Improved atmospheric weighted average temperature modeling method

Abstract

The invention discloses an improved atmospheric weighted average temperature modeling method, and relates to the technical field of meteorological modeling, and the method comprises the following steps: S1, obtaining multi-source data related to atmospheric weighted average temperature; s2, preprocessing the multi-source data; s3, constructing a least square method linear regression model, and performing preliminary modeling on the atmospheric weighted average temperature; s4, constructing an improved convolutional long-short term memory network model, training, calculating an error between a predicted value and a true value, and updating parameters of the improved convolutional long-short term memory network model; s5, fusing predicted values of the least square method linear regression model and the improved convolution long-short term memory network model; and S6, performing regional adaptation optimization on the fused atmospheric weighted average temperature predicted value based on the regional correction factor. The method combines the least square method linear regression with the improved convolution long-short term memory network, accurately models the atmospheric weighted average temperature, and has the advantages of high precision, strong adaptability and outstanding regional optimization capability.

Description

Technical Field

[0001] The invention relates to the technical field of meteorological modeling, and in particular to an improved atmospheric weighted average temperature modeling method. Background Art

[0002] The weighted average temperature of the atmosphere is a key parameter in the tropospheric delay correction of the global navigation satellite system. Its accuracy directly affects the positioning accuracy of the global navigation satellite system, the inversion effect of meteorological parameters and the computational performance of the numerical weather forecast model. In the prior art, the calculation methods of the weighted average temperature of the atmosphere mainly include empirical models, simple regression models and some physics-based atmospheric models. Although these methods perform well under certain specific conditions, they are difficult to fully adapt to the needs of different regions and complex meteorological conditions due to the limitations of the model itself and the complexity of the calculation data.

[0003] Empirical models are built based on fixed parameters and usually rely on observational data from mid-latitudes, making them less applicable to tropical, high-latitude, and high-altitude regions. The physical assumptions of these models are usually based on static temperature and pressure distributions, ignoring dynamic changes in the troposphere, such as strong convective activity or extreme weather conditions. In addition, empirical models are difficult to integrate multi-source data (such as satellite remote sensing data, ground meteorological station data, and numerical weather forecast data), and their simplicity also leads to limited processing capabilities for complex nonlinear relationships. In practical applications, this approach often introduces systematic errors, limiting the accuracy of tropospheric delay corrections in global navigation satellite systems.

[0004] Simple linear regression model is another commonly used method, which calculates the atmospheric weighted average temperature by linearly fitting the relationship between meteorological characteristic variables (such as air pressure, temperature) and atmospheric weighted average temperature. Although this method is slightly more flexible than the empirical model, the limitations of the linear assumption make it difficult to capture the complex nonlinear relationship between atmospheric variables, especially under dynamic meteorological conditions. In addition, with the increase in data volume and the expansion of meteorological characteristic dimensions, simple regression models are susceptible to the "dimensionality disaster", resulting in a decrease in the robustness of the model and the prediction accuracy. At the same time, most of the existing regression models lack error correction mechanisms for specific regions, resulting in large differences in their adaptability in different geographical regions and climatic conditions.

[0005] In recent years, the application of deep learning technology in the field of meteorology has provided new possibilities for modeling atmospheric weighted average temperature. Convolutional neural networks perform well in spatial feature extraction, while long short-term memory networks have advantages in processing time series data. However, these technologies still face many problems in practical applications. On the one hand, traditional convolutional neural networks and long short-term memory network models often cannot fully utilize the spatiotemporal coupling characteristics contained in meteorological data, resulting in limited model prediction capabilities. On the other hand, existing deep learning models usually lack a mechanism for dynamically weighting the importance of key areas and time steps of data, and cannot effectively capture complex spatiotemporal characteristics. In addition, the "black box" nature of deep learning models increases the difficulty of model interpretability and limits their application in fields with high precision requirements.

