A surface rainfall forecast method based on BeiDou GNSS water vapor

Abstract

The present invention discloses a surface rainfall forecasting method based on Beidou GNSS water vapor, which specifically includes the following steps: first, collecting original observation data and ephemeris data through GNSS equipment; then obtaining surface water vapor distribution; finally, constructing and training a surface rainfall spatiotemporal prediction model, wherein, for rainfall prediction, in order to suppress the model's erroneous output of predicted rainfall when there is no rainfall and improve the model's ability to accurately output rainfall, a fusion prediction layer based on classification forecast is designed in the decoder part. The present invention adopts the above-mentioned surface rainfall forecasting method based on Beidou GNSS water vapor, which can achieve high-precision monitoring and prediction of surface rainfall conditions in a certain area, has all-weather capabilities, is not affected by weather and environment, has high data reliability, and has low equipment cost, making this technology have broad application prospects in small and medium-sized river basins with limited resources and economically underdeveloped areas.

Description

Technical Field

[0001] The present invention relates to the technical field of weather forecasting, and in particular to a surface rainfall forecasting method based on Beidou GNSS water vapor. Background Art

[0002] Surface rainfall refers to the average rainfall in a specific area or river basin within a certain period of time. It is one of the most important physical quantities in hydrological and meteorological operations and scientific research. It is also an important basis for flood and drought prevention, water resources assessment and estimation, engineering design, and mountain torrent and geological disaster prevention and reduction within the river basin.

[0003] Traditional methods for calculating surface rainfall primarily rely on interpolating point rainfall measurements from ground-based rain gauges to determine regional surface rainfall. Currently, meteorological and hydrological departments widely use arithmetic mean, Thiessen polygons, and kriging. These methods are computationally simple, but they also have significant drawbacks: they fail to account for the uneven distribution of precipitation and are affected by the density of rain gauges, leading to significant measurement errors at sparse sites. Surface rainfall measurements obtained from satellite and radar measurements suffer from poor temporal or spatial resolution, and are costly and expensive to maintain, making them unsuitable for surface rainfall monitoring in small and medium-sized watersheds or economically underdeveloped regions.

[0004] With the rapid development of the new generation of global navigation satellite systems, GNSS-based atmospheric water vapor telemetry technology has the advantages of high precision, all-weather operation, long-term stability, and high spatiotemporal resolution, providing important data support for atmospheric water vapor monitoring. Among the various meteorological factors that affect weather changes, atmospheric water vapor is one of the most important parameters. Its effective monitoring and rational use are important prerequisites and keys to achieving accurate forecasts of various types of disastrous weather events. However, traditional GNSS water vapor-based rainfall forecasts mainly use the accumulated water vapor above the station detected by the GNSS station to predict whether it will rain or the amount of rainfall. It does not make sufficient use of the regional water vapor distribution information within the detection range of the GNSS station. Summary of the Invention

[0005] The purpose of the present invention is to provide a surface rainfall forecasting method based on Beidou GNSS water vapor, which is a new type of surface rainfall forecasting method that can achieve high-precision monitoring and prediction of surface rainfall conditions in a certain area. It has all-weather capabilities, is not affected by weather and environment, has high data reliability, and low equipment cost. It can also achieve overall rainfall monitoring in a region, avoiding the limitations of single-point measurements of traditional rainfall monitoring instruments. Compared with traditional methods, it can achieve more accurate and real-time predictions, and is particularly suitable for application scenarios with high requirements for temporal and spatial resolution. In addition, the low cost and easy deployment characteristics make this technology have broad application prospects in small and medium-sized river basins with limited resources and economically underdeveloped areas. In the future, with the further improvement of the Beidou system and the development of deep learning technology, the prediction accuracy and application scope of this patented technology will be further expanded.

[0006] To achieve the above object, the present invention provides a surface rainfall forecasting method based on BeiDou GNSS water vapor, which specifically includes the following steps:

[0007] S1, data collection;

[0008] S2, surface water vapor distribution acquisition;

[0009] S3. Build and train a spatiotemporal rainfall prediction model.

