Physical constraint fused multi-scale extreme gale event AI identification method and device and medium

Abstract

The invention discloses a physical constraint fused multi-scale extreme gale event AI identification method and device and a medium, and relates to the technical field of atmospheric science and artificial intelligence. According to the method, a novel multi-scale extreme gale event identification model is constructed by integrating multi-scale meteorological data. The model effectively captures extreme gale features under different scales through a multi-scale feature extraction module, a feature fusion sharing module and a multi-task output module. In addition, a physical constraint loss design is introduced into the model to ensure that outputs such as wind speed and air pressure conform to meteorological dynamics rules. Through deep learning analysis of meteorological data and combination of physical mechanism understanding, different types of extreme gale events such as hurricane, thunderstorm gale and tornado can be effectively distinguished, and reliable data support and scientific basis are provided for meteorological prediction and emergency response.

Description

Technical Field

[0001] The present invention relates to the field of atmospheric science and artificial intelligence technology, and more specifically, to a multi-scale AI identification method, device and medium for extreme gale events integrating physical constraints. Background Art

[0002] In the field of identification of extreme wind events, current research is mainly divided into tasks such as small-scale thunderstorm gale identification and large-scale gale prediction, and a series of technologies and methods have been developed based on different tasks.

[0003] Radar observation data, with its high temporal and spatial resolution of wind field information, provides an important basis for identifying small-scale strong local wind events. Many studies have used radar echo characteristics, such as echo intensity and rotation, combined with deep learning methods to develop recognition technologies for small-scale strong wind events such as thunderstorms and tornadoes. This type of deep learning-based method can accurately capture the characteristics of small-scale strong wind events by mining the complex patterns contained in radar data, providing timely information for early warning and emergency response. However, these methods are mainly limited to the identification of small-scale strong wind events with strong local characteristics, and it is difficult to fully cover extreme strong winds of different scales.

[0004] Numerical weather forecast models play an important role in large-scale extreme wind events, such as those caused by tropical cyclones and strong cold fronts. These models based on physical mechanisms, such as global climate models and regional climate models, can better simulate large-scale wind events caused by large weather systems by solving atmospheric dynamics equations. In recent years, a series of large artificial intelligence meteorological models (such as Pangu) have also begun to be applied to the identification and prediction of large-scale weather systems. These models perform well in dealing with large-scale weather phenomena, but still face certain challenges in capturing refined small-scale features, which affects the accuracy of identification and prediction.

[0005] In general, existing methods are difficult to fully cover extreme wind events of different scales, and new technical solutions are urgently needed to fill this gap. Therefore, how to effectively integrate multi-source data, extract multi-scale features, and combine physical mechanisms to achieve refined identification of extreme wind events of different scales has become an important research direction. Summary of the invention

[0006] In order to solve the above technical problems, the present invention provides a multi-scale AI identification method, device and medium for extreme gale events integrating physical constraints. Aiming at the diversity and complexity of extreme gale events, an identification scheme integrating physical mechanisms is proposed to solve the deficiencies of traditional methods in event classification and identification accuracy, thereby improving the detection capability and identification accuracy of extreme gale events. First, by integrating multi-source meteorological data (such as satellite remote sensing data, meteorological reanalysis data and radar echo data), the characteristics of extreme gale events of different scales are fully captured to ensure the consistency and high quality of the data, providing a solid foundation for model training; secondly, a multi-scale extreme gale event identification model MEWE-Net is constructed, and through multi-scale feature extraction and multi-task physical constraint loss function, a refined identification of various extreme gale events is achieved, the accuracy and stability of the model are improved, and reliable support is provided for meteorological warning and emergency response.

[0007] Specifically, the present invention aims to significantly improve the accuracy and stability of extreme wind identification by integrating multi-scale meteorological data and performing multi-task deep learning modeling, thereby providing more reliable support for meteorological warning and emergency response.

[0008] The model processes meteorological data of different scales through three feature extraction paths. Each path adjusts the network structure according to the characteristics of the input scale to extract large-scale features (using ResNet-50), medium-scale features (using ResNet-34) and small-scale features (using MobileNetV2+SE Block). Then, the spatial attention mechanism is used to fuse multi-scale features to enhance the focus on key spatial areas. Finally, through the multi-task output layer, the areas of hurricanes, thunderstorms and tornadoes are predicted respectively. ResNet is a deep convolutional neural network that uses residual learning to effectively alleviate the gradient vanishing problem in deep network training; MobileNetV2 is a lightweight network that optimizes computational efficiency in design and is suitable for resource-constrained environments. It also reduces computational complexity through deep separable convolutions; the spatial attention mechanism (SAM Block) can automatically focus on important areas in the input features and enhance the model's perception of key information. This design can take into account the characteristics of data of different scales and realize the advantages of multi-task learning through shared layers, further improving the ability to identify extreme wind events.

[0009] In a first aspect, the present invention provides a multi-scale AI recognition method for extreme gale events integrating physical constraints, the method comprising:

[0010] Acquiring meteorological data, and performing cleaning and standardization processing on the meteorological data to obtain preprocessed data;

[0011] Based on the preprocessed data, time-space alignment and interpolation are performed to obtain model input data; wherein the model input data is represented as X={X sat ,X re ,X radar}∈R H×W×C , X sat represents satellite remote sensing data, X re is the meteorological reanalysis data, X radar is the radar echo data, H and W are the spatial dimensions, C is the number of channels of the data, and R is a set of real numbers;

[0012] Constructing a recognition model; wherein the recognition model includes a multi-scale feature extraction module, a feature fusion sharing layer and a multi-task output layer, wherein the multi-scale feature extraction module responds to the model input data, and for different data sources and different scales, adopts different convolution structures to perform high-level feature extraction, obtains multi-scale features and feeds them to the feature fusion sharing layer; the feature fusion sharing layer uses a splicing layer and a spatial attention mechanism module to fuse the multi-scale features, obtains shared features and feeds them to the multi-task output layer; the multi-task output layer is used to pass the shared features to the multi-task fully connected layer to respectively output the occurrence probabilities of hurricanes, thunderstorms and tornadoes at various locations;