[0006] Existing technologies also have deficiencies in the fusion and feature extraction of multi-source meteorological data. Meteorological data usually include ground meteorological station observations, satellite remote sensing and numerical weather forecast output data, which are often inconsistent in time and spatial resolution. Missing data, noise and outliers also pose challenges to modeling. Traditional data preprocessing methods lack systematicity and standardization, making it difficult to effectively utilize information in multi-source data, resulting in insufficient input features to support the needs of complex models. In addition, current modeling techniques usually fail to fully consider the temporal, spatial and vertical profile characteristics of meteorological data, and fail to fully tap the potential of multidimensional meteorological data.

[0007] In terms of regional adaptability, the existing atmospheric weighted mean temperature models lack dynamic optimization mechanisms for different geographical regions and climate conditions, and usually rely on static empirical corrections or offline optimization. This method is difficult to adapt to real-time changing meteorological conditions, resulting in significant differences in prediction accuracy between regions. For regions with significant climate differences, such as tropical and high-latitude regions, the errors of existing models are more obvious. In addition, the regional adaptation of the model often ignores the quantification and correction of systematic deviations, further limiting the practical application effect of the model.

[0008] In summary, the existing technologies in atmospheric weighted average temperature modeling mainly have the following deficiencies: the accuracy of empirical models and simple regression models is limited, and it is difficult to adapt to complex meteorological conditions; the multi-source data fusion and feature extraction capabilities are insufficient, and the multi-dimensional information in meteorological data cannot be effectively utilized; the feature extraction and time series modeling capabilities of deep learning models are limited, and the complex relationships in the data cannot be fully captured; the model lacks regional adaptability and cannot be effectively optimized for different geographical regions and climate conditions. Therefore, there is an urgent need for an improved method that combines statistical modeling and deep learning technology, which can make full use of multi-source data, enhance feature extraction capabilities, and dynamically optimize regional correction mechanisms to improve the accuracy, applicability, and stability of atmospheric weighted average temperature modeling. Summary of the invention

[0009] One object of the present invention is to propose an improved atmospheric weighted average temperature modeling method. The present invention combines the least squares linear regression model with the improved convolutional long short-term memory network, makes full use of multi-source meteorological data, and accurately models the atmospheric weighted average temperature by integrating statistical modeling and deep learning technology. A dynamic regional correction optimization mechanism is proposed, which has the advantages of high calculation accuracy, strong adaptability, and outstanding dynamic optimization capability for regional errors.

[0010] An improved atmospheric weighted average temperature modeling method according to an embodiment of the present invention comprises the following steps: S1. Obtain multi-source data related to atmospheric weighted average temperature; S2. Preprocessing the multi-source data to generate a multi-dimensional meteorological feature input data set; S3. Construct a least squares linear regression model. Based on the multidimensional meteorological characteristic input data set, use the least squares linear regression model to preliminarily model the atmospheric weighted average temperature, fit the relationship between the atmospheric weighted average temperature and the meteorological characteristic variables, and generate the predicted value of the least squares linear regression model. S4. Construct an improved convolutional long short-term memory network model, use the multidimensional meteorological feature input data set as a training sample, train the convolutional long short-term memory network model, calculate the error between the predicted value and the true value according to the loss function, update the parameters of the improved convolutional long short-term memory network model, and generate the predicted value of the improved convolutional long short-term memory network model; S5, fusion least squares linear regression model and improved convolutional long short-term memory network model prediction value, generate fusion atmospheric weighted average temperature prediction value; S6. Based on the fused atmospheric weighted average temperature prediction value, the fused atmospheric weighted average temperature prediction value is regionally adapted and optimized in combination with the regional correction factor.

[0011] Optionally, the multi-source data includes temperature, humidity, air pressure observed by ground meteorological stations and satellite remote sensing data.

[0012] Optionally, the preprocessing includes time and space alignment, data cleaning, standardization processing and feature variable extraction, and the multidimensional meteorological feature input data set includes time features, space features and vertical profile features.