[0010] Preferably, in step S1, original observation data and ephemeris data are collected by GNSS equipment, and the oblique tropospheric delay in the signal path of each satellite above the site and the corresponding satellite absolute position are obtained by solving the precise single point positioning algorithm.

[0011] Preferably, in step S2, based on the oblique tropospheric delays of different satellite paths and the corresponding satellite absolute positions calculated in step S1, a tomography algorithm is used to perform three-dimensional tomography on the data collected by multiple GNSS stations.

[0012] Preferably, the specific process of performing three-dimensional tomography on the data collected by multiple GNSS stations using a tomography algorithm is as follows:

[0013] S21, dividing the regional three-dimensional grid into voxel blocks with exponentially increasing vertical intervals, and calculating an intercept matrix;

[0014] S22. Setting horizontal and vertical constraints through the boundary constraint and initialization module, and obtaining a priori background field according to the numerical model;

[0015] S23. Through the water vapor density inversion module, the observation equation and constraint equation are established to construct the tomographic model. The non-iterative singular value decomposition (SVD) is used to solve the tomographic equation to obtain the three-dimensional tomographic wet refractive index grid data. The wet refractive index of the Z axis is integrated and then calculated to obtain the two-dimensional surface water vapor grid data. The steps are repeated to obtain the final three-dimensional water vapor spatiotemporal distribution sequence.

[0016] Preferably, for any two-dimensional grid point in step S23, the total water vapor value PWV is calculated by the following formula for the wet refractive index integral of the Z-axis height axis:

[0017]

[0018] PWV (x,y) =Π·ZWD (x,y) ;

[0019] Among them, ZWD (x,y) is the total wet delay of the two-dimensional surface grid at row x and column y, H i is the height of the i-th point on the Z-axis height axis, H i+1 is the height of the i+1th point on the Z-axis height axis of this point, is the wet refractive index of the i-th point on the height axis, is the wet refractive index of the i+1th point on the height axis, Π is the coefficient calculated based on temperature and pressure, PWV (x,y) is the total water vapor value of the two-dimensional surface grid at row x and column y.

[0020] Preferably, in step S3, the specific process of constructing and training the spatiotemporal prediction model for surface rainfall is as follows:

[0021] S31. Training and testing data set collection:

[0022] Collect historical GNSS data, rainfall data, temperature and pressure meteorological data, and calculate the historical spatiotemporal data of surface water vapor distribution based on the GNSS data according to the steps in S2. Divide the data into model training and testing data sets in a ratio of 3:1.

[0023] S32. Build model:

[0024] A Transformer-based spatiotemporal rainfall prediction model is constructed. The prediction model includes a time series embedding layer and a time series encoder-decoder. The time series embedding layer projects the input matrix into another dimensional space. The encoder then uses a Transformer-based encoder-decoder to convert the input time series into a fixed-dimensional feature map. The decoder then decodes the feature map to generate the future time series. The decoder includes a rainfall generation module and a level prediction module. The final output of the spatiotemporal rainfall prediction model is the output of the rainfall generation module multiplied by the output of the level prediction module.

[0025] S33. Training model:

[0026] The constructed historical dataset is used to pre-train the spatial and temporal rainfall prediction model. The network structure and hyperparameters are adjusted by traversal, and the loss function is set to the improved root mean square error function.

[0027] The improved root mean square error loss function takes into account the rainfall events that are of primary concern in area rainfall prediction. The greater the rainfall, the greater the weight of the loss function. The calculation formula is as follows:

[0028]

[0029] Among them, L is the loss function, n is the number of samples, y i is the i-th real rainfall value, is the corresponding predicted rainfall value, α is the weight hyperparameter;

[0030] S34. Fine-tune the model based on measured data:

[0031] According to the general and universal meteorological characteristics, correlations and regularities of surface water vapor and surface rainfall learned by the pre-training model obtained by pre-training in step S33, in the fine-tuning stage, the pre-training model is fine-tuned using the local meteorological data of the target area or station. During the fine-tuning process, the weights of all other layers are frozen, the weight of the output layer is released, and the neural network is trained using the enhanced data. Only the weight of the output layer is updated for fine-tuning.

