forecast values of each observation site in the target region; when the processing unit 402 determines the second regional concentration forecast results of the respective sub-regions based on the site ensemble forecast data, it may specifically be configured to: For any sub-region in the target region, determine the site pollutant concentration forecast values of each observation site within the any sub-region from the site ensemble forecast data; Using the site pollutant concentration forecast values of the respective observation sites, calculate the second regional concentration forecast result of the any sub-region.

[0059] In another implementation manner, when the processing unit 402 corrects the initial air quality forecast data based on the forecast deviation coefficients of the respective sub-regions to obtain target air quality forecast data, it may specifically be configured to: For any target grid in the target region, determine the forecast deviation coefficient of the sub-region where the any target grid is located from the forecast deviation coefficients of the respective sub-regions; Using the forecast deviation coefficient of the sub-region where any of the target grids is located, correct the initial pollutant concentration forecast value of any of the target grids to obtain the target pollutant concentration forecast value of any of the target grids, and the target air quality forecast data includes the target pollutant concentration forecast values of each of the target grids.

[0060] In another embodiment, when the obtaining unit 401 obtains the initial air quality forecast data, it may specifically be used for: Obtain low-resolution grid concentration prediction data and obtain downscaling indication data, where the low-resolution grid concentration prediction data includes the concentration prediction results of each initial grid in the target area, the resolution corresponding to each initial grid is lower than the resolution corresponding to each target grid, and the downscaling indication data includes the grid indication data of each target grid; Call a target statistical downscaling model, and perform downscaling prediction based on the low-resolution grid concentration prediction data and the downscaling indication data to obtain the initial air quality forecast data.

[0061] In another embodiment, when the obtaining unit 401 obtains the site ensemble forecast data, it may specifically be used for: Obtain the data to be forecasted, and respectively call each air quality numerical model in the air quality numerical model ensemble, and based on the data to be forecasted, determine the air quality forecast data under each air quality numerical model; Respectively determine the site forecast data to be corrected under each air quality numerical model from the air quality forecast data under each air quality numerical model; Determine the deviation correction parameter set under each air quality numerical model, and respectively correct the site forecast data to be corrected under each air quality numerical model based on the deviation correction parameter set under each air quality numerical model to obtain the corrected site forecast data under each air quality numerical model; Based on the corrected site forecast data under each air quality numerical model, determine the site ensemble forecast data.

[0062] According to an embodiment of the present invention, Figure 4Each unit in the air quality forecasting device shown can be separately or all combined into one or several other units to form, or a certain one (or some) of the units can be further split into multiple smaller units in terms of function to form, which can achieve the same operations without affecting the realization of the technical effects of the embodiments of the present invention. The above units are divided based on logical functions. In practical applications, the function of one unit can also be realized by multiple units, or the functions of multiple units can be realized by one unit. In other embodiments of the present invention, any air quality forecasting device can also include other units. In practical applications, these functions can also be assisted by other units and can be realized by the cooperation of multiple units.

[0063] According to another embodiment of the present invention, it can be achieved by running a computer program (including program code) capable of executing the respective steps involved in the corresponding method shown in Figure 1 or Figure 2 on a general-purpose electronic device such as a computer including processing elements and storage elements such as a central processing unit (CPU), a random access storage medium (RAM), a read-only storage medium (ROM), etc., to construct an air quality forecasting device as shown in Figure 4 and to implement the air quality forecasting method of the embodiments of the present invention. The computer program can be recorded on, for example, a computer storage medium, loaded into the above-mentioned electronic device through the computer storage medium, and run therein.

[0064] The embodiments of the present invention can obtain initial air quality forecasting data and obtain site set forecasting data. The initial air quality forecasting data includes the initial pollutant concentration forecasting values of each target grid in the target area, and each target grid is the grid division result of the target area at a high resolution. Then, based on the initial air quality forecasting data, the first area concentration forecasting results of each sub-area in the target area can be determined, and based on the site set forecasting data, the second area concentration forecasting results of each sub-area can be determined. Based on this, the forecasting deviation coefficients of each sub-area can be determined respectively based on the first area concentration forecasting results and the second area concentration forecasting results of each sub-area; thus, based on the forecasting deviation coefficients of each sub-area, the initial air quality forecasting data can be corrected to obtain the target air quality forecasting data. It can be seen that the embodiments of the present invention can determine the forecasting deviation coefficients of each sub-area through the site set forecasting data, etc., so as to correct the initial air quality forecasting data to obtain the target air quality forecasting data with higher accuracy, which can effectively improve the accuracy of high-resolution air quality forecasting, that is, the target air quality forecasting data can not only retain the characteristics of high-precision spatial distribution but also have a relatively accurate pollutant concentration forecasting level.

