Weather forecast downscaling correction method and device based on generative model

Abstract

The invention provides a weather forecast downscaling correction method and device based on a generative model, relates to the technical field of data processing, and solves the technical problem of low accuracy of weather forecast downscaling data. The method comprises the following steps: generating a low-resolution forecasting sample and a high-resolution live analysis sample based on weather live analysis data and weather forecast historical data; determining the high-resolution live analysis sample as a first-stage training data set for a variational auto-encoder, and determining the low-resolution prediction sample and a specified live constraint condition as a second-stage training data set for a diffusion model; constructing a downscaling correction model based on the diffusion model and the variational auto-encoder; and performing first-stage model training on the variational auto-encoder part by using the first-stage training data set to obtain a pre-training network, and performing second-stage model training on the diffusion model part by using the second-stage training data set based on the pre-training network to obtain a trained target downscaling correction model.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method and device for downscaling and correcting weather forecasts based on a generative model. Background Art

[0002] At present, the way to downscale low-resolution weather forecasts is based on the convolutional neural network downscaling method, represented by SRCNN (Super-Resolution Convolution Neural Network), which directly learns the mapping relationship from low-resolution (LR) images to high-resolution (HR) images. The loss function is the mean square error, and what is optimized is the pixel-level difference between low-resolution and high-resolution images. In the fields of meteorology, oceanography and other earth sciences, this type of image super-resolution algorithm is used to establish the relationship between two meteorological data products with different resolutions to achieve the purpose of downscaling low-resolution weather forecasts.

[0003] However, the above-mentioned direct training model learns the pixel differences from the low-resolution weather forecast grid field to the high-resolution actual reanalysis. The reconstructed image is prone to be too smooth and lacks high-frequency details and authenticity, resulting in low data accuracy of the existing weather forecast downscaling method. Summary of the invention

[0004] The purpose of the present invention is to provide a weather forecast downscaling correction method and device based on a generative model to solve the technical problem of low data accuracy of existing weather forecast downscaling methods.

[0005] In a first aspect, the present application provides a method for downscaling and correcting meteorological forecasts based on a generative model, the method comprising: Acquire meteorological real-time analysis data and meteorological forecast historical data, and generate low-resolution forecast samples and high-resolution real-time analysis samples based on the meteorological real-time analysis data and the meteorological forecast historical data; wherein the resolution of the high-resolution real-time analysis samples is higher than that of the low-resolution forecast samples; Determine the high-resolution live analysis samples as a first-stage training data set for a variational autoencoder, and determine the low-resolution forecast samples and specified live constraints as a second-stage training data set for a diffusion model, wherein the training labels corresponding to the second-stage training data set are the high-resolution live analysis samples; A downscaling correction model is constructed based on the diffusion model and the variational autoencoder; wherein the variational autoencoder part in the downscaling correction model is used to encode and decode meteorological variables in a specified low-dimensional latent space, and the diffusion model part in the downscaling correction model is used to refine the preliminary results generated by the variational autoencoder part to improve the high-resolution correction results; Using the first-stage training data set to perform first-stage model training on the variational autoencoder part to obtain a pre-trained network, and based on the pre-trained network, using the second-stage training data set to perform second-stage model training on the diffusion model part to obtain a trained target downscaling correction model; Based on the weather forecast data to be corrected and the specified actual constraints, downscaling correction processing is performed using the target downscaling correction model to obtain corrected high-resolution weather forecast data; wherein the resolution of the corrected high-resolution weather forecast data is higher than that of the weather forecast data to be corrected.

[0006] In one possible implementation, the loss function corresponding to the variational autoencoder part includes a reconstruction loss function and a KL divergence loss function; wherein the reconstruction loss function is used to measure the gap between the reconstructed data and the real high-resolution data corresponding to the high-resolution live analysis sample, and the KL divergence loss function is used to control the closeness of the potential distribution to the standard normal distribution N(0, I).

[0007] In one possible implementation, the diffusion process corresponding to the diffusion model part includes a forward process and a reverse process; wherein the forward process includes gradually adding noise to the preliminary result generated by the variational autoencoder part, and the reverse process includes learning to gradually denoise the noise back to the high-resolution corrected result corresponding to the high-resolution live analysis sample.