[0013] Based on the three tasks of hurricane, thunderstorm and tornado, a multi-task physical constraint loss function is constructed, and a physical constraint loss term is designed for each task to obtain a hurricane physical constraint function, a thunderstorm physical constraint function and a tornado physical constraint function. According to the hurricane physical constraint function, the thunderstorm physical constraint function and the tornado physical constraint function, weight hyperparameters are added to obtain a comprehensive loss function;

[0014] The recognition model is trained and evaluated. During the training process, the comprehensive loss function is used as the optimization target, and the model parameters are continuously adjusted through back propagation to minimize the multi-task cross entropy loss.

[0015] Furthermore, the preprocessed data includes preprocessed satellite remote sensing data, meteorological reanalysis data and radar echo data. Based on the preprocessed data, time-space alignment and interpolation are performed to obtain model input data, including:

[0016] The pre-processed satellite remote sensing data, meteorological reanalysis data and radar echo data are converted to the same geographic coordinate system to ensure that the meteorological characteristics can be consistently reflected at the same spatial scale;

[0017] The spatial resolution is adjusted by bilinear interpolation method to obtain the same grid size H×W;

[0018] The time steps of different data sources are aligned through time interpolation, the preprocessed satellite remote sensing data are interpolated from the 10-minute frequency to the per-minute step of the preprocessed satellite radar echo data, and the missing time data are filled to obtain structured model input data.

[0019] Further, the multi-scale feature extraction module includes a large-scale channel, a medium-scale channel and a small-scale channel;

[0020] In the large-scale channel, the model input data X first passes through the convolution layer Conv 7×7 Perform dimensionality reduction and obtain the first feature:

[0021]

[0022] In the formula, Represents the first feature, Conv 7×7 Represents a two-dimensional convolution layer with a convolution kernel of (7,7);

[0023] After batch normalization and maximum pooling layers, the second feature is extracted through the pre-trained ResNet-50 backbone network. The calculation formula is:

[0024]

[0025] In the formula, Represents the second feature, ResNet-50 represents the 50-layer residual neural network pre-trained in the torchvision library, BN represents batch normalization, and MaxPool represents maximum pooling;

[0026] The second feature is restored to its spatial resolution through an upsampling layer to obtain a large-scale feature F lar ;

[0027] In the mesoscale channel, the third characteristic is obtained by the following formula:

[0028]

[0029] In the formula, Represents the third feature, Conv 5×5 Represents a two-dimensional convolution layer with a convolution kernel of (5,5);

[0030] The fourth feature is extracted using batch normalization, average pooling layer and ResNet-34 network, and the calculation formula is:

[0031]

[0032] In the formula, ResNet-34 represents the 34-layer residual neural network pre-trained in the torchvision library, and AvgPool represents average pooling.

[0033] The fourth feature is passed through the upsampling layer to obtain the mid-scale feature F mid ;

[0034] In the small-scale channel, the fifth characteristic is obtained by the following formula:

[0035]

[0036] In the formula, Represents the fifth feature, Conv 3×3 Represents a two-dimensional convolution layer with a convolution kernel of (5,5);

[0037] After further feature processing by batch normalization layer and average pooling layer, the sixth feature is obtained:

[0038]

[0039] In the formula, Represents the sixth feature; MobileNetV2 represents the lightweight convolutional backbone network pre-trained in the torchvision library;

[0040] Based on the sixth feature, the SE module is used to enhance the feature weights between channels to obtain small-scale features:

[0041]

[0042] Among them, F small represents small-scale features, W1 and W2 represent the first weight matrix and the second weight matrix respectively, σ represents the Sigmoid function used to map the weights to the range of [0,1], ReLU is the activation function used for nonlinear mapping in the SE module, ⊙ represents element-by-element multiplication; i represents the row index of the feature map, and j represents the column index of the feature map.

[0043] Furthermore, in the feature fusion sharing layer, the received multi-scale feature large-scale feature F is converted into lar , mesoscale features F mid and small-scale features F small Splicing is performed on the channel dimension to obtain fusion features:

[0044] F concat =Concat(F lar ,F mid ,F small )

[0045] In the formula, Fconcat Represents fusion features, Concat represents concatenation operation;

[0046] The fusion feature F concat Input into the spatial attention mechanism module to obtain shared features:

[0047] F SAM =σ(Conv 7×7 (Concat(F avg ,F max )))⊙F concat

[0048] In the formula, F avg and F max Respectively represent F concat The feature map of average pooling and maximum pooling in the channel dimension, F SAM Represents shared features.

[0049] Furthermore, in the multi-task output layer, the shared feature F SAM Adding the features of each scale and passing through the fully connected layer branches of each task, we get the forecast outputs of hurricanes, thunderstorms, and tornadoes:

[0050]

[0051] In the formula, and They represent the prediction outputs of hurricanes, thunderstorms, and tornadoes respectively. FC1, FC2, and FC3 represent the fully connected layer branches with weights and biases corresponding to the hurricane, thunderstorm, and tornado tasks respectively.

[0052] Furthermore, the multi-task physical constraint loss function is expressed as:

[0053]

[0054] Where, L CE represents the multi-task physical constraint loss function, represents the model output of the i-th task, Y i Represents the binary occurrence label of the true i-th task at each position, i = 1, 2, 3, corresponding to hurricane, thunderstorm and tornado, respectively.