[0065] Based on the descriptions of the above method embodiments and apparatus embodiments, an exemplary embodiment of the present invention further provides an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor. The memory stores a computer program executable by the at least one processor, and when executed by the at least one processor, the computer program is used to cause the electronic device to execute the method according to the embodiments of the present invention.

[0066] An exemplary embodiment of the present invention further provides a non-transitory computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor of a computer, the computer is caused to execute the method according to the embodiments of the present invention.

[0067] An exemplary embodiment of the present invention further provides a computer program product, including a computer program, wherein when the computer program is executed by a processor of a computer, the computer is caused to execute the method according to the embodiments of the present invention.

[0068] Referring to Figure 5 , a block diagram of an electronic device 500 that can be a server or a client of the present invention will now be described. It is an example of a hardware device that can be applied to various aspects of the present invention. The electronic device is intended to represent various forms of digital electronic computer devices, such as, a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, a personal digital processor, a cellular phone, a smart phone, a wearable device, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0069] As Figure 5 shown, the electronic device 500 includes a computing unit 501, which can execute various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the electronic device 500 can also be stored. The computing unit 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0070] Multiple components in the electronic device 500 are connected to the I / O interface 505, including: an input unit 506, an output unit 507, a storage unit 508, and a communication unit 509. The input unit 506 can be any type of device capable of inputting information into the electronic device 500. The input unit 506 can receive input digital or character information and generate key signal inputs related to the user settings and / or function controls of the electronic device. The output unit 507 can be any type of device capable of presenting information and can include, but is not limited to, a display, a speaker, a video / audio output terminal, a vibrator, and / or a printer. The storage unit 508 can include, but is not limited to, magnetic disks and optical discs. The communication unit 509 allows the electronic device 500 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks and can include, but is not limited to, a modem, a network card, an infrared communication device, a wireless communication transceiver, and / or a chipset, such as a BluetoothTM device, a WiFi device, a WiMax device, a cellular communication device, and / or the like.

[0071] The computing unit 501 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 501 executes the various methods and processes described above. For example, in some embodiments, the air quality forecasting method can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 508. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 500 via the ROM 502 and / or the communication unit 509. In some embodiments, the computing unit 501 can be configured to execute the air quality forecasting method in any other suitable manner (e.g., by means of firmware).

[0072] The program code for implementing the method of the present invention can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the program codes are executed by the processor or controller, the functions / operations specified in the flowchart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, executed partially on the machine as an independent software package and partially on a remote machine, or executed entirely on a remote machine or server.

[0073] In the context of the present invention, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0074] As used in the present invention, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, apparatus, and / or device (e.g., a disk, an optical disk, a memory, a programmable logic device (PLD)) used to provide machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term "machine-readable signal" refers to any signal used to provide machine instructions and / or data to a programmable processor.

[0075] In order to provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0076] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected to each other by digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), and the Internet.

[0077] A computer system can include clients and servers. The clients and servers are generally remote from each other and typically interact through a communication network. The client-server relationship is created by computer programs that run on the respective computers and have a client-server relationship with each other.

[0078] Also, it should be understood that the above-disclosed is only a preferred embodiment of the present invention, and of course it cannot be used to limit the scope of the rights of the present invention. Therefore, equivalent changes made according to the claims of the present invention still fall within the scope covered by the present invention.

Claims

1. An air quality forecasting method, characterized in that, Including: Obtaining initial air quality forecast data and obtaining site ensemble forecast data, where the initial air quality forecast data includes initial pollutant concentration forecast values for each target grid in the target area, and each of the target grids is the grid division result of the target area at a high resolution; Based on the initial air quality forecast data, determining first regional concentration forecast results for each sub-region in the target area, and based on the site ensemble forecast data, determining second regional concentration forecast results for each sub-region; Respectively based on the first regional concentration forecast results and the second regional concentration forecast results of each sub-region, determining a forecast deviation coefficient for each sub-region; Based on the forecast deviation coefficients of each sub-region, correcting the initial air quality forecast data to obtain target air quality forecast data.