[0008] In a possible implementation, the construction diffusion process corresponding to the forward process is performed by the following formula: ; in, is the forward process at time step t The data is used to represent the intermediate state to which a certain amount of noise has been added; is the forward process at time step t -1 data is used to represent the output of the previous step; is the noise attenuation coefficient in the forward process, which is used to determine the mixing ratio of the original data and the noise; is the standard normal distribution noise; The weights to hold for the data; is the weight of the noise.

[0009] In a possible implementation, in the reverse process, the following formula is used to learn how to gradually denoise the noise and return to the high-resolution correction result: ; in, t is the time step, is the noise state at time step t, is the noise state closer to the real data at time step t-1, is the probability distribution data, N () represents the Gaussian distribution form, represents the center value of the data after denoising, is the mean, is the covariance matrix, which is used to represent the control generation Diversity of time.

[0010] In a possible implementation, the input conditions corresponding to the diffusion model part include: taking additional conditional variables as model inputs in each step so that the model generation results meet the physical constraints.

[0011] In a possible implementation, the additional condition variables include any one or more of the following: terrain, climate zone information, and recent actual conditions.

[0012] In a possible implementation, the obtaining of weather forecast historical data includes: Acquire ocean current data at multiple meteorological data collection locations at different time points in an ocean area; wherein the ocean current data includes at least one of sea water temperature, air temperature, wind speed, and water flow speed; Comparing the plurality of ocean current data corresponding to the plurality of meteorological data collection location points to obtain a comparison result, and determining at least two groups of target ocean current data having a data similarity greater than a preset similarity according to the comparison result; Determine the target meteorological data collection location points and the corresponding target time points corresponding to the at least two groups of target ocean current data, and determine the ocean current dynamic data based on at least two of the target meteorological data collection location points and at least two of the target time points; wherein the ocean current dynamic data includes the dynamic direction and movement time difference of the target ocean current between at least two of the target meteorological data collection location points; Based on the ocean current dynamics data, historical weather forecast data for the ocean area is determined.

[0013] In a second aspect, the present application provides a weather forecast downscaling correction device based on a generative model, the device comprising: An acquisition module, used to acquire meteorological real-time analysis data and meteorological forecast historical data, and generate low-resolution forecast samples and high-resolution real-time analysis samples based on the meteorological real-time analysis data and the meteorological forecast historical data; wherein the resolution of the high-resolution real-time analysis samples is higher than that of the low-resolution forecast samples; A determination module, configured to determine the high-resolution live analysis samples as a first-stage training data set for a variational autoencoder, and determine the low-resolution prediction samples and the specified live constraint conditions as a second-stage training data set for a diffusion model, wherein the training label corresponding to the second-stage training data set is the high-resolution live analysis sample; A construction module is used to construct a downscaling correction model based on the diffusion model and the variational autoencoder; wherein the variational autoencoder part in the downscaling correction model is used to encode and decode meteorological variables in a specified low-dimensional latent space, and the diffusion model part in the downscaling correction model is used to refine the preliminary results generated by the variational autoencoder to improve the high-resolution correction results; A training module, configured to perform first-stage model training on the variational autoencoder part using the first-stage training data set to obtain a pre-trained network, and perform second-stage model training on the diffusion model part based on the pre-trained network using the second-stage training data set to obtain a trained target downscaling correction model; A correction module is used to perform downscaling correction processing using the target downscaling correction model based on the weather forecast data to be corrected and the specified actual constraints to obtain corrected high-resolution weather forecast data; wherein the resolution of the corrected high-resolution weather forecast data is higher than that of the weather forecast data to be corrected.

[0014] In a third aspect, the present application provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, and when the processor executes the computer program, the method described in the first aspect is implemented.

[0015] In a fourth aspect, the present application further provides a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to execute the method described in the first aspect.