[0055] Furthermore, physical constraint loss items are designed for each task to obtain hurricane physical constraint functions, thunderstorm physical constraint functions and tornado physical constraint functions. According to the hurricane physical constraint functions, thunderstorm physical constraint functions and tornado physical constraint functions, weight hyperparameters are added to obtain a comprehensive loss function, including:

[0056] Based on the wind speed and air pressure information, the physical prior probability is defined to obtain the hurricane physical constraint function L h ; The thunderstorm wind forecast result is constrained by the local wind speed increase, and the thunderstorm wind physical constraint function L is obtained w ; Constrain the tornado prediction results by convection intensity, and obtain the tornado physical constraint function L t , calculated as:

[0057]

[0058] In the formula, σ represents the Sigmoid function, V represents the wind speed, P represents the air pressure, ΔV represents the degree of wind speed increase, S represents the convection intensity, and k1, k2, k3 and k4 all represent the adjustment coefficients;

[0059] According to the hurricane physical constraint function, thunderstorm gale physical constraint function and tornado physical constraint function, weight hyperparameters are added to obtain a comprehensive loss function, which is expressed as:

[0060] L=L CE +λ1L h +λ2L w +λ3L t

[0061] Wherein, λ1, λ2 and λ3 represent the first weight hyperparameter, the second weight hyperparameter and the third weight hyperparameter, respectively, which are used to adjust the influence of hurricane physical constraints, thunderstorm gale physical constraints and tornado physical constraints, respectively.

[0062] Furthermore, the recognition model is trained and evaluated, including:

[0063] During the model evaluation phase, accuracy, recall, and F1 scores were used to quantify the accuracy and robustness of the model in identifying hurricanes, thunderstorms, and tornadoes. At the same time, physical consistency indicators were introduced to evaluate the physical relationship between wind speed and air pressure and detect its rationality in model predictions. The error distribution of the model on different types of events was analyzed through the confusion matrix to discover and improve the probability of misclassification. Spatial distribution visualization was used to further verify the model's recognition ability in different geographic regions to ensure the reliability and applicability of the model in actual meteorological environments.

[0064] In a second aspect, the present invention provides a multi-scale AI recognition device for extreme gale events integrating physical constraints, the device comprising:

[0065] A preprocessing unit is configured to obtain meteorological data, and clean and standardize the meteorological data to obtain preprocessed data;

[0066] The alignment and interpolation unit is configured to perform spatiotemporal alignment and interpolation based on the preprocessed data to obtain model input data; wherein the model input data is represented by X={X sat ,X re ,X radar}∈R H×W×C , X sat represents satellite remote sensing data, X re is the meteorological reanalysis data, X radar is the radar echo data, H and W are the spatial dimensions, C is the number of channels of the data, and R is a set of real numbers;

[0067] A model building unit is configured to build a recognition model; wherein the recognition model includes a multi-scale feature extraction module, a feature fusion sharing layer and a multi-task output layer, wherein the multi-scale feature extraction module responds to the model input data, and for different data sources and different scales, adopts different convolution structures to perform high-level feature extraction, obtains multi-scale features and feeds them to the feature fusion sharing layer; the feature fusion sharing layer uses a splicing layer and a spatial attention mechanism module to fuse the multi-scale features, obtains shared features and feeds them to the multi-task output layer; the multi-task output layer is used to pass the shared features to the multi-task fully connected layer to respectively output the occurrence probabilities of hurricanes, thunderstorms and tornadoes at various locations;

[0068] The function construction unit is configured to construct a multi-task physical constraint loss function based on the three tasks of hurricane, thunderstorm and tornado, and design a physical constraint loss term for each task to obtain a hurricane physical constraint function, a thunderstorm physical constraint function and a tornado physical constraint function, and add a weight hyperparameter according to the hurricane physical constraint function, the thunderstorm physical constraint function and the tornado physical constraint function to obtain a comprehensive loss function;

[0069] The training and evaluation unit is configured to train and evaluate the recognition model. During the training process, the comprehensive loss function is used as the optimization target, and the model parameters are continuously adjusted through back propagation to minimize the multi-task cross entropy loss.

[0070] In a third aspect, the present invention provides a readable storage medium, wherein the readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the method as described above.

[0071] The present invention has at least the following beneficial effects:

[0072] The present invention realizes high-precision identification of hurricanes, thunderstorms and tornadoes by constructing the MEWE-Net model. The method is based on multi-source data, including satellite remote sensing, meteorological reanalysis and radar echo data, to ensure the spatiotemporal consistency of input data. First, the data preprocessing stage completes the cleaning, standardization and spatiotemporal alignment of multi-source data to form structured input data. Then, the MEWE-Net model uses a multi-scale feature extraction module and a spatial attention mechanism to fuse multi-scale features, and combines a multi-task output layer to accurately predict the probability of occurrence of extreme wind events.

[0073] 2) The present invention adopts a multi-task physical constraint loss design. This method uses meteorological dynamics prior knowledge to introduce physical constraints for hurricane, thunderstorm and tornado tasks respectively. The identification of hurricanes is based on high wind speed and low pressure, thunderstorms are constrained by local wind speed changes, and tornadoes are output in combination with convective intensity constraints. Finally, a comprehensive loss function is constructed to balance classification accuracy and physical consistency. In model training, by optimizing the comprehensive loss function, the model not only improves classification accuracy, but also ensures the physical rationality of the output.

[0074] 3) During the evaluation phase, the present invention comprehensively verifies the prediction performance of the model through classification indicators and physical consistency indicators, and further evaluates the recognition ability of the model in different regions in combination with spatial distribution visualization, thereby ensuring its applicability and reliability in the actual meteorological environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 A flowchart of a multi-scale AI identification method for extreme gale events integrating physical constraints according to an embodiment of the present invention is shown.

[0076] Figure 2 A sequence screening flow chart according to an embodiment of the present invention is shown.