[0013] Optionally, the S3 specifically includes: S31, obtaining air pressure and temperature data in a multi-dimensional meteorological feature input data set; S32. Construct a least squares linear regression model to fit the atmospheric weighted average temperature Nonlinear relationship between pressure and temperature data: ; ; ; in, represents the weighted mean temperature of the atmosphere, , , , , , and represents the coefficient of the regression model, which is determined by data fitting, P represents air pressure, T represents temperature, DOY represents the day of the year, and Indicates the number of days; S33, using the least squares method to perform parameter fitting on historical meteorological data to obtain optimal model parameters; S34. Calculate the mean square error (MSE) of the least squares linear regression model to evaluate the performance of the least squares linear regression model: ; Among them, MSE represents the mean square error of the improved polynomial regression model, represents the true value of the i-th sample, represents the predicted value of the i-th sample, N represents the number of samples, represents the true value mean; S35. Output the predicted value of the improved polynomial regression model .

[0014] Optionally, the S4 specifically includes: S41. Construct an improved convolutional long short-term memory network model, including: The input layer receives a multidimensional meteorological feature input data set and sets the dimension of the input tensor to N×Y×M, where N represents the number of samples, Y represents the time step, and M represents the number of meteorological feature variables; The improved convolution layer introduces a spatial attention mechanism to enhance the feature extraction capability of important areas of spatial correlation. The output dimension of the improved convolution layer is N×Y×C, where C in the output dimension of the improved convolution layer represents the number of channels. The improved long short-term memory network layer introduces a gated attention mechanism to enhance the ability to allocate weights to important information at different time steps. The output dimension of the improved long short-term memory network layer is N×H, where H in the output dimension of the improved long short-term memory network layer represents the number of hidden units. Output layer, generates the predicted value of atmospheric weighted average temperature, and the output dimension is N×1; S42, the multi-dimensional meteorological feature input data set is input into the improved convolution layer for feature extraction, and the feature Z extracted by the convolution layer introduces the spatial attention weight: ; in, represents the enhanced features of the convolutional layer output, represents the spatial attention weight matrix, and They represent the convolution kernel weight and bias of the spatial attention mechanism respectively, Sigmoid represents the activation function, represents the convolution calculation, represents the point-by-point product calculation, and Z represents the features extracted by the improved convolutional layer; S43. Adding temporal attention weights to the input gate, forget gate, and output gate formulas of the traditional long short-term memory network layer Form an improved long short-term memory network and enhance the features Input to the improved long short-term memory network: ; ; ; ; ; ; in, , and Represent the activation values ​​of the input gate, forget gate and output gate respectively, with a dimension of H. and Represent the unit state and hidden state respectively, with a dimension of H, , , and Represents the weight matrix input to each gate and state, with a dimension of H×C, , , and represents the bias term, with dimension H, , , and Represents the weight matrix from hidden state to each gate and state, with dimension H×H, represents the time attention weight, with dimension Y, represents the temporal attention weight matrix, with dimension Y×(C+H), represents the temporal attention bias term, with dimension Y, represents the concatenation of the input at time step t and the hidden state at the previous time step, represents the input of time step t, tanh represents the hyperbolic tangent activation function, ReLU activation function. S44. Define the mean square error loss function and use the Adam algorithm to update the parameters: ; in, represents the mean square error loss function, represents the true value of the i-th sample, represents the predicted value of the i-th sample, N represents the number of samples, Represents the parameters of the improved convolutional long short-term memory network model; S45, the output layer generates the predicted value of the improved convolutional long short-term memory network model: ; in, represents the predicted value of the improved convolutional long short-term memory network model, represents the output layer weight, Represents the output layer bias.

[0015] Optionally, the S6 specifically includes: S61. Define the regional correction factor , quantifies the systematic deviation of the prediction error in region i: ; in, represents the correction factor for region i, represents the number of samples in region i, represents the real atmospheric weighted average temperature of the jth sample, represents the predicted value of the fused atmospheric weighted average temperature of the jth sample; S62. Divide region i according to geographical location and meteorological conditions, and each region contains the latitude and longitude coordinate range corresponding to the sample: ; ; in, and Respectively represent the latitude and longitude of the jth sample, , , and Indicates the upper and lower limits of the geographical range of region i; S63, use area correction factor Fusion atmospheric weighted mean temperature forecast Make corrections to generate the predicted value after regional adaptation optimization: ; in, represents the predicted value of atmospheric weighted average temperature after regional adaptation optimization, represents the correction factor of region i to which sample j belongs, represents the predicted value of the fused atmospheric weighted average temperature of the jth sample; S64. Dynamically update the regional correction factor in real-time prediction , and correct it with the new observation data: ; in, represents the updated regional correction factor, represents the historical area correction factor, represents the correction amount based on new observation data, Indicates the update rate.