[0016] This application brings the following beneficial effects: The present application provides a weather forecast downscaling correction method and device based on a generative model, which can obtain meteorological actual analysis data and meteorological forecast historical data, and generate low-resolution forecast samples and high-resolution actual analysis samples based on the meteorological actual analysis data and the meteorological forecast historical data; wherein the resolution of the high-resolution actual analysis samples is higher than that of the low-resolution forecast samples; the high-resolution actual analysis samples are determined as the first-stage training data set for a variational autoencoder, and the low-resolution forecast samples and specified actual constraints are determined as the second-stage training data set for a diffusion model, wherein the training labels corresponding to the second-stage training data set are the high-resolution actual analysis samples; a downscaling correction model is constructed based on the diffusion model and the variational autoencoder; wherein the variational autoencoder in the downscaling correction model The encoder part is used to encode and decode the meteorological variables in the specified low-dimensional latent space, and the diffusion model part in the downscaling correction model is used to refine the preliminary results generated by the variational autoencoder part to improve the high-resolution correction results; the variational autoencoder part is trained with the first-stage training data set to obtain a pre-trained network, and the diffusion model part is trained with the second-stage training data set based on the pre-trained network to obtain a trained target downscaling correction model; based on the meteorological forecast data to be corrected and the specified actual constraints, the target downscaling correction model is used to perform downscaling correction processing to obtain corrected high-resolution meteorological forecast data; wherein the resolution of the corrected high-resolution meteorological forecast data is higher than that of the meteorological forecast data to be corrected. In this scheme, not only the diffusion model is used to gradually generate real high-resolution meteorological data from noise, but also the real-time reanalysis data is used to constrain the downscaling process, thereby achieving the downscaling and correction of meteorological forecasts at the same time. Not only can the low-resolution forecast be downscaled, but it can also ensure that the details generated by the model are close to the real situation rather than forged details, thereby improving the data resolution and details, thereby improving the accuracy of the downscaled meteorological forecast data, and solving the technical problem of low data accuracy of the existing meteorological forecast downscaling method.

[0017] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the specific implementation methods of the present application or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 A schematic diagram of a flow chart of a weather forecast downscaling correction method based on a generative model provided in an embodiment of the present application; Figure 2 A schematic diagram of a method for constructing a revised model for weather forecasting based on a mobile observation point provided in an embodiment of the present application; Figure 3 A schematic diagram of the structure of a weather forecast downscaling correction device based on a generative model provided in an embodiment of the present application; Figure 4 A schematic structural diagram of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution of the present application will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application.

[0021] The terms "including" and "having" and any variations thereof mentioned in the embodiments of the present application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device including a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or devices.

[0022] At present, the way to downscale low-resolution weather forecasts also includes downscaling methods based on generative adversarial networks, including SRGAN (Super-Resolution Generative Adversarial Network) and ESRGAN (Enhanced Super-Resolution Generative Adversarial Network). Basically, the generator and discriminator of the generative adversarial network (GAN) are trained in an adversarial way. The generator is responsible for generating realistic high-resolution images, while the discriminator is used to distinguish between generated images and real images. This type of method focuses on optimizing image and perceptual quality rather than simple pixel-level errors, so it can generate high-frequency details, and the image is sharper and closer to human eye perception. In visual performance, due to the traditional pixel-based model. In the existing technology, such methods can be used for downscaling of meteorological data, but such methods are difficult to train, and the adversarial loss may introduce forged details that are inconsistent with the actual situation.

[0023] The above-mentioned downscaling method based on generative adversarial networks has many technical problems, including: the training process of generative adversarial networks using image data is already very unstable, and the loss function is more likely to become a nan value when using meteorological data for training, that is, the gradient disappears or explodes; the adversarial loss of the generative adversarial network is prone to introduce fake details. In this case, although the data resolution is improved and there are more details, it is often inconsistent with the actual situation.

[0024] Based on this, an embodiment of the present application provides a method and device for downscaling and correcting weather forecasts based on a generative model, by which the technical problem of low data accuracy of existing weather forecast downscaling methods can be solved.

[0025] The embodiments of the present invention are further described below in conjunction with the accompanying drawings.

[0026] Figure 1 The present invention provides a flow chart of a method for downscaling and correcting weather forecasts based on a generative model. Figure 1 As shown, the method includes: Step S110, obtaining meteorological real-time analysis data and meteorological forecast historical data, and generating low-resolution forecast samples and high-resolution real-time analysis samples based on the meteorological real-time analysis data and meteorological forecast historical data.

[0027] Among them, the resolution of high-resolution actual analysis samples is higher than that of low-resolution forecast samples.

[0028] As a possible implementation, Figure 2 As shown, firstly, the global meteorological actual reanalysis data and meteorological forecast historical data are collected to produce a training data set, that is, to establish low-resolution forecast samples and high-resolution actual reanalysis samples (that is, high-resolution actual analysis samples).