[0077] Figure 3 A structural diagram of a multi-scale extreme wind event AI identification device integrating physical constraints according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0078] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. The embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings and specific embodiments, but are not intended to limit the present invention. For the various steps described herein, if there is no necessity for a causal relationship between each other, the order in which they are described as examples herein should not be regarded as a limitation, and those skilled in the art should know that they can be adjusted in order, as long as the logic between them is not destroyed, resulting in the inability to implement the entire process.

[0079] In the field of atmospheric science and weather forecasting, extreme wind events have attracted widespread attention as a meteorological phenomenon with significant environmental and social impacts. However, due to the diversity and complexity of extreme wind events, existing technologies have difficulty in comprehensively modeling multi-source data at different scales, and there are certain limitations.

[0080] In current research, the identification of extreme gale events often has analytical limitations for single-scale or specific types of events, and lacks a method for comprehensive identification of multi-scale extreme gale events. Existing numerical weather forecast models and machine learning methods often cannot effectively integrate multi-source data when capturing storm characteristics at different scales, thus limiting the comprehensive understanding and prediction of gale events. The shortcomings of this method make it difficult to accurately identify local and large-scale gale events when faced with complex meteorological phenomena. In addition, existing technologies lack flexibility and adaptability in dealing with the diversity of extreme weather events, and cannot fully utilize the complementarity of various types of meteorological data, thus affecting the accuracy and reliability of the identification results. Therefore, developing a new method that can comprehensively identify extreme gale events at different scales has become an urgent task in current research.

[0081] Based on this, the embodiment of the present invention provides a multi-scale AI recognition method for extreme wind events that integrates physical constraints. This method constructs a multi-scale extreme wind event recognition model MEWE-Net for multi-source data, designs multi-task physical constraint losses, and realizes refined recognition of three types of events: hurricanes, thunderstorms, and tornadoes. Specifically, Figure 1 As shown, the method includes steps Step 1 to Step 5, which are described in detail as follows.

[0082] Step 1: Data collection and preprocessing: Collect satellite remote sensing, meteorological reanalysis and radar echo data, clean and standardize them, and handle missing and outlier values ​​to ensure data quality.

[0083] In step 1, meteorological data is obtained, and the meteorological data is cleaned and standardized to obtain pre-processed data.

[0084] For example, meteorological data from different sources are first integrated to construct a high-quality input set with temporal and spatial consistency. Specifically, three main data sources are collected: satellite remote sensing data X sat , meteorological reanalysis dataX re and radar echo data X radar, which is used to capture detailed information about multi-scale weather events. Satellite remote sensing images provide large-scale wind field characteristics, reanalysis data supplement atmospheric background parameters (such as temperature and air pressure), and radar echo data reflect small-scale local weather characteristics with high resolution. These data are cleaned and standardized, missing values ​​are filled, outliers are removed, dimension differences are eliminated, and data are compressed to a unified numerical range to ensure consistency and high quality of various types of data.

[0085] Step 2: Spatiotemporal alignment and interpolation: Unify the temporal frequency and spatial resolution of different data sources, complete projection conversion, spatial alignment and temporal interpolation, and ensure that the data has consistent spatiotemporal distribution.

[0086] In this embodiment, step 2 is a step of performing spatiotemporal alignment and interpolation on step 1. The preprocessed data may be subjected to spatiotemporal alignment and interpolation to obtain model input data; wherein the model input data is represented as X={X sat ,X re ,X radar}∈R H×W×C , X sat represents satellite remote sensing data, X re is the meteorological reanalysis data, X radar is the radar echo data, H and W are the spatial dimensions, C is the number of channels of the data, and R is a set of real numbers.

[0087] In an exemplary embodiment, in step Step 2, the temporal frequency and spatial resolution of these data are further unified. First, the satellite, reanalysis, and radar data are projected to the same geographic coordinate system to ensure that the meteorological characteristics can be consistently reflected at the same spatial scale. The spatial resolution is adjusted by the bilinear interpolation method to obtain the same grid size H×W. In addition, the time steps of different data sources are aligned by time interpolation, the satellite remote sensing images are interpolated from a 10-minute frequency to a minute step of the radar data, and the missing time data is filled to ensure the integrity of the time series. After the above steps, structured model input data is obtained.

[0088] Step 3: Build the recognition model MEWE-Net: design multi-scale feature extraction, feature fusion sharing and multi-task output layers to form a complete deep neural network structure.

[0089] In this embodiment, the recognition model MEWE-Net includes a multi-scale feature extraction module, a feature fusion sharing layer and a multi-task output layer. The multi-scale feature extraction module responds to the model input data, and uses different convolution structures for different data sources and different scales to extract high-level features, obtains multi-scale features and feeds them to the feature fusion sharing layer; the feature fusion sharing layer uses the splicing layer and the spatial attention mechanism module to fuse the multi-scale features, obtains shared features and feeds them to the multi-task output layer; the multi-task output layer is used to pass the shared features to the multi-task fully connected layer to output the occurrence probabilities of hurricanes, thunderstorms and tornadoes at various locations respectively;

[0090] In an exemplary embodiment, the recognition model MEWE-Net includes three parts: a multi-scale feature extraction module, a feature fusion sharing layer, and a multi-task output layer. It uses a multi-scale feature extraction and multi-task learning strategy to analyze different types of meteorological data to achieve accurate recognition of hurricanes, thunderstorms, and tornadoes. Figure 2 Specifically, for the input data, first in the multi-scale feature extraction module, different convolution structures (such as different convolution kernel sizes) are used for high-level feature extraction for different data sources and different scales; then in the feature fusion sharing layer, the splicing layer and the spatial attention mechanism module (SAM Block) are used to fuse the multi-scale features to capture complex spatial relationships; further through the multi-task output layer, the shared features are passed to the multi-task fully connected layer to output the occurrence probabilities of hurricanes, thunderstorms, and tornadoes at each location.

