Lightning disaster early warning method based on meteorological data

Abstract

The invention discloses a lightning disaster early warning method based on meteorological data. The method comprises the following steps: fusing meteorological data and topographic data, and generating a spatial-temporal distribution data set of atmospheric parameters and topographic features in a complex mountainous area environment through weighted stacking and spatial-temporal interpolation methods; analyzing an electromagnetic wave multi-path propagation effect according to the spatial-temporal distribution data set, calculating an attenuation coefficient with a high-frequency component greater than 1 gigahertz and a propagation loss with a low-frequency component less than 1 gigahertz, and generating a spectrum degradation preliminary quantization data set; and if a high-frequency component attenuation coefficient in the spectrum degradation preliminary quantization data set exceeds an attenuation threshold value, optimizing propagation path parameters through a genetic algorithm, generating a corrected spectrum degradation model and the like. The precision and the real-time performance of lightning disaster early warning in the complex mountainous area can be remarkably improved, and the disaster risk is effectively reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of lightning disaster early warning, and in particular to a lightning disaster early warning method based on meteorological data. Background Art

[0002] Lightning disaster warning is a key research direction in the field of disaster prevention and mitigation, directly related to the safety of life and property, and economic and social stability. Especially in complex mountainous environments, where lightning activity is frequent and highly destructive, accurate warnings are crucial for ensuring the safety of transportation, power facilities, and residents in mountainous areas. Currently, lightning warning methods based on meteorological data have achieved certain results in flat terrain or stable meteorological conditions. However, when faced with complex mountainous scenarios, existing methods are clearly insufficiently adaptable. Traditional technologies often ignore the combined effects of terrain undulations, vegetation cover, and dynamic meteorological conditions on lightning signal propagation, resulting in reduced warning accuracy. This is especially true in severe convective weather, where short-term meteorological changes further exacerbate the difficulty of prediction. These limitations make existing methods difficult to meet the actual needs of lightning disaster warnings in mountainous areas.

[0003] In complex mountainous environments, spectrum degradation of electromagnetic wave propagation is a core factor affecting the accuracy of early warnings. Due to terrain obstruction and vegetation absorption, the high-frequency components of electromagnetic waves are significantly attenuated, while the low-frequency components are relatively stable, resulting in signal spectrum distortion, which in turn affects the identification and positioning accuracy of lightning types. More importantly, the dynamic changes in atmospheric humidity and rainfall intensity have a particularly prominent nonlinear effect on high-frequency signals. For example, during localized heavy rainfall, short-term high humidity environments will accelerate the attenuation of high-frequency signals, making it difficult for early warning algorithms based on fixed parameters to accurately distinguish between cloud-to-ground lightning, resulting in positioning errors or missed reports. The complexity of spectrum degradation stems from the multipath effect and the transient nature of meteorological conditions. Existing technologies have difficulty in quantifying the attenuation patterns of signals in different frequency bands in real time, and are unable to dynamically adjust algorithm parameters to adapt to the rapidly changing mountain meteorological environment.

[0004] Therefore, how to integrate multi-source meteorological data such as atmospheric humidity and rainfall intensity in complex mountainous scenarios, accurately quantify the differentiated impact of spectrum degradation on signals in different frequency bands, and dynamically optimize the warning algorithm parameters has become a key issue in improving the accuracy of lightning disaster warnings. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to propose a lightning disaster warning method based on meteorological data, which can significantly improve the accuracy and real-time performance of lightning disaster warning in complex mountainous areas and effectively reduce disaster risks.

[0006] According to one aspect of the present invention, a lightning disaster early warning method based on meteorological data is provided, the method comprising: By integrating meteorological data with topographic data and using weighted overlay and spatiotemporal interpolation methods, we can generate a spatiotemporal distribution dataset of atmospheric parameters and topographic characteristics in complex mountainous environments. Based on the spatiotemporal distribution dataset, the multipath propagation effects of electromagnetic waves were analyzed, the attenuation coefficients of high-frequency components greater than 1 GHz and the propagation losses of low-frequency components less than 1 GHz were calculated, and a preliminary quantitative dataset of spectral degradation was generated. If the attenuation coefficient of the high-frequency component in the preliminary quantified data set of spectrum degradation exceeds the attenuation threshold, the propagation path parameters are optimized by genetic algorithm to generate a modified spectrum degradation model; Based on the modified spectrum degradation model, the atmospheric humidity change rate and rainfall intensity change rate in the dynamic meteorological data are extracted, the nonlinear attenuation coefficient of the high-frequency signal is calculated, and a frequency band differential impact data set is generated; Based on the frequency band differential impact dataset and signal waveform characteristics, lightning types are classified, the characteristic distribution of cloud-to-ground lightning is distinguished, and a lightning type identification dataset and confidence level are generated. If the confidence level of the lightning type recognition dataset is lower than the confidence threshold, the real-time meteorological transient characteristic data is integrated, the support vector machine kernel function and penalty coefficient are adjusted, and an optimized lightning type recognition model is generated; Based on the lightning type and characteristic distribution generated by the optimized lightning type identification model and combined with dynamic meteorological data, the state of the lightning signal propagation path is estimated to generate a lightning location accuracy dataset; Based on the lightning location accuracy dataset, real-time meteorological data and terrain data are updated, the dynamic parameters of the corrected spectrum degradation model are adjusted, and a lightning disaster warning parameter dataset is generated. Based on the lightning disaster warning parameter dataset, it is transmitted to the warning system to complete lightning disaster warning in complex mountainous environments.