[0029] Step S120: determine the high-resolution actual analysis samples as the first-stage training data set for the variational autoencoder, and determine the low-resolution prediction samples and the specified actual constraint conditions as the second-stage training data set for the diffusion model.

[0030] Among them, the training labels corresponding to the second-stage training data set are high-resolution real-life analysis samples.

[0031] In an optional embodiment, a first-stage training data set for a variational autoencoder is established using high-resolution live reanalysis samples; and a second-stage training data set for a diffusion model is constructed using low-resolution forecasts, live constraints, and live reanalysis data.

[0032] The downscaling process with real-world constraints ensures that the generated details are close to the real world while enhancing the image resolution. Moreover, by integrating the real-world constraints with the low-resolution forecast, the forecast is corrected while downscaling, so that the subsequent process does not need to be divided into two steps: downscaling and correction.

[0033] Step S130, constructing a downscaling correction model based on the diffusion model and the variational autoencoder.

[0034] Among them, the variational autoencoder part in the downscaling correction model is used to encode and decode the meteorological variables in a specified low-dimensional latent space, and the diffusion model part in the downscaling correction model is used to refine the preliminary results generated by the variational autoencoder part to improve the high-resolution correction results.

[0035] In this step, a downscaling correction model based on variational autoencoder and diffusion model is constructed, in which the main task of the variational autoencoder part is to encode and decode the low-dimensional latent space of meteorological variables. The loss functions corresponding to the variational autoencoder part include reconstruction loss function and KL divergence loss function; among them, the reconstruction loss function is used to measure the gap between the reconstructed data and the real high-resolution data corresponding to the high-resolution live analysis sample, usually using mean square error (MSE) or other correlation indicators; KL divergence loss function is used to control the closeness of the potential distribution to the standard normal distribution N(0, I).

[0036] The role of the above diffusion model part is to refine the preliminary results generated by VAE (variational autoencoder), capture more delicate patterns, and improve the effect of high-resolution correction. The diffusion process corresponding to the diffusion model part includes the forward process (Forward Process) and the reverse process (Reverse Process); among them, the forward process includes gradually adding noise to the preliminary results generated by the variational autoencoder part, and the reverse process includes learning to gradually denoise from the noise back to the high-resolution correction result corresponding to the high-resolution live analysis sample.

[0037] In an optional implementation, the construction diffusion process corresponding to the forward process is performed by the following formula: ; in, is the forward process at time step t The data is used to represent the intermediate state to which a certain amount of noise has been added; is the forward process at time step t -1 data is used to represent the output of the previous step; is the noise attenuation coefficient in the forward process, which is used to determine the mixing ratio of the original data and the noise; is the standard normal distribution noise; The weights to hold for the data; is the weight of the noise. Through the processing method of the above formula, the data processing process can be made more accurate.

[0038] In the embodiment of the present application, in the reverse process, the following formula is used to learn how to gradually remove noise and return to the high-resolution correction result: ; in, t is the time step, is the noise state at time step t, is the noise state closer to the real data at time step t-1, is the probability distribution data, N () represents the Gaussian distribution form, represents the center value of the data after denoising, is the mean, is the covariance matrix, which is used to represent the control generation The above formula can make the data processing process more accurate.

[0039] It should be noted that this formula represents the reverse process, which describes how to use the current data at time step t. Estimate the last time step . Represents the conditional probability distribution of the reverse process, which means: at time step t, given the current noise state , calculate the previous time step The probability distribution of .

[0040] Since the transition at each time step is assumed to be Gaussian, it is a conditional normal distribution. Its goal is to learn how to get from Gradually denoise to get data closer to the real data .

[0041] For the Gaussian distribution , among which is the mean, which means that at time step t, Predict The "most likely value" of . It is a function learned by a neural network (usually U-Net or Transformer). Of course, it can also be understood intuitively as: Represents the center value of the data after denoising. The closer to the real data, the better. is the covariance matrix, which represents the generation The uncertainty in time controls the diversity of the generated data. In addition, in many diffusion models, the covariance matrix It may be set to a diagonal matrix or even directly set to a constant (to simplify calculations). It can also be intuitively understood as: it determines the degree of "jitter" of denoising in the reverse process. If it is set too small, the generated data will lack diversity; if it is too large, the data will become unstable.