[0091] The multi-scale feature extraction module is divided into three channels: large-scale, medium-scale and small-scale. In the large-scale channel, the input data X first passes through the convolution layer Conv 7×7 Perform dimensionality reduction to obtain features

[0092]

[0093] In the formula, Represents the first feature, Conv 7×7 Represents a two-dimensional convolution layer with a convolution kernel of (7,7);

[0094] After batch normalization and maximum pooling layers, the second feature is extracted through the pre-trained ResNet-50 backbone network. The calculation formula is:

[0095]

[0096] In the formula, Represents the second feature, ResNet-50 represents a 50-layer residual neural network pre-trained in the torchvision library, BN represents batch normalization, and MaxPool represents maximum pooling.

[0097] Then the spatial resolution is restored through the upsampling layer to obtain the final feature F lar .

[0098] In the medium-scale channel, the convolution kernel of the convolution layer is set to (5,5) size to adapt to the extraction of medium-scale features, resulting in:

[0099]

[0100] In the formula, Represents the third feature, Conv 5×5 Represents a two-dimensional convolutional layer with a convolution kernel of (5,5).

[0101] We further use batch normalization, average pooling layer and ResNet-34 network to extract mid-scale features, namely

[0102]

[0103] In the formula, ResNet-34 represents the 34-layer residual neural network pre-trained in the torchvision library, and AvgPool represents average pooling.

[0104] Finally, after the upsampling layer, the mid-scale feature F is obtained mid .

[0105] In the small-scale channel, a convolution kernel of size (3,3) is used for feature mapping to obtain the feature

[0106]

[0107] In the formula, Represents the fifth feature, Conv 3×3 Represents a two-dimensional convolutional layer with a convolution kernel of (5,5).

[0108] Subsequently, the feature processing is further performed through the batch normalization layer and the average pooling layer to reduce the risk of overfitting. This channel uses the lightweight model MobileNetV2 to avoid the inefficiency caused by too much small-scale information, namely:

[0109]

[0110] In the formula, Represents the sixth feature; MobileNetV2 represents the lightweight convolutional backbone network pre-trained in the torchvision library.

[0111] In this channel, SE Block (Squeeze-and-Excitation Block) is used to enhance the feature weights between channels to obtain the final small-scale feature F small :

[0112]

[0113] Among them, F small represents small-scale features, W1 and W2 represent the first weight matrix and the second weight matrix respectively, σ represents the Sigmoid function used to map the weights to the range of [0,1], ReLU is the activation function used for nonlinear mapping in the SE module, ⊙ represents element-by-element multiplication; i represents the row index of the feature map, and j represents the column index of the feature map.

[0114] In the feature fusion sharing layer, the received multi-scale features F lar 、F mid and F small By splicing in the channel dimension, we get the following fusion features:

[0115] F concat =Concat(F lar ,F mid ,F small )

[0116] Then F concat Input into the spatial attention mechanism module (SAM Block) to further enhance the features of important spatial locations, namely:

[0117] F concat =Concat(F lar ,F mid ,F small )

[0118] Among them, σ represents the Sigmoid function, Conv 7×7 represents a convolutional layer with a kernel size of (7,7), F avg and F max Respectively represent F concat The feature map of average pooling and maximum pooling is performed on the channel dimension, and ⊙ represents element-by-element multiplication.

[0119] Finally, in the multi-task output layer, the fusion feature F SAM By adding the features of each scale and passing through the fully connected layer branches of each task, we can get the prediction output of three extreme wind events: hurricane, thunderstorm and tornado:

[0120]

[0121] In the formula, and They represent the prediction outputs of hurricanes, thunderstorms, and tornadoes respectively. FC1, FC2, and FC3 represent the fully connected layer branches with weights and biases corresponding to the hurricane, thunderstorm, and tornado tasks respectively.

[0122] Step 4: Multi-task physical constraint loss: Construct a multi-task loss function, introduce classification and regression losses, and combine physical constraints to ensure that the output results are physically reasonable.

[0123] In an exemplary embodiment, for the three tasks of hurricane, thunderstorm and tornado, a multi-task physical constraint loss function is constructed to ensure that the output of the model on each task fits the true label value while ensuring compliance with the laws of meteorological dynamics. First, the cross entropy loss function used to fit the predicted value and the true value can be defined as:

[0124]

[0125] Where, L CE represents the multi-task physical constraint loss function, represents the model output of the i-th task, Y i Represents the binary occurrence label of the true i-th task at each position, i = 1, 2, 3, corresponding to hurricane, thunderstorm and tornado, respectively.

[0126] Based on the multi-task cross entropy loss function, a physical constraint loss term is designed for each extreme wind task. Since hurricanes usually occur when the wind speed is extremely high and the air pressure is extremely low, the physical prior probability is defined based on the wind speed and air pressure information to obtain the hurricane physical constraint function L h The probability of thunderstorms and gale is closely related to the increase of local wind speed. The prediction results are constrained by the increase of local wind speed, and the physical constraint function L of thunderstorms and gale is obtained. w ; Tornadoes are often accompanied by strong convective activities. The prediction results are constrained by the convective intensity in the input data, and the tornado physical constraint function L is obtained. t , calculated as:

[0127]

[0128] Where σ represents the Sigmoid function, V represents wind speed, P represents air pressure, ΔV represents the degree of wind speed increase, S represents convection intensity, and k1, k2, k3, and k4 all represent adjustment coefficients. The above physical constraint loss terms are combined into the multi-task loss to obtain the final loss function:

[0129] L=L CE +λ1L h +λ2L w +λ3L t

[0130] Wherein, λ1, λ2 and λ3 represent the first weight hyperparameter, the second weight hyperparameter and the third weight hyperparameter, respectively, which are used to adjust the influence of hurricane physical constraints, thunderstorm gale physical constraints and tornado physical constraints, respectively.