[0016] The beneficial effects of the present invention are: First, in the modeling process, the present invention uses the least squares linear regression model to preliminarily model the atmospheric weighted average temperature. The introduction of the least squares method provides a good physical interpretation for modeling the linear relationship between meteorological variables and the atmospheric weighted average temperature, and evaluates the model performance through the mean square error, making the model more robust and controllable.

[0017] Secondly, in order to further improve the ability to capture nonlinear relationships, the present invention designs a deep learning model based on an improved convolutional long short-term memory network. The convolutional layer introduces a spatial attention mechanism to dynamically identify important areas in multi-source meteorological data, enhances the ability to extract spatial features, and significantly improves the adaptability of the model to complex geographical conditions. At the same time, the improved long short-term memory network layer uses a gated attention mechanism to dynamically assign weights to time series data, thereby accurately capturing the time-dependent characteristics of the atmospheric weighted average temperature. This series-connected deep learning structure fully integrates spatial and temporal characteristics, providing a powerful tool for nonlinear modeling of the atmospheric weighted average temperature.

[0018] In addition, the present invention further adopts a model fusion strategy to combine the prediction results of the least squares linear regression model with the prediction results of the improved convolutional long short-term memory network model to generate a more accurate and stable atmospheric weighted average temperature prediction value. By integrating the interpretability of traditional statistical models and the nonlinear modeling capabilities of deep learning models, the limitations of a single model are overcome, and a high-precision prediction effect can be maintained under different meteorological conditions.

[0019] Finally, in response to the regional adaptability problem, the present invention innovatively proposes a dynamic optimization mechanism based on regional correction factors. By quantifying the systematic deviation of the prediction error in the region and combining it with a real-time update strategy, the present invention can dynamically adjust the prediction results of the model to adapt to changes in different geographical regions and meteorological conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 A flowchart of an improved atmospheric weighted average temperature modeling method proposed by the present invention; Figure 2 A schematic diagram of the structure of an improved convolutional long short-term memory network model for an improved atmospheric weighted average temperature modeling method proposed in the present invention. DETAILED DESCRIPTION

[0021] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0022] refer to Figure 1 and Figure 2 , an improved atmospheric weighted average temperature modeling method, comprising the following steps: S1. Obtain multi-source data related to atmospheric weighted average temperature; S2. Preprocessing the multi-source data to generate a multi-dimensional meteorological feature input data set; S3. Construct a least squares linear regression model. Based on the multidimensional meteorological characteristic input data set, use the least squares linear regression model to preliminarily model the atmospheric weighted average temperature, fit the relationship between the atmospheric weighted average temperature and the meteorological characteristic variables, and generate the predicted value of the least squares linear regression model. S4. Construct an improved convolutional long short-term memory network model, use the multidimensional meteorological feature input data set as a training sample, train the convolutional long short-term memory network model, calculate the error between the predicted value and the true value according to the loss function, update the parameters of the improved convolutional long short-term memory network model, and generate the predicted value of the improved convolutional long short-term memory network model; S5, fusion least squares linear regression model and improved convolutional long short-term memory network model prediction value, generate fusion atmospheric weighted average temperature prediction value; S6. Based on the fused atmospheric weighted average temperature prediction value, the fused atmospheric weighted average temperature prediction value is regionally adapted and optimized in combination with the regional correction factor.

[0023] In this embodiment, the multi-source data includes temperature, humidity, air pressure observed by ground meteorological stations and satellite remote sensing data.