2. The method according to claim 1, wherein The respectively based on the first regional concentration forecast results and the second regional concentration forecast results of each sub-region to determine the forecast deviation coefficient for each sub-region includes: For the m-th sub-region in the target area, taking the ratio between the second regional concentration forecast result of the m-th sub-region and the first regional concentration forecast result of the m-th sub-region as the forecast deviation coefficient of the m-th sub-region, where m ∈ [1, M], and M is the number of sub-regions in the target area.

3. The method according to claim 1 or 2, characterized in that The based on the initial air quality forecast data to determine the first regional concentration forecast results for each sub-region in the target area includes: For any sub-region in the target area, based on the initial air quality forecast data, determining site prediction results for each observation site within the any sub-region; Using the site prediction results of each observation site to calculate the first regional concentration forecast result of the any sub-region.

4. The method according to claim 1 or 2, characterized in that, The site ensemble forecast data includes site pollutant concentration forecast values for each observation site in the target area; the based on the site ensemble forecast data to determine the second regional concentration forecast results for each sub-region includes: For any sub-region in the target area, determining the site pollutant concentration forecast values for each observation site within the any sub-region from the site ensemble forecast data; Using the site pollutant concentration forecast values of each observation site to calculate the second regional concentration forecast result of the any sub-region.

5. The method according to claim 1 or 2, characterized in that, The based on the forecast deviation coefficients of each sub-region to correct the initial air quality forecast data to obtain target air quality forecast data includes: For any target grid in the target area, determining the forecast deviation coefficient of the sub-region where the any target grid is located from the forecast deviation coefficients of each sub-region; Using the forecast deviation coefficient of the sub-region where the any target grid is located to correct the initial pollutant concentration forecast value of the any target grid to obtain the target pollutant concentration forecast value of the any target grid, and the target air quality forecast data includes the target pollutant concentration forecast values for each target grid.

6. The method according to claim 1 or 2, characterized in that, The obtaining the initial air quality forecast data includes: Obtain low-resolution grid concentration prediction data and obtain downscaling indication data. The low-resolution grid concentration prediction data includes the concentration prediction results of each initial grid in the target area, and the resolution corresponding to each initial grid is lower than the resolution corresponding to each target grid. The downscaling indication data includes the grid indication data of each target grid. Call a target statistical downscaling model, and perform downscaling prediction based on the low-resolution grid concentration prediction data and the downscaling indication data to obtain initial air quality forecast data.

7. The method according to claim 1 or 2, characterized in that, The obtaining of the station ensemble forecast data includes: Obtain the data to be forecast, and respectively call each air quality numerical model in the air quality numerical model ensemble. Based on the data to be forecast, determine the air quality forecast data under each air quality numerical model. Respectively determine the station forecast data to be corrected under each air quality numerical model from the air quality forecast data under each air quality numerical model. Determine the set of bias correction parameters under each air quality numerical model, and respectively correct the station forecast data to be corrected under each air quality numerical model based on the set of bias correction parameters under each air quality numerical model to obtain the corrected station forecast data under each air quality numerical model. Based on the corrected station forecast data under each air quality numerical model, determine the station ensemble forecast data.

8. An air quality forecasting device, characterized in that, The device includes: An obtaining unit, configured to obtain initial air quality forecast data and obtain station ensemble forecast data. The initial air quality forecast data includes the initial pollutant concentration forecast values of each target grid in the target area, and each target grid is the grid division result of the target area at high resolution. A processing unit, configured to determine the first regional concentration forecast result of each sub-region in the target area based on the initial air quality forecast data, and determine the second regional concentration forecast result of each sub-region based on the station ensemble forecast data. The processing unit is further configured to respectively determine the forecast deviation coefficient of each sub-region based on the first regional concentration forecast result and the second regional concentration forecast result of each sub-region. The processing unit is further configured to correct the initial air quality forecast data based on the forecast deviation coefficient of each sub-region to obtain the target air quality forecast data.

9. An electronic device, characterized in that, Includes: A processor; And A memory storing a program, wherein the program includes instructions that, when executed by the processor, cause the processor to execute the method according to any one of claims 1-7.

10. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause a computer to execute the method according to any one of claims 1-7.

Images