[0032] Therefore, the present invention adopts the above-mentioned surface rainfall forecasting method based on Beidou GNSS water vapor, and the beneficial effects are as follows:

[0033] (1) The present invention can realize high-precision monitoring and prediction of surface rainfall conditions in a certain area, is all-weather, is not affected by weather and environment, has high data reliability, and has low equipment cost.

[0034] (2) The present invention can also monitor rainfall across an entire region, avoiding the limitations of single-point measurements by traditional rainfall monitoring instruments. Compared to traditional methods, it can achieve more accurate and real-time predictions, making it particularly suitable for applications requiring high spatial and temporal resolution.

[0035] (3) The low cost and easy deployment of the present invention give it broad application prospects in small and medium-sized river basins with limited resources and economically underdeveloped areas. In the future, with the further improvement of the Beidou system and the development of deep learning technology, the prediction accuracy and application scope of the present invention will be further expanded.

[0036] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a schematic diagram of the overall process of an embodiment of a surface rainfall forecasting method based on Beidou GNSS water vapor according to the present invention;

[0038] Figure 2 This is a diagram of the spatiotemporal prediction model architecture of a surface rainfall forecasting method based on Beidou GNSS water vapor according to an embodiment of the present invention;

[0039] Figure 3 This is a schematic diagram of the equipment location in Hong Kong, my country and the Thiessen polygon results of the reference rainfall station in an embodiment of the surface rainfall forecast method based on Beidou GNSS water vapor of the present invention;

[0040] Figure 4 This is an overview diagram of hourly comparison of the predicted surface rainfall in Hong Kong, my country, based on the GNN network of an embodiment of the present invention based on Beidou GNSS water vapor.

[0041] Figure 5 This is a comparison diagram of the GNSS predicted surface rainfall at different times and the interpolation distribution of the reference rainfall station in an embodiment of the surface rainfall forecast method based on Beidou GNSS water vapor of the present invention. DETAILED DESCRIPTION

[0042] The technical solution of the present invention is further described below with reference to the accompanying drawings and embodiments.

[0043] Unless otherwise defined, technical or scientific terms used in the present invention shall have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs.

[0044] like Figure 1 As shown in FIG, a surface rainfall forecast method based on BeiDou GNSS water vapor includes the following steps:

[0045] S1, data collection;

[0046] The original observation data and ephemeris data are collected by GNSS equipment, and the oblique tropospheric delay and the corresponding satellite absolute position in the signal path of each satellite above the site are calculated using the precise single-point positioning algorithm.

[0047] S2, surface water vapor distribution acquisition;

[0048] According to the oblique tropospheric delays of different satellite paths and the corresponding satellite absolute positions obtained by step S1, a tomography algorithm is used to perform three-dimensional tomography on the data collected by multiple GNSS stations.

[0049] The specific process is as follows:

[0050] S21, dividing the regional three-dimensional grid into voxel blocks with exponentially increasing vertical intervals, and calculating an intercept matrix;

[0051] S22. Simultaneously, through the boundary constraint and initialization module, set horizontal and vertical constraint conditions, and obtain a priori background field according to the numerical model;

[0052] S23. Through the water vapor density inversion module, the observation equation and constraint equation are established to construct the tomographic model. The non-iterative singular value decomposition (SVD) is used to solve the tomographic equation to obtain the three-dimensional tomographic wet refractive index grid data. The wet refractive index of the Z axis is integrated and then calculated to obtain the two-dimensional surface water vapor grid data. The steps are repeated to obtain the final three-dimensional water vapor spatiotemporal distribution sequence.

[0053] For any two-dimensional grid point, the total water vapor value (PWV) is calculated by integrating the wet refractive index along the z-axis (i.e., the total water vapor value after integrating the three-dimensional grid along the z-axis) using the following formula:

[0054]

[0055] PWV (x,y) =Π·ZWD (x,y) ;

[0056] Among them, ZWD (x,y) is the total wet delay of the two-dimensional surface grid at row x and column y, H i is the height of the i-th point on the Z axis (height axis), H i+1 is the height of the i+1th point on the Z axis (height axis) of this point, is the wet refractive index of the i-th point on the Z axis (height axis), is the wet refractive index of the i+1th point on the Z axis (height axis), Π is the coefficient calculated based on temperature and pressure, PWV (x,y) is the total water vapor value of the two-dimensional surface grid at row x and column y.