[0024] In this implementation, the preprocessing includes time and space alignment, data cleaning, standardization processing and feature variable extraction, and the multidimensional meteorological feature input data set includes time features, space features and vertical profile features.

[0025] In this implementation, S3 specifically includes: S31, obtaining air pressure and temperature data in a multi-dimensional meteorological feature input data set; S32. Construct a least squares linear regression model to fit the atmospheric weighted average temperature Nonlinear relationship between pressure and temperature data: ; ; ; in, represents the weighted mean temperature of the atmosphere, , , , , , and represents the coefficient of the regression model, which is determined by data fitting, P represents air pressure, T represents temperature, DOY represents the day of the year, and Indicates the number of days; S33, using the least squares method to perform parameter fitting on historical meteorological data to obtain optimal model parameters; S34. Calculate the mean square error (MSE) of the least squares linear regression model to evaluate the performance of the least squares linear regression model: ; Among them, MSE represents the mean square error of the improved polynomial regression model, represents the true value of the i-th sample, represents the predicted value of the i-th sample, N represents the number of samples, represents the true value mean; S35. Output the predicted value of the improved polynomial regression model .

[0026] In this implementation, S4 specifically includes: S41. Construct an improved convolutional long short-term memory network model, including: The input layer receives a multidimensional meteorological feature input data set and sets the dimension of the input tensor to N×Y×M, where N represents the number of samples, Y represents the time step, and M represents the number of meteorological feature variables; The improved convolution layer introduces a spatial attention mechanism to enhance the feature extraction capability of important areas of spatial correlation. The output dimension of the improved convolution layer is N×Y×C, where C in the output dimension of the improved convolution layer represents the number of channels. The improved long short-term memory network layer introduces a gated attention mechanism to enhance the ability to allocate weights to important information at different time steps. The output dimension of the improved long short-term memory network layer is N×H, where H in the output dimension of the improved long short-term memory network layer represents the number of hidden units. Output layer, generates the predicted value of atmospheric weighted average temperature, and the output dimension is N×1; S42, the multi-dimensional meteorological feature input data set is input into the improved convolution layer for feature extraction, and the feature Z extracted by the convolution layer introduces the spatial attention weight: ; in, represents the enhanced features of the convolutional layer output, represents the spatial attention weight matrix, and They represent the convolution kernel weight and bias of the spatial attention mechanism respectively, Sigmoid represents the activation function, represents the convolution calculation, represents the point-by-point product calculation, and Z represents the features extracted by the improved convolutional layer; S43. Adding temporal attention weights to the input gate, forget gate, and output gate formulas of the traditional long short-term memory network layer Form an improved long short-term memory network and enhance the features Input to the improved long short-term memory network: ; ; ; ; ; ; in, , and Represent the activation values ​​of the input gate, forget gate and output gate respectively, with a dimension of H. and Represent the unit state and hidden state respectively, with a dimension of H, , , and Represents the weight matrix input to each gate and state, with a dimension of H×C, , , and represents the bias term, with dimension H, , , and Represents the weight matrix from hidden state to each gate and state, with dimension H×H, represents the time attention weight, with dimension Y, represents the temporal attention weight matrix, with dimension Y×(C+H), represents the temporal attention bias term, with dimension Y, represents the concatenation of the input at time step t and the hidden state at the previous time step, represents the input of time step t, tanh represents the hyperbolic tangent activation function, ReLU activation function. S44. Define the mean square error loss function and use the Adam algorithm to update the parameters: ; in, represents the mean square error loss function, represents the true value of the i-th sample, represents the predicted value of the i-th sample, N represents the number of samples, Represents the parameters of the improved convolutional long short-term memory network model; S45, the output layer generates the predicted value of the improved convolutional long short-term memory network model: ; in, represents the predicted value of the improved convolutional long short-term memory network model, represents the output layer weight, Represents the output layer bias.