[0042] In addition, the input conditions corresponding to the diffusion model part include: in each step, additional conditional variables are used as model inputs to make the model generation results meet physical constraints. The additional conditional variables include any one or more of the following: terrain, climate zone information, and recent actual conditions.

[0043] Step S140, using the first-stage training data set to perform first-stage model training on the variational autoencoder part to obtain a pre-trained network, and based on the pre-trained network, using the second-stage training data set to perform second-stage model training on the diffusion model part to obtain a trained target downscaling correction model.

[0044] In practical applications, the training of the entire correction network is divided into two stages. The first stage is to train the autoencoder using real-time analysis data, and the second stage is to train the diffusion model using real-time constraints, low-resolution forecast data and high-resolution real-time reanalysis samples.

[0045] In the model training process, high-resolution actual reanalysis data is first used to train the variational autoencoder, which is the first training stage, to obtain a pre-trained network. After the pre-trained network, conditional constraints and low-resolution forecasts are used as model inputs, and actual reanalysis data are used as labels to train the diffusion model, which is the second training stage.

[0046] In the present application embodiment, Figure 2 As shown in the figure, a constrained generative downscaling correction algorithm based on diffusion (diffusion model) uses meteorological reanalysis data at continuous moments to train the variational encoder. After the training is completed, the diffusion model is trained to downscale the low-resolution meteorological forecast under the constraint of real-time reanalysis data at nearby moments, thereby obtaining high-resolution and realistically detailed results.

[0047] In practical applications, if the diffusion model is directly used to downscale meteorological data, this method can also obtain a downscaling result. However, this method lacks actual constraints and may lead to falsified details that are inconsistent with the actual situation and lack correction effect.

[0048] By constructing and training a constrained downscaling correction dataset through a diffusion model, it is possible to downscale the meteorological forecast using a diffusion model under the constraints of reanalysis conditions, so that downscaling and correction can be completed simultaneously.

[0049] Step S150, based on the weather forecast data to be corrected and the specified actual constraints, downscaling correction processing is performed using the target downscaling correction model to obtain corrected high-resolution weather forecast data.

[0050] Among them, the resolution of the revised high-resolution weather forecast data is higher than that of the weather forecast data to be revised.

[0051] After the model training is completed, the actual constraints and the low-resolution forecast data to be corrected are obtained, and the actual constraints and the low-resolution forecast data to be corrected are used together as the input of the trained target downscaling correction model to output high-resolution downscaled corrected forecast data.

[0052] In the embodiment of the present application, not only a diffusion model is used to gradually generate real high-resolution meteorological data from noise, but also the actual reanalysis data is used to constrain the downscaling process, thereby achieving downscaling and correction of meteorological forecasts at the same time. Not only can the low-resolution forecast be downscaled, but it can also ensure that the details generated by the model are close to the real situation rather than forged details, thereby improving data resolution and details, thereby improving the accuracy of meteorological forecast downscaled data.

[0053] Different from the downscaling method based on convolutional neural network in the prior art, the embodiment of the present application adopts a generative model, using a step-by-step diffusion algorithm, starting from real data, gradually adding Gaussian noise, and finally obtaining a distribution close to random noise, and then learning to gradually restore from noise to a clear image, so that the model can learn the complex distribution of data more carefully. Furthermore, the scheme provided by the embodiment of the present application is also different from the downscaling method based on generative adversarial network in the prior art. In the embodiment of the present application, a variational encoder and a diffusion model are used, and the entire training is stable and controllable, which can avoid the training collapse of the generative adversarial network. Moreover, by constructing a diffusion model with conditional constraints, real-world information is introduced while downscaling, guiding the model to approach the real situation while downscaling, and realizing downscaling and correction at the same time, solving the problem of forged details, and avoiding the problems such as the details generated by the AI ​​downscaling model are too smooth, and the generated forged details are inconsistent with the real situation.

[0054] In some embodiments, the process of obtaining weather forecast historical data in the above step S110 may specifically include the following steps: obtaining ocean current data at multiple meteorological data collection locations at different time points for an ocean area; wherein the ocean current data includes at least one of sea water temperature, air temperature, wind speed, and water flow speed; comparing the multiple ocean current data corresponding to the multiple meteorological data collection locations to obtain a comparison result, and determining at least two groups of target ocean current data whose data similarity is greater than a preset similarity based on the comparison result; determining the target meteorological data collection locations and the corresponding target time points for collection corresponding to the at least two groups of target ocean current data, and determining ocean current dynamic data based on at least two of the target meteorological data collection locations and at least two of the target time points; wherein the ocean current dynamic data includes the dynamic direction and moving time difference of the target ocean current between at least two of the target meteorological data collection locations; and determining weather forecast historical data for the ocean area based on the ocean current dynamic data.