[0057] S3. Build and train a spatiotemporal rainfall prediction model:

[0058] According to the three-dimensional spatiotemporal distribution sequence of water vapor in the region, it is input into the spatiotemporal prediction model of surface rainfall to predict the spatiotemporal distribution sequence of surface rainfall in the current region in the next 1-3 hours.

[0059] The specific process of building and training the spatiotemporal rainfall prediction model is as follows:

[0060] S31. Training and testing data set collection:

[0061] Collect historical GNSS data and rainfall data, temperature, air pressure and other meteorological data, and calculate the historical spatiotemporal data of surface water vapor distribution based on the GNSS data according to step S2.

[0062] like Figure 3 As shown, this embodiment conducts experiments in Hong Kong, China, using eight GNSS stations to train and test the BeiDou GNSS areal rainfall forecast model. Twenty-two reference rain gauges are used as reference values ​​to verify the accuracy of the BeiDou GNSS areal rainfall forecast model. Data from the Hong Kong GNSS station and temperature and pressure data measured by the station's meteorological sensors are collected and divided into a 3:1 ratio to form the model training and test datasets. GNSS data can be downloaded from the International GNSS Service (IGS) official website, and ERA5 reanalysis data can be downloaded from the European Centre for Medium-Range Weather Forecasts (ECMWF) official website.

[0063] S32. Build model:

[0064] like Figure 2 As shown in the figure, a Transformer-based spatiotemporal rainfall prediction model is constructed. The model input is a historical 6-hour spatiotemporal distribution sequence of water vapor, and the output is a spatiotemporal rainfall distribution sequence for the next 1-3 hours. The Transformer base model used in this model performs well in processing spatiotemporal series data and can effectively capture dynamic changes in both time and space dimensions.

[0065] The prediction model consists of a time series embedding layer and a time series encoder-decoder. The time series embedding layer projects the input matrix into another dimensional space, and then uses a Transformer-based encoder-decoder to first convert the input time series into a fixed-dimensional feature map by the encoder, and then decode the feature map to generate the future time series through the decoder.

[0066] Among them, the learning framework diagram of the Transformer-based surface rainfall prediction model is as follows:

[0067] The specific process of building a prediction model is as follows:

[0068] Step 1: Take the historical water vapor spatiotemporal sequence X of continuous α steps as input and output a series of future values ​​Y of β time steps, where and

[0069] Step 2: In the first stage, the time series embedding layer is designed, which includes the scalar projection module, position encoding and time encoding. First, the water vapor spatiotemporal sequence X needs to be projected into the required dimension in the scalar projection part, and the position information is added. Finally, the time-related encoding is added to obtain the data M input to the model. emb ;

[0070] Step 3. In the second stage, a Transformer-based time series encoder-decoder is designed. The time series encoder-decoder consists of two parts: the encoder and the decoder. Therefore, the model effectively captures both neighbor dependencies and long-distance dependencies through convolutional attention modules and multi-head attention mechanisms.

[0071] The encoder consists of a convolutional attention module and a downsampling module, and the decoder consists of a convolutional attention module, a multi-head attention mechanism, and a fully connected layer. The input of the time series encoder-decoder is M emb , the output is The model embeds the time series into the output of the layer M emb As input to the time series encoder-decoder.

[0072] Use the convolutional attention module in the encoder to the encoder input M emb Perform calculations to further obtain advanced features and capture neighbor dependencies to obtain M cbam ; Then M cbam Input to the downsampling module, through the maximum pooling layer, reduce parameters, filter features, and obtain the encoder output M encoder ;

[0073] The input of the decoder is the output M of the encoder encoder , the output is The model first uses the convolutional attention module in the decoder to further capture the neighbor dependency and prevent the receptive field from growing exponentially with the number of layers, which leads to attention to unimportant features. Then use the multi-head attention mechanism to capture long-distance dependencies, and finally use the fully connected layer to project the dimension to the original dimension of the time series to obtain the output

[0074] In addition, in the time series encoder-decoder, the model uses a convolutional attention module and a maximum pooling layer of a downsampling module to reduce the GPU memory usage during training.