[0027] In this implementation manner, S6 specifically includes: S61. Define the regional correction factor , quantifies the systematic deviation of the prediction error in region i: ; in, represents the correction factor for region i, represents the number of samples in region i, represents the real atmospheric weighted average temperature of the jth sample, represents the predicted value of the fused atmospheric weighted average temperature of the jth sample; S62. Divide region i according to geographical location and meteorological conditions, and each region contains the latitude and longitude coordinate range corresponding to the sample: ; ; in, and Respectively represent the latitude and longitude of the jth sample, , , and Indicates the upper and lower limits of the geographical range of region i; S63, use area correction factor Fusion atmospheric weighted mean temperature forecast Make corrections to generate the predicted value after regional adaptation optimization: ; in, represents the predicted value of atmospheric weighted average temperature after regional adaptation optimization, represents the correction factor of region i to which sample j belongs, represents the predicted value of the fused atmospheric weighted average temperature of the jth sample; S64. Dynamically update the regional correction factor in real-time prediction , and correct it with the new observation data: ; in, represents the updated regional correction factor, represents the historical area correction factor, represents the correction amount based on new observation data, Indicates the update rate.

[0028] Embodiment 1: In order to verify the feasibility of the present invention in implementation, the present invention was applied to a certain actual global navigation satellite system tropospheric delay correction scenario, and ground meteorological observation data and satellite remote sensing data in central China were selected as the main research objects. The data collection range covers the three provinces of Henan, Hubei and Hunan, and the time span is from January to December 2023. Ground observation data include air pressure, temperature and humidity, and remote sensing data provides vertical atmospheric profile information and cloud top temperature. The selected area has typical monsoon climate characteristics, and the meteorological conditions are changeable throughout the year, including high temperature and humidity in summer and low temperature and dryness in winter, which is suitable for evaluating the modeling accuracy and adaptability of the present invention.

[0029] First, the multi-source meteorological data were preprocessed, including time synchronization, spatial alignment, and outlier removal. With the time step of hours, each meteorological characteristic variable was standardized to generate a multidimensional meteorological characteristic input data set. Then, the least squares linear regression model was used to preliminarily model the atmospheric weighted average temperature. By fitting the key meteorological variables extracted by principal component analysis (such as the daily variation of air pressure and the annual cycle of temperature), preliminary prediction results were obtained.

[0030] Subsequently, an improved convolutional long short-term memory network model was introduced for deep learning modeling. In the model design, the convolution layer effectively extracts the features of key areas in the pressure field and temperature field by introducing a spatial attention mechanism, and the long short-term memory network layer strengthens the modeling ability of time series dependencies through a gated attention mechanism. In order to further improve the regional adaptability of the model, we introduced a regional correction factor based on the fusion of the prediction values ​​of the two models. By dividing the geographical regions (such as plains and hilly areas), the prediction errors in each region are dynamically corrected. K in the table represents Kelvin.

[0031] Table 1 Experimental data comparison table ; From the performance comparison table of different models verified by the above experiments in modeling atmospheric weighted average temperature, it can be seen that the improved model proposed in the present invention has significantly better prediction accuracy and adaptability than traditional methods under various meteorological conditions and geographical regions. Specifically, the traditional empirical model is relatively stable in mid-latitude regions, but in areas with complex meteorological conditions such as the tropics and polar regions, the prediction error is large, with the mean square error reaching 4.5K and 5.0K respectively, and the overall regional average error is 4.2K. This shows that the empirical model is subject to its static assumptions and regional applicability limitations, and it is difficult to provide reliable predictions under complex conditions.

[0032] In contrast, the least squares linear regression model significantly reduced the prediction errors in mid-latitudes and tropical regions through principal component analysis and linear fitting of meteorological characteristic variables, with the mean square error in mid-latitudes reduced to 2.8 K and the mean square error in tropical regions to 3.8 K. However, in polar regions, the model still showed certain limitations, with the mean square error only reduced from 5.0 K to 4.5 K, indicating that it is still difficult to capture nonlinear changes under complex meteorological conditions under the linear assumption.