[0007] Among the above technical solutions, one is aimed at improving the accuracy and effectiveness of lightning disaster warnings in complex mountainous environments.

[0008] First, meteorological and topographic data are integrated, using weighted overlay and spatiotemporal interpolation methods to generate a spatiotemporal distribution dataset of atmospheric parameters and topographic characteristics in complex mountainous environments. This step fully accounts for the impact of complex mountainous terrain on meteorological elements, constructing a more realistic spatiotemporal distribution. Based on the spatiotemporal distribution dataset, the effects of electromagnetic wave multipath propagation are analyzed, and the attenuation coefficients of high-frequency components greater than 1 GHz and the propagation losses of low-frequency components less than 1 GHz are calculated, generating a preliminary spectral degradation quantification dataset. This analysis helps understand the variations in the propagation characteristics of electromagnetic waves in complex mountainous environments across different frequency bands. When the attenuation coefficient of the high-frequency component in the preliminary spectral degradation quantification dataset exceeds the attenuation threshold, a genetic algorithm is used to optimize the propagation path parameters, generating a revised spectral degradation model. The introduction of the genetic algorithm effectively searches for more optimal propagation path parameters, further improving the model's description of spectral degradation and making it more accurate in describing the complex and changing lightning propagation environment in mountainous areas. Based on the revised spectral degradation model, the atmospheric humidity change rate and rainfall intensity change rate are extracted from the dynamic meteorological data, and the nonlinear attenuation coefficient of the high-frequency signal is calculated to generate a frequency band-differentiated impact dataset. This step, combined with dynamic meteorological characteristics, further explores the differential impact of meteorological factors on signals in different frequency bands, providing more detailed data for subsequent lightning type identification. Based on the frequency band differential impact dataset and signal waveform characteristics, lightning types are classified, the characteristic distributions of cloud-to-ground flashes are clarified, and a lightning type identification dataset and confidence level are generated. If the confidence level falls below a set threshold, real-time meteorological transient characteristic data is integrated, and the support vector machine kernel function and penalty coefficient are adjusted to generate an optimized lightning type identification model. This confidence-based feedback optimization mechanism continuously improves the accuracy and reliability of lightning type identification, ensuring the scientific nature of lightning type judgments. Based on the lightning type and characteristic distribution derived from the optimized lightning type identification model and combined with dynamic meteorological data, a state estimation of the lightning signal propagation path is performed to generate a lightning location accuracy dataset. This step helps accurately determine the propagation trajectory of lightning signals and locate the lightning occurrence location, providing more precise spatial information for subsequent warnings. Based on the lightning location accuracy dataset, real-time meteorological and terrain data are updated, and the dynamic parameters of the modified spectral degradation model are adjusted to generate a lightning disaster warning parameter dataset. This dataset is then transmitted to the warning system, ultimately completing lightning disaster warnings in complex mountainous environments. This entire process forms a closed-loop dynamic update and warning mechanism that can promptly issue accurate lightning disaster warnings based on the latest data and model optimization results, buying valuable time for disaster prevention and mitigation.

[0009] This meteorological data-based lightning disaster warning method integrates closely and collaboratively, from data fusion to model optimization and ultimately to warning implementation. It fully considers the numerous challenges facing lightning disaster warning in complex mountainous environments, such as the impact of terrain on meteorological conditions and electromagnetic wave propagation, and the difficulty in identifying lightning types. By utilizing a variety of advanced algorithms and data processing techniques, it is expected to significantly improve the accuracy and timeliness of lightning disaster warnings, providing strong support for lightning protection efforts in complex mountainous areas and reducing losses caused by lightning disasters.