[0075] Among them, for rainfall prediction, in order to suppress the model's erroneous output of predicted rainfall when there is no rainfall and improve the model's ability to accurately output rainfall, a fusion prediction layer based on classification forecast is designed in the decoder part, including a rainfall generation module and a level prediction module. The rainfall generation module consists of a multi-layer perceptron with a hidden layer of 1024 and outputs the grid rainfall prediction result; the level prediction module consists of another multi-layer perceptron with a hidden layer of 512 and a Softmax layer, predicting whether there will be rainfall in the next three hours and classifying it into different rainfall levels: no rain (0) and rain (1). The two modules are trained separately, using the root mean square error and cross entropy as the loss function respectively. After training, the final output of the surface rainfall spatiotemporal prediction model is the output result of the rainfall generation module multiplied by the model level prediction module.

[0076] S33. Training model:

[0077] The constructed historical dataset (a dataset consisting of water vapor data and rainfall data inverted from historical GNSS, and temperature and pressure meteorological data) is used to pre-train the surface rainfall spatiotemporal prediction model. By traversing and adjusting the network structure and hyperparameters, the loss function of the rainfall generation module is set to the improved root mean square error function, and the loss function of the level prediction module is set to the cross entropy loss function. After 1000 rounds of training, the model with the smallest loss function is obtained, which is the optimal model.

[0078] Among them, the improved root mean square error loss function considers the rainfall events that are the main focus in area rainfall prediction, and the greater the rainfall, the greater the weight of the loss function. The calculation formula is as follows:

[0079]

[0080] Among them, L is the loss function, n is the number of samples, y i is the i-th real rainfall value, is the corresponding predicted rainfall value, α is the weight hyperparameter;

[0081] S34. Fine-tune the model based on measured data:

[0082] According to the general and universal meteorological characteristics, correlations and regularities of surface water vapor and surface rainfall learned by the pre-training model obtained by pre-training in step S33, in the fine-tuning stage, the pre-training model is fine-tuned using the local meteorological data of the target area or station. During the fine-tuning process, the weights of all other layers are frozen, the weight of the output layer is released, and the neural network is trained using the enhanced data. Only the weight of the output layer is updated for fine-tuning.

[0083] like Figure 4Figure 2 shows a comparison of the average surface rainfall forecast results for Hong Kong, my country, using the BeiDou GNSS-derived water vapor forecast model and the average surface rainfall calculated using Thiessen polygons at reference rain gauge stations. As can be seen, the BeiDou GNSS surface rainfall forecast model provides excellent one-hour surface rainfall forecast results for the eight stations in Hong Kong, China, largely consistent with the reference surface rainfall, effectively capturing the entire rainfall process.

[0084] Through the above detailed implementation steps, the spatiotemporal rainfall prediction model can make full use of BeiDou GNSS water vapor data, as well as other meteorological data such as temperature and air pressure, to achieve high-precision prediction of the surface rainfall in a specific area within the next 1-3 hours. Figure 5 Figure (a) shows a comparison of the spatial distribution of rainfall predicted by the GNSS forecast model one hour ahead with the Kriging interpolation of the reference rainfall station. Figure (b) shows a comparison of the spatial distribution of rainfall predicted by the GNSS forecast model two hours ahead with the Kriging interpolation of the reference rainfall station. This method not only improves the accuracy and efficiency of rainfall forecasts but also effectively reduces related costs, possessing important application value in fields such as water resources management, disaster prevention, and meteorological research.

[0085] Therefore, the present invention utilizes the aforementioned Beidou GNSS water vapor-based surface rainfall forecasting method to monitor rainfall across an entire region, avoiding the limitations of single-point measurements by traditional rainfall monitoring instruments. Compared to traditional methods, this method offers more accurate and real-time forecasts with high data reliability, making it particularly suitable for applications requiring high spatial and temporal resolution.