[0131] Step 5: Model training and evaluation: Jointly optimize multi-task losses to evaluate the accuracy and stability of model recognition.

[0132] In an exemplary embodiment, during the model training process, the comprehensive loss function L is used as the optimization target, and the model parameters are continuously adjusted through back propagation to minimize the multi-task cross entropy loss to improve the classification accuracy of hurricanes, thunderstorms, and tornadoes. At the same time, the physical constraint loss term ensures that the model output probability conforms to the dynamic laws of meteorology, thereby maintaining the classification performance while improving the physical consistency and credibility of the model prediction results. Ultimately, by optimizing the comprehensive loss function of the model, the model will be able to accurately predict the probability of occurrence of extreme weather events such as hurricanes, thunderstorms, and tornadoes.

[0133] In the model evaluation phase, a series of indicators are used to comprehensively measure the prediction performance and physical rationality of the model. First, classification evaluation indicators such as accuracy, recall, and F1 score are used to quantify the accuracy and robustness of the model in identifying extreme events such as hurricanes, thunderstorms, and tornadoes. At the same time, in order to ensure that the prediction results conform to the laws of meteorological physics, a physical consistency index is introduced to evaluate the physical relationship between wind speed and air pressure and detect its rationality in model prediction. In addition, the error distribution of the model on different types of events is analyzed through the confusion matrix to discover and improve possible misclassifications. Finally, spatial distribution visualization is used to further verify the recognition ability of the model in different geographical regions to ensure the reliability and applicability of the model in actual meteorological environments. Through these multi-level evaluations, the performance and physical rationality of the model are fully verified, providing support for its effectiveness in practical applications.

[0134] In summary, the innovation of the present invention lies in the construction of a multi-scale extreme wind event identification model MEWE-Net for multi-source data, and the introduction of a multi-task physical constraint loss function to achieve refined identification based on physical features. By integrating satellite remote sensing, meteorological reanalysis and radar echo data, the model can effectively capture the spatiotemporal characteristics of events such as hurricanes, thunderstorms and tornadoes at multiple scales, thereby achieving high-precision identification of extreme wind events. This technology integrates meteorological dynamics knowledge into model training through physical constraint loss, ensuring the physical consistency of model predictions and improving the reliability of the model.

[0135] In terms of model architecture, the MEWE-Net of the present invention consists of multi-scale feature extraction, feature fusion sharing layer and multi-task output layer. It extracts meteorological features at different spatial scales through large, medium and small scale channels, and uses spatial attention mechanism to enhance the feature response of important spatial positions. In addition, the multi-task physical constraint loss sets unique physical prior constraints for each high wind event. Through physical conditions such as the relationship between hurricane wind speed and air pressure, thunderstorm wind speed increment, and tornado convection intensity, it ensures that the output of the model on each task conforms to the actual physical laws, further improving the prediction accuracy and physical consistency of the model.

[0136] In addition, MEWE-Net of the present invention adopts multimodal data fusion and multi-task learning strategies, and establishes fine-grained prediction models for different types of gale events through spatiotemporal alignment and multi-scale processing of multi-source meteorological data. This design enables MEWE-Net to more accurately identify the location and probability of extreme gale events, providing a more valuable reference for extreme weather warnings.

[0137] This invention significantly improves the recognition accuracy and physical consistency of extreme wind events in meteorological disaster prediction through innovative data fusion, physical mechanism constraints and multi-task learning methods, and has good prospects for promotion and application and industry value.

[0138] The embodiment of the present invention also provides a multi-scale extreme wind event AI recognition device integrating physical constraints, such as Figure 3 As shown, the device comprises:

[0139] The preprocessing unit 301 is configured to obtain meteorological data, and clean and standardize the meteorological data to obtain preprocessed data;

[0140] The alignment and interpolation unit 302 is configured to perform spatiotemporal alignment and interpolation based on the preprocessed data to obtain model input data; wherein the model input data is represented by X={X sat ,X re ,X radar}∈R H×W×C , X sat represents satellite remote sensing data, X re is the meteorological reanalysis data, X radar is the radar echo data, H and W are the spatial dimensions, C is the number of channels of the data, and R is a set of real numbers;

[0141] The model building unit 303 is configured to build a recognition model; wherein the recognition model includes a multi-scale feature extraction module, a feature fusion sharing layer and a multi-task output layer, wherein the multi-scale feature extraction module responds to the model input data, and for different data sources and different scales, adopts different convolution structures to perform high-level feature extraction, obtains multi-scale features and feeds them to the feature fusion sharing layer; the feature fusion sharing layer uses a splicing layer and a spatial attention mechanism module to fuse the multi-scale features, obtains shared features and feeds them to the multi-task output layer; the multi-task output layer is used to pass the shared features to the multi-task fully connected layer to output the occurrence probabilities of hurricanes, thunderstorms and tornadoes at various locations respectively;

[0142] The function construction unit 304 is configured to construct a multi-task physical constraint loss function based on the three tasks of hurricane, thunderstorm and tornado, and design a physical constraint loss term for each task to obtain a hurricane physical constraint function, a thunderstorm physical constraint function and a tornado physical constraint function, and add a weight hyperparameter according to the hurricane physical constraint function, the thunderstorm physical constraint function and the tornado physical constraint function to obtain a comprehensive loss function;

[0143] The training evaluation unit 305 is configured to train and evaluate the recognition model. During the training process, the comprehensive loss function is used as the optimization target, and the model parameters are continuously adjusted through back propagation to minimize the multi-task cross entropy loss.

[0144] It should be noted that the various device structures described in this embodiment belong to the same technical concept as the previously described method, and achieve the same technical effect through the same principle, which will not be repeated here.

[0145] An embodiment of the present invention further provides a readable storage medium, which stores one or more programs. The one or more programs can be executed by one or more processors to implement the method described in any of the above embodiments.