[0055] Through the above-mentioned dynamic processing method for ocean current data, the meteorological data for ocean areas in the acquired historical weather forecast data can be made more comprehensive and accurate, so that the subsequent model training process and weather forecast data processing process can have higher data accuracy.

[0056] Figure 3 A schematic diagram of the structure of a weather forecast downscaling correction device based on a generative model is provided. Figure 3 As shown, the weather forecast downscaling correction device 300 based on the generative model includes: The acquisition module 301 is used to acquire the meteorological real-time analysis data and the meteorological forecast historical data, and generate a low-resolution forecast sample and a high-resolution real-time analysis sample based on the meteorological real-time analysis data and the meteorological forecast historical data; wherein the resolution of the high-resolution real-time analysis sample is higher than that of the low-resolution forecast sample; A determination module 302 is used to determine the high-resolution live analysis samples as a first-stage training data set for a variational autoencoder, and to determine the low-resolution prediction samples and the specified live constraint conditions as a second-stage training data set for a diffusion model, wherein the training label corresponding to the second-stage training data set is the high-resolution live analysis sample; A construction module 303 is used to construct a downscaling correction model based on the diffusion model and the variational autoencoder; wherein the variational autoencoder part in the downscaling correction model is used to encode and decode meteorological variables in a specified low-dimensional latent space, and the diffusion model part in the downscaling correction model is used to refine the preliminary results generated by the variational autoencoder to improve the high-resolution correction results; A training module 304 is used to perform first-stage model training on the variational autoencoder part using the first-stage training data set to obtain a pre-trained network, and perform second-stage model training on the diffusion model part based on the pre-trained network using the second-stage training data set to obtain a trained target downscaling correction model; The correction module 305 is used to perform downscaling correction processing based on the weather forecast data to be corrected and the specified actual constraints using the target downscaling correction model to obtain corrected high-resolution weather forecast data; wherein the resolution of the corrected high-resolution weather forecast data is higher than that of the weather forecast data to be corrected.

[0057] In some embodiments, the loss function corresponding to the variational autoencoder part includes a reconstruction loss function and a KL divergence loss function; wherein the reconstruction loss function is used to measure the gap between the reconstructed data and the actual high-resolution data corresponding to the high-resolution live analysis sample, and the KL divergence loss function is used to control the closeness of the potential distribution to the standard normal distribution N(0, I).

[0058] In some embodiments, the diffusion process corresponding to the diffusion model part includes a forward process and a reverse process; wherein, the forward process includes gradually adding noise to the preliminary result generated by the variational autoencoder part, and the reverse process includes learning to gradually denoise the noise back to the high-resolution corrected result corresponding to the high-resolution live analysis sample.

[0059] In some embodiments, the construction diffusion process corresponding to the forward process is performed by the following formula: ; in, is the forward process at time step t The data is used to represent the intermediate state to which a certain amount of noise has been added; is the forward process at time step t -1 data is used to represent the output of the previous step; is the noise attenuation coefficient in the forward process, which is used to determine the mixing ratio of the original data and the noise; is the standard normal distribution noise; The weights to hold for the data; is the weight of the noise.

[0060] In some embodiments, in the reverse process, the following formula is used to learn to gradually denoise the noise back to the high-resolution correction result: ; in, t is the time step, is the noise state at time step t, is the noise state closer to the real data at time step t-1, is the probability distribution data, N () represents the Gaussian distribution form, represents the center value of the data after denoising, is the mean, is the covariance matrix, which is used to represent the control generation Diversity of time.

[0061] In some embodiments, the input conditions corresponding to the diffusion model part include: using additional conditional variables as model inputs in each step to make the model generation results meet physical constraints.

[0062] In some embodiments, the additional condition variables include any one or more of the following: topography, climate zone information, and recent conditions.

[0063] The weather forecast downscaling correction device based on a generative model provided in the embodiment of the present application has the same technical features as the weather forecast downscaling correction method based on a generative model provided in the above-mentioned embodiment, so it can also solve the same technical problems and achieve the same technical effects.