[0086] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the same. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solutions of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A surface rainfall forecasting method based on BeiDou GNSS water vapor, characterized in that: The specific steps include: S1, data collection; S2, surface water vapor distribution acquisition; S3. Build and train a spatiotemporal rainfall prediction model; In step S1, the original observation data and ephemeris data are collected by the GNSS equipment, and the slant tropospheric delay and the corresponding satellite absolute position in the signal path of each satellite above the site are calculated by the precise single point positioning algorithm; In step S2, a tomography algorithm is used to perform three-dimensional tomography on the data collected by multiple GNSS stations based on the oblique tropospheric delays of different satellite paths and the corresponding satellite absolute positions obtained in step S1; In step S3, the specific process of building and training the spatiotemporal rainfall prediction model is as follows: S31. Training and testing data set collection: Collect historical GNSS data, rainfall data, temperature and pressure meteorological data, and calculate the historical spatiotemporal data of surface water vapor distribution based on the GNSS data according to the steps in S2. Divide the data into model training and testing data sets in a ratio of 3:

1. S32. Build model: A Transformer-based spatiotemporal rainfall prediction model is constructed. The prediction model includes a time series embedding layer and a time series encoder-decoder. The time series embedding layer projects the input matrix into another dimensional space. The encoder then uses a Transformer-based encoder-decoder to convert the input time series into a fixed-dimensional feature map. The decoder then decodes the feature map to generate the future time series. The decoder includes a rainfall generation module and a level prediction module. The final output of the spatiotemporal rainfall prediction model is the output of the rainfall generation module multiplied by the output of the level prediction module. S33. Training model: The constructed historical dataset is used to pre-train the spatial and temporal rainfall prediction model. The network structure and hyperparameters are adjusted by traversal, and the loss function is set to the improved root mean square error loss function. The improved root mean square error loss function takes into account the rainfall events that are of primary concern in area rainfall prediction. The greater the rainfall, the greater the weight of the loss function. The calculation formula is as follows: ； in, is the loss function, is the number of samples, For the The actual rainfall value, is the corresponding predicted rainfall value, is the weight hyperparameter; S34. Fine-tune the model based on measured data: According to the meteorological characteristics, correlations and regularities of surface water vapor and surface rainfall learned by the pre-training model obtained by pre-training in step S33, in the fine-tuning stage, the pre-training model is fine-tuned using the local meteorological data of the target area or station. During the fine-tuning process, the weights of all other layers are frozen, the weights of the output layer are released, and the neural network is trained using the enhanced data. Only the weights of the output layer are updated for fine-tuning.

2. The surface rainfall forecasting method based on BeiDou GNSS water vapor according to claim 1 is characterized in that: The specific process of using the tomography algorithm to perform three-dimensional tomography on the data collected by multiple GNSS stations is as follows: S21, dividing the regional three-dimensional grid into voxel blocks with exponentially increasing vertical intervals, and calculating the intercept matrix; S22. Setting horizontal and vertical constraints through the boundary constraint and initialization module, and obtaining a priori background field according to the numerical model; S23. Through the water vapor density inversion module, the observation equation and constraint equation are established to construct the tomographic model. The non-iterative singular value decomposition (SVD) is used to solve the tomographic equation to obtain the three-dimensional tomographic wet refractive index grid data. The wet refractive index of the Z axis is integrated and then calculated to obtain the two-dimensional surface water vapor grid data. The steps are repeated to obtain the final three-dimensional water vapor spatiotemporal distribution sequence.

3. The method for surface rainfall forecasting based on BeiDou GNSS water vapor according to claim 2, characterized in that: At any two-dimensional grid point in step S23, the total water vapor value PWV is calculated by the following formula for the wet refractive index integral of the Z-axis height axis: ； ； in, is the total wet delay of the 2D surface grid at row x and column y, is the height of the i-th point on the Z-axis height axis, is the height of the i+1th point on the Z-axis height axis, is the wet refractive index of the i-th point on the height axis, is the wet refractive index of the i+1th point on the height axis, is the coefficient calculated based on temperature and pressure, is the total water vapor value of the two-dimensional surface grid at row x and column y.

Images