[0146] In addition, although exemplary embodiments have been described herein, the scope includes any and all embodiments based on the present invention with equivalent elements, modifications, omissions, combinations (e.g., various embodiments intersecting schemes), adaptations or changes. The elements in the claims will be interpreted broadly based on the language adopted in the claims, and are not limited to the examples described in this specification or during the implementation of this application, and the examples will be interpreted as non-exclusive. Therefore, this specification and examples are intended to be considered as examples only, and the true scope and spirit are indicated by the following claims and the full scope of their equivalents.

[0147] The above description is intended to be illustrative rather than restrictive. For example, the above examples (or one or more of them) can be used in combination with each other. For example, those of ordinary skill in the art can use other embodiments when reading the above description. In addition, in the above-mentioned specific embodiments, various features can be grouped together to simplify the present invention. This should not be interpreted as a feature of an invention that is not claimed for protection being necessary for any claim. On the contrary, the subject matter of the present invention may be less than all the features of the embodiments of a specific invention. Thus, the following claims are incorporated into the specific embodiments as examples or embodiments, wherein each claim is independently used as a separate embodiment, and it is considered that these embodiments can be combined with each other in various combinations or arrangements. The scope of the present invention should be determined with reference to the full scope of the equivalent forms of the attached claims and these claims.

Claims

1. A multi-scale AI recognition method for extreme wind events integrating physical constraints, characterized in that: The method comprises: Acquiring meteorological data, and performing cleaning and standardization processing on the meteorological data to obtain preprocessed data; Based on the preprocessed data, time-space alignment and interpolation are performed to obtain model input data; wherein the model input data is represented as X={X sat ,X re ,X radar }∈R H×W×C , X sat represents satellite remote sensing data, X re is the meteorological reanalysis data, X radar is the radar echo data, H and W are the spatial dimensions, C is the number of channels of the data, and R is a set of real numbers; Constructing a recognition model; wherein the recognition model includes a multi-scale feature extraction module, a feature fusion sharing layer and a multi-task output layer, wherein the multi-scale feature extraction module responds to the model input data, and for different data sources and different scales, adopts different convolution structures to perform high-level feature extraction, obtains multi-scale features and feeds them to the feature fusion sharing layer; the feature fusion sharing layer uses a splicing layer and a spatial attention mechanism module to fuse the multi-scale features, obtains shared features and feeds them to the multi-task output layer; the multi-task output layer is used to pass the shared features to the multi-task fully connected layer to respectively output the occurrence probabilities of hurricanes, thunderstorms and tornadoes at various locations; Based on the three tasks of hurricane, thunderstorm and tornado, a multi-task physical constraint loss function is constructed, and a physical constraint loss term is designed for each task to obtain a hurricane physical constraint function, a thunderstorm physical constraint function and a tornado physical constraint function. According to the hurricane physical constraint function, the thunderstorm physical constraint function and the tornado physical constraint function, weight hyperparameters are added to obtain a comprehensive loss function; The recognition model is trained and evaluated. During the training process, the comprehensive loss function is used as the optimization target, and the model parameters are continuously adjusted through back propagation to minimize the multi-task cross entropy loss.

2. The method according to claim 1, characterized in that The preprocessed data includes preprocessed satellite remote sensing data, meteorological reanalysis data and radar echo data. Based on the preprocessed data, time-space alignment and interpolation are performed to obtain model input data, including: The pre-processed satellite remote sensing data, meteorological reanalysis data and radar echo data are converted to the same geographic coordinate system to ensure that the meteorological characteristics can be consistently reflected at the same spatial scale; The spatial resolution is adjusted by bilinear interpolation method to obtain the same grid size H×W; The time steps of different data sources are aligned through time interpolation, the preprocessed satellite remote sensing data are interpolated from the 10-minute frequency to the per-minute step of the preprocessed satellite radar echo data, and the missing time data are filled to obtain structured model input data.

3. The method according to claim 1, characterized in that The multi-scale feature extraction module includes a large-scale channel, a medium-scale channel and a small-scale channel; In the large-scale channel, the model input data X first passes through the convolution layer Conv 7×7 Perform dimensionality reduction and obtain the first feature: In the formula, Represents the first feature, Conv 7×7 Represents a two-dimensional convolution layer with a convolution kernel of (7,7); After batch normalization and maximum pooling layers, the second feature is extracted through the pre-trained ResNet-50 backbone network. The calculation formula is: In the formula, Represents the second feature, ResNet-50 represents the 50-layer residual neural network pre-trained in the torchvision library, BN represents batch normalization, and MaxPool represents maximum pooling; The second feature is restored to its spatial resolution through an upsampling layer to obtain a large-scale feature F lar ; In the mesoscale channel, the third characteristic is obtained by the following formula: In the formula, Represents the third feature, Conv 5×5 Represents a two-dimensional convolution layer with a convolution kernel of (5,5); The fourth feature is extracted using batch normalization, average pooling layer and ResNet-34 network, and the calculation formula is: In the formula, ResNet-34 represents the 34-layer residual neural network pre-trained in the torchvision library, and AvgPool represents average pooling. The fourth feature is passed through the upsampling layer to obtain the mid-scale feature F mid ; In the small-scale channel, the fifth characteristic is obtained by the following formula: In the formula, Represents the fifth feature, Conv 3×3 Represents a two-dimensional convolution layer with a convolution kernel of (5,5); After further feature processing by batch normalization layer and average pooling layer, the sixth feature is obtained: In the formula, Represents the sixth feature; MobileNetV2 represents the lightweight convolutional backbone network pre-trained in the torchvision library; Based on the sixth feature, the SE module is used to enhance the feature weights between channels to obtain small-scale features: Among them, F small represents small-scale features, W1 and W2 represent the first weight matrix and the second weight matrix respectively, σ represents the Sigmoid function used to map the weights to the range of [0,1], ReLU is the activation function used for nonlinear mapping in the SE module, ⊙ represents element-by-element multiplication; i represents the row index of the feature map, and j represents the column index of the feature map.