[0064] An electronic device provided in an embodiment of the present application is Figure 4 As shown, the electronic device 400 includes a processor 402 and a memory 401, wherein the memory stores a computer program that can be run on the processor, and the processor implements the steps of the method provided in the above embodiment when executing the computer program.

[0065] See also Figure 4 The electronic device further includes: a bus 403 and a communication interface 404, a processor 402, a communication interface 404 and a memory 401 are connected via the bus 403; the processor 402 is used to execute an executable module stored in the memory 401, such as a computer program.

[0066] The memory 401 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk storage. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 404 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used.

[0067] The bus 403 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0068] Among them, the memory 401 is used to store programs, and the processor 402 executes the program after receiving the execution instruction. The method executed by the device defined by the process disclosed in any embodiment of the present application can be applied to the processor 402 or implemented by the processor 402.

[0069] The processor 402 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 402. The above processor 402 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present application can be directly embodied as a hardware decoding processor to execute, or the hardware and software modules in the decoding processor are combined to execute. The software module may be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 401, and the processor 402 reads the information in the memory 401 and completes the steps of the above method in combination with its hardware.

[0070] Corresponding to the above-mentioned weather forecast downscaling correction method based on the generative model, an embodiment of the present application also provides a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are called and executed by the processor, the computer-executable instructions prompt the processor to execute the steps of the above-mentioned weather forecast downscaling correction method based on the generative model.

[0071] The weather forecast downscaling correction device based on the generative model provided in the embodiment of the present application can be specific hardware on the device or software or firmware installed on the device. The device provided in the embodiment of the present application, its implementation principle and the technical effect produced are the same as those in the aforementioned method embodiment. For the sake of brief description, the parts not mentioned in the device embodiment can refer to the corresponding contents in the aforementioned method embodiment. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, devices and units described above can all refer to the corresponding processes in the aforementioned method embodiment, and will not be repeated here.

[0072] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0073] For another example, the flowchart and block diagram in the accompanying drawings show the possible architecture, function and operation of the device, method and computer program product according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or the flowchart, and the combination of the boxes in the block diagram and / or the flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0074] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0075] In addition, each functional unit in the embodiments provided in the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0076] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the weather forecast downscaling correction method based on the generative model described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., and other media that can store program codes.

[0077] It should be noted that similar numbers and letters represent similar items in the following figures. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are only used to distinguish the description and are not to be understood as indicating or implying relative importance.

[0078] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present application, which are used to illustrate the technical solution of the present application, rather than to limit it. The protection scope of the present application is not limited thereto. Although the present application is described in detail with reference to the aforementioned embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solution recorded in the aforementioned embodiments within the technical scope disclosed in the present application, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solution deviate from the scope of the technical solution of the embodiment of the present application. They should all be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A weather forecast downscaling correction method based on a generative model, characterized in that: The method comprises: Acquire meteorological real-time analysis data and meteorological forecast historical data, and generate low-resolution forecast samples and high-resolution real-time analysis samples based on the meteorological real-time analysis data and the meteorological forecast historical data; wherein the resolution of the high-resolution real-time analysis samples is higher than that of the low-resolution forecast samples; Determine the high-resolution live analysis samples as a first-stage training data set for a variational autoencoder, and determine the low-resolution forecast samples and specified live constraints as a second-stage training data set for a diffusion model, wherein the training labels corresponding to the second-stage training data set are the high-resolution live analysis samples; A downscaling correction model is constructed based on the diffusion model and the variational autoencoder; wherein the variational autoencoder part in the downscaling correction model is used to encode and decode meteorological variables in a specified low-dimensional latent space, and the diffusion model part in the downscaling correction model is used to refine the preliminary results generated by the variational autoencoder part to improve the high-resolution correction results; Using the first-stage training data set to perform first-stage model training on the variational autoencoder part to obtain a pre-trained network, and based on the pre-trained network, using the second-stage training data set to perform second-stage model training on the diffusion model part to obtain a trained target downscaling correction model; Based on the weather forecast data to be corrected and the specified actual constraints, downscaling correction processing is performed using the target downscaling correction model to obtain corrected high-resolution weather forecast data; wherein the resolution of the corrected high-resolution weather forecast data is higher than that of the weather forecast data to be corrected.