[0033] After the introduction of the improved convolutional long short-term memory network model, the overall prediction performance has been improved. The model enhances the ability to capture spatiotemporal features through the spatial attention mechanism of the convolutional layer and the temporal attention mechanism of the long short-term memory network. The mean square error in the mid-latitude, tropical and polar regions dropped to 1.5K, 2.0K and 2.8K respectively, and the overall regional average MSE was 2.1K. This performance improvement shows that the improved deep learning model performs better than traditional statistical models under complex meteorological conditions, especially in terms of modeling nonlinear and dynamic characteristics.

[0034] Further through model fusion and regional correction optimization, the final prediction results of the present invention have achieved higher accuracy in all regions. For example, in mid-latitudes, the corrected mean square error of the fusion model is reduced from 1.3 K to 1.0K; in tropical regions, it is reduced from 1.8K to 1.5K; in polar regions, it is reduced from 2.5K to 2.0K. The overall regional average mean square error is reduced to 1.5K, which is 59.5% lower than the least squares model and 28.6% lower than the improved convolutional long short-term memory network model. In addition, in tropospheric delay correction, the error of the traditional empirical model is 4.0cm, while the fusion model of the present invention (after regional correction and optimization) reduces the error to 2.1 cm, showing excellent performance in practical applications.

[0035] In summary, the improved method of the present invention has demonstrated excellent performance in atmospheric weighted average temperature modeling through the combination of multiple models, regional correction optimization and the introduction of deep learning technology. It not only improves the prediction accuracy, but also effectively solves the adaptability problem of traditional methods in complex areas.

[0036] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An improved atmospheric weighted average temperature modeling method, characterized in that: The steps include: S1. Obtain multi-source data related to atmospheric weighted average temperature; S2. Preprocessing the multi-source data to generate a multi-dimensional meteorological feature input data set; S3. Construct a least squares linear regression model. Based on the multidimensional meteorological characteristic input data set, use the least squares linear regression model to preliminarily model the atmospheric weighted average temperature, fit the relationship between the atmospheric weighted average temperature and the meteorological characteristic variables, and generate the predicted value of the least squares linear regression model. S4. Construct an improved convolutional long short-term memory network model, use the multidimensional meteorological feature input data set as a training sample, train the convolutional long short-term memory network model, calculate the error between the predicted value and the true value according to the loss function, update the parameters of the improved convolutional long short-term memory network model, and generate the predicted value of the improved convolutional long short-term memory network model; S5, fusion least squares linear regression model and improved convolutional long short-term memory network model prediction value, generate fusion atmospheric weighted average temperature prediction value; S6. Based on the fused atmospheric weighted average temperature prediction value, the fused atmospheric weighted average temperature prediction value is regionally adapted and optimized in combination with the regional correction factor.

2. The improved atmospheric weighted average temperature modeling method according to claim 1, characterized in that: The multi-source data include temperature, humidity, air pressure observed by ground meteorological stations and satellite remote sensing data.

3. The improved atmospheric weighted average temperature modeling method according to claim 1, characterized in that: The preprocessing includes time and space alignment, data cleaning, standardization and feature variable extraction, and the multidimensional meteorological feature input data set includes time features, space features and vertical profile features.

4. The improved atmospheric weighted average temperature modeling method according to claim 1, characterized in that: The S3 specifically includes: S31, obtaining air pressure and temperature data in a multi-dimensional meteorological feature input data set; S32. Construct a least squares linear regression model to fit the atmospheric weighted average temperature Nonlinear relationship between pressure and temperature data: ; ; ; in, represents the weighted mean temperature of the atmosphere, , , , , , and represents the coefficient of the regression model, which is determined by data fitting, P represents air pressure, T represents temperature, DOY represents the day of the year, and Indicates the number of days; S33, using the least squares method to perform parameter fitting on historical meteorological data to obtain optimal model parameters; S34. Calculate the mean square error (MSE) of the least squares linear regression model to evaluate the performance of the least squares linear regression model: ; Among them, MSE represents the mean square error of the improved polynomial regression model, represents the true value of the i-th sample, represents the predicted value of the i-th sample, N represents the number of samples, represents the true value mean; S35. Output the predicted value of the improved polynomial regression model .