4. The method according to claim 3, characterized in that: In the feature fusion sharing layer, the received multi-scale feature large-scale feature F is converted into lar , mesoscale features F mid and small-scale features F small Splicing is performed on the channel dimension to obtain fusion features: F concat =Concat(F lar ,F mid ,F small ) In the formula, F concat Represents fusion features, Concat represents concatenation operation; The fusion feature F concat Input into the spatial attention mechanism module to obtain shared features: F SAM =σ(Conv 7×7 (Concat(F avg ,F max )))⊙F concat In the formula, F avg and F max Respectively represent F concat The feature map of average pooling and maximum pooling in the channel dimension, F SAM Represents shared features.

5. The method according to claim 4, characterized in that In the multi-task output layer, the shared feature F SAM Adding the features of each scale and passing through the fully connected layer branches of each task, we get the forecast outputs of hurricanes, thunderstorms, and tornadoes: In the formula, and They represent the prediction outputs of hurricanes, thunderstorms, and tornadoes respectively. FC1, FC2, and FC3 represent the fully connected layer branches with weights and biases corresponding to the hurricane, thunderstorm, and tornado tasks respectively.

6. The method according to claim 5, characterized in that The multi-task physical constraint loss function is expressed as: Where, L CE represents the multi-task physical constraint loss function, represents the model output of the i-th task, Y i Represents the binary occurrence label of the true i-th task at each position, i = 1, 2, 3, corresponding to hurricane, thunderstorm and tornado, respectively.

7. The method according to claim 6, characterized in that A physical constraint loss term is designed for each task to obtain a hurricane physical constraint function, a thunderstorm physical constraint function, and a tornado physical constraint function. According to the hurricane physical constraint function, the thunderstorm physical constraint function, and the tornado physical constraint function, a weight hyperparameter is added to obtain a comprehensive loss function, including: Based on the wind speed and air pressure information, the physical prior probability is defined to obtain the hurricane physical constraint function L h ; The thunderstorm wind forecast result is constrained by the local wind speed increase, and the thunderstorm wind physical constraint function L is obtained w ; Constrain the tornado prediction results by convection intensity, and obtain the tornado physical constraint function L t , calculated as: In the formula, σ represents the Sigmoid function, V represents the wind speed, P represents the air pressure, ΔV represents the degree of wind speed increase, S represents the convection intensity, and k1, k2, k3 and k4 all represent the adjustment coefficients; According to the hurricane physical constraint function, thunderstorm gale physical constraint function and tornado physical constraint function, weight hyperparameters are added to obtain a comprehensive loss function, which is expressed as: L=L CE +λ1L h +λ2L w +λ3L t Wherein, λ1, λ2 and λ3 represent the first weight hyperparameter, the second weight hyperparameter and the third weight hyperparameter, respectively, which are used to adjust the influence of hurricane physical constraints, thunderstorm gale physical constraints and tornado physical constraints, respectively.

8. The method according to claim 1, characterized in that The recognition model is trained and evaluated, including: During the model evaluation phase, accuracy, recall, and F1 scores were used to quantify the accuracy and robustness of the model in identifying hurricanes, thunderstorms, and tornadoes. At the same time, physical consistency indicators were introduced to evaluate the physical relationship between wind speed and air pressure and detect its rationality in model predictions. The error distribution of the model on different types of events was analyzed through the confusion matrix to discover and improve the probability of misclassification. Spatial distribution visualization was used to further verify the model's recognition ability in different geographic regions to ensure the reliability and applicability of the model in actual meteorological environments.

9. A multi-scale AI recognition device for extreme gale events integrating physical constraints, characterized in that: The device comprises: A preprocessing unit is configured to obtain meteorological data, and clean and standardize the meteorological data to obtain preprocessed data; The alignment and interpolation unit is configured to perform spatiotemporal alignment and interpolation based on the preprocessed data to obtain model input data; wherein the model input data is represented by X={X sat ,X re ,X radar }∈R H×W×C , X sat represents satellite remote sensing data, X re is the meteorological reanalysis data, X radar is the radar echo data, H and W are the spatial dimensions, C is the number of channels of the data, and R is a set of real numbers; A model building unit is configured to build a recognition model; wherein the recognition model includes a multi-scale feature extraction module, a feature fusion sharing layer and a multi-task output layer, wherein the multi-scale feature extraction module responds to the model input data, and for different data sources and different scales, adopts different convolution structures to perform high-level feature extraction, obtains multi-scale features and feeds them to the feature fusion sharing layer; the feature fusion sharing layer uses a splicing layer and a spatial attention mechanism module to fuse the multi-scale features, obtains shared features and feeds them to the multi-task output layer; the multi-task output layer is used to pass the shared features to the multi-task fully connected layer to respectively output the occurrence probabilities of hurricanes, thunderstorms and tornadoes at various locations; The function construction unit is configured to construct a multi-task physical constraint loss function based on the three tasks of hurricane, thunderstorm and tornado, and design a physical constraint loss term for each task to obtain a hurricane physical constraint function, a thunderstorm physical constraint function and a tornado physical constraint function, and add a weight hyperparameter according to the hurricane physical constraint function, the thunderstorm physical constraint function and the tornado physical constraint function to obtain a comprehensive loss function; The training and evaluation unit is configured to train and evaluate the recognition model. During the training process, the comprehensive loss function is used as the optimization target, and the model parameters are continuously adjusted through back propagation to minimize the multi-task cross entropy loss. 10 . A non-transitory computer-readable storage medium storing instructions, which, when executed by a processor, perform the method according to claim 1 .

Images