2. The method according to claim 1, characterized in that The loss function corresponding to the variational autoencoder part includes a reconstruction loss function and a KL divergence loss function; wherein the reconstruction loss function is used to measure the gap between the reconstructed data and the real high-resolution data corresponding to the high-resolution live analysis sample, and the KL divergence loss function is used to control the closeness of the potential distribution to the standard normal distribution N(0, I).

3. The method according to claim 1, characterized in that The diffusion process corresponding to the diffusion model part includes a forward process and a reverse process; wherein the forward process includes gradually adding noise to the preliminary result generated by the variational autoencoder part, and the reverse process includes learning to gradually denoise the noise back to the high-resolution corrected result corresponding to the high-resolution live analysis sample.

4. The method according to claim 3, characterized in that The construction diffusion process corresponding to the forward process is performed by the following formula: ； in, is the forward process at time step t The data is used to represent the intermediate state to which a certain amount of noise has been added; is the forward process at time step t -1 data is used to represent the output of the previous step; is the noise attenuation coefficient in the forward process, which is used to determine the mixing ratio of the original data and the noise; is the standard normal distribution noise; The weights to hold for the data; is the weight of the noise.

5. The method according to claim 3, characterized in that: In the reverse process, the following formula is used to learn how to gradually denoise the noise and return to the high-resolution correction result: ； in, t is the time step, is the noise state at time step t, is the noise state closer to the real data at time step t-1, is the probability distribution data, N () represents the Gaussian distribution form, represents the center value of the data after denoising, is the mean, is the covariance matrix, which is used to represent the control generation Diversity of time.

6. The method according to claim 3, characterized in that The input conditions corresponding to the diffusion model part include: taking additional conditional variables as model inputs in each step so that the model generation results meet the physical constraints.

7. The method according to claim 6, characterized in that The additional condition variables include any one or more of the following: Topography, climate zone information, and recent conditions.

8. The method according to claim 1, characterized in that The obtaining of weather forecast historical data includes: Acquire ocean current data at multiple meteorological data collection locations at different time points in an ocean area; wherein the ocean current data includes at least one of sea water temperature, air temperature, wind speed, and water flow speed; Comparing the plurality of ocean current data corresponding to the plurality of meteorological data collection location points to obtain a comparison result, and determining at least two groups of target ocean current data having a data similarity greater than a preset similarity according to the comparison result; Determine the target meteorological data collection location points and the corresponding target time points corresponding to the at least two groups of target ocean current data, and determine the ocean current dynamic data based on at least two of the target meteorological data collection location points and at least two of the target time points; wherein the ocean current dynamic data includes the dynamic direction and movement time difference of the target ocean current between at least two of the target meteorological data collection location points; Based on the ocean current dynamics data, historical weather forecast data for the ocean area is determined.

9. A weather forecast downscaling correction device based on a generative model, characterized in that: The device comprises: An acquisition module, used to acquire meteorological real-time analysis data and meteorological forecast historical data, and generate low-resolution forecast samples and high-resolution real-time analysis samples based on the meteorological real-time analysis data and the meteorological forecast historical data; wherein the resolution of the high-resolution real-time analysis samples is higher than that of the low-resolution forecast samples; A determination module, configured to determine the high-resolution live analysis samples as a first-stage training data set for a variational autoencoder, and determine the low-resolution prediction samples and the specified live constraint conditions as a second-stage training data set for a diffusion model, wherein the training label corresponding to the second-stage training data set is the high-resolution live analysis sample; A construction module is used to construct a downscaling correction model based on the diffusion model and the variational autoencoder; wherein the variational autoencoder part in the downscaling correction model is used to encode and decode meteorological variables in a specified low-dimensional latent space, and the diffusion model part in the downscaling correction model is used to refine the preliminary results generated by the variational autoencoder to improve the high-resolution correction results; A training module, configured to perform first-stage model training on the variational autoencoder part using the first-stage training data set to obtain a pre-trained network, and perform second-stage model training on the diffusion model part based on the pre-trained network using the second-stage training data set to obtain a trained target downscaling correction model; A correction module is used to perform downscaling correction processing using the target downscaling correction model based on the weather forecast data to be corrected and the specified actual constraints to obtain corrected high-resolution weather forecast data; wherein the resolution of the corrected high-resolution weather forecast data is higher than that of the weather forecast data to be corrected.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to execute the method according to any one of claims 1 to 8.

Images