5. The improved atmospheric weighted average temperature modeling method according to claim 2, characterized in that: The S4 specifically includes: S41. Construct an improved convolutional long short-term memory network model, including: The input layer receives a multidimensional meteorological feature input data set and sets the dimension of the input tensor to N×Y×M, where N represents the number of samples, Y represents the time step, and M represents the number of meteorological feature variables; The improved convolution layer introduces a spatial attention mechanism to enhance the feature extraction capability of important areas of spatial correlation. The output dimension of the improved convolution layer is N×Y×C, where C in the output dimension of the improved convolution layer represents the number of channels. The improved long short-term memory network layer introduces a gated attention mechanism to enhance the ability to allocate weights to important information at different time steps. The output dimension of the improved long short-term memory network layer is N×H, where H in the output dimension of the improved long short-term memory network layer represents the number of hidden units. Output layer, generates the predicted value of atmospheric weighted average temperature, and the output dimension is N×1; S42, the multi-dimensional meteorological feature input data set is input into the improved convolution layer for feature extraction, and the feature Z extracted by the convolution layer introduces the spatial attention weight: ; in, represents the enhanced features of the convolutional layer output, represents the spatial attention weight matrix, and They represent the convolution kernel weight and bias of the spatial attention mechanism respectively, Sigmoid represents the activation function, represents the convolution calculation, represents the point-by-point product calculation, and Z represents the features extracted by the improved convolutional layer; S43. Adding temporal attention weights to the input gate, forget gate, and output gate formulas of the traditional long short-term memory network layer Form an improved long short-term memory network to enhance the features Input to the improved long short-term memory network: ; ; ; ; ; ; in, , and Represent the activation values ​​of the input gate, forget gate and output gate respectively, with a dimension of H. and Represent the unit state and hidden state respectively, with a dimension of H, , , and Represents the weight matrix input to each gate and state, with a dimension of H×C, , , and represents the bias term, with dimension H, , , and Represents the weight matrix from hidden state to each gate and state, with dimension H×H, represents the time attention weight, with dimension Y, represents the temporal attention weight matrix, with dimension Y×(C+H), represents the temporal attention bias term, with dimension Y, represents the concatenation of the input at time step t and the hidden state at the previous time step, represents the input of time step t, tanh represents the hyperbolic tangent activation function, ReLU activation function. S44. Define the mean square error loss function and use the Adam algorithm to update the parameters: ; in, represents the mean square error loss function, represents the true value of the i-th sample, represents the predicted value of the i-th sample, N represents the number of samples, Represents the parameters of the improved convolutional long short-term memory network model; S45, the output layer generates the predicted value of the improved convolutional long short-term memory network model: ; in, represents the predicted value of the improved convolutional long short-term memory network model, represents the output layer weight, Represents the output layer bias.

6. The improved atmospheric weighted average temperature modeling method according to claim 1, characterized in that: The S6 specifically includes: S61. Define the regional correction factor , quantifies the systematic deviation of the prediction error in region i: ; in, represents the correction factor for region i, represents the number of samples in region i, represents the real atmospheric weighted average temperature of the jth sample, represents the predicted value of the fused atmospheric weighted average temperature of the jth sample; S62. Divide region i according to geographical location and meteorological conditions, and each region contains the latitude and longitude coordinate range corresponding to the sample: ; ; in, and Respectively represent the latitude and longitude of the jth sample, , , and Indicates the upper and lower limits of the geographical range of region i; S63, use area correction factor Fusion atmospheric weighted mean temperature forecast Make corrections to generate the predicted value after regional adaptation optimization: ; in, represents the predicted value of atmospheric weighted average temperature after regional adaptation optimization, represents the correction factor of region i to which sample j belongs, represents the predicted value of the fused atmospheric weighted average temperature of the jth sample; S64. Dynamically update the regional correction factor in real-time prediction , and correct it with the new observation data: ; in, represents the updated regional correction factor, represents the historical area correction factor, represents the correction amount based on new observation data, Indicates the update rate.

Images