[0010] In some embodiments, meteorological data and terrain data are integrated to generate a spatiotemporal distribution dataset of atmospheric parameters and terrain characteristics in a complex mountainous environment through weighted superposition and spatiotemporal interpolation methods, including: Obtain atmospheric humidity, rainfall intensity, temperature from meteorological data, and terrain relief and vegetation cover from terrain data, and generate standardized meteorological and terrain datasets through preprocessing; A weighted superposition method is used to assign different weights to atmospheric humidity, rainfall intensity, and temperature, and a comprehensive meteorological impact factor is generated through linear combination to obtain a comprehensive meteorological dataset. By using the spatiotemporal interpolation method, the spatiotemporal distribution characteristics are extracted from the comprehensive meteorological dataset and terrain data to generate a continuous meteorological and terrain spatiotemporal distribution dataset; If the rainfall intensity in the comprehensive meteorological dataset exceeds the intensity threshold, the rainfall intensity is Gaussian smoothed to generate a smoothed meteorological spatiotemporal distribution dataset; Based on the smoothed meteorological and terrain spatiotemporal distribution datasets, the Kriging interpolation method is used to generate the spatiotemporal distribution datasets of atmospheric parameters and terrain characteristics in complex mountainous environments.

[0011] The above technical solution focuses on integrating meteorological data with terrain data to generate a spatiotemporal distribution dataset of atmospheric parameters and terrain characteristics in complex mountainous environments. The entire process uses a variety of methods to process, analyze, and optimize the data, aiming to provide an accurate and spatiotemporally continuous data foundation for subsequent related research or applications.

[0012] First, key elements such as atmospheric humidity, rainfall intensity, and temperature from meteorological data, as well as topographic relief and vegetation cover from topographic data, are obtained and preprocessed to generate standardized meteorological and topographic datasets. This step is fundamental to the entire process. Standardization unifies data from different sources and formats, paving the way for subsequent integration and analysis. Using a weighted overlay method, different weights are assigned to atmospheric humidity, rainfall intensity, and temperature. A comprehensive meteorological impact factor is generated through linear combination to produce a comprehensive meteorological dataset. This weighting is assigned based on the importance of different meteorological elements in their actual environmental impact, enabling a more scientific and comprehensive consideration of the combined effects of these factors, making the resulting comprehensive meteorological dataset more representative and valuable. Spatiotemporal distribution features are extracted from the comprehensive meteorological and topographic datasets, and a continuous spatiotemporal distribution dataset of meteorological and topographic data is generated using spatiotemporal interpolation methods. The application of spatiotemporal interpolation methods compensates for data gaps in time and space, generating a continuous spatiotemporal distribution dataset that provides more complete data support for subsequent research on the spatiotemporal variations of atmospheric parameters and topographic features. If the rainfall intensity in the comprehensive meteorological dataset exceeds the intensity threshold, Gaussian smoothing is performed on the rainfall intensity to generate a smoothed meteorological spatiotemporal distribution dataset. This optimization process is specifically designed for rainfall intensity data. When rainfall intensity is excessively high, Gaussian smoothing can smooth out sudden changes in the data, reducing the impact of extreme values ​​and making the meteorological spatiotemporal distribution dataset more reasonable and stable. Based on the smoothed meteorological and topographic spatiotemporal distribution datasets, the Kriging interpolation method is used to generate a spatiotemporal distribution dataset of atmospheric parameters and topographic features in complex mountainous environments. The Kriging interpolation method fully accounts for characteristics such as the spatial correlation of the data, further improving the accuracy and reliability of the final spatiotemporal distribution dataset, making it more consistent with the actual characteristics of complex mountainous environments.

[0013] The above steps effectively integrated meteorological data with topographic data, and rationally processed and optimized the data, ultimately generating a spatiotemporal distribution dataset of atmospheric parameters and topographic features in a complex mountainous environment.

[0014] In some embodiments, based on the spatiotemporal distribution dataset, analyzing the multipath propagation effect of electromagnetic waves, calculating the attenuation coefficient of high-frequency components greater than 1 GHz and the propagation loss of low-frequency components less than 1 GHz, and generating a preliminary quantified dataset of spectrum degradation includes: Obtain a spatiotemporal distribution data set, use preprocessing methods to clean the data, remove noise points and outliers, and obtain a standardized data set; The electromagnetic wave propagation path is simulated on the standardized data set by ray tracing method, and the reflection, refraction and diffraction paths in the multipath effect are determined to obtain the propagation path set; For a set of propagation paths, the attenuation coefficient of electromagnetic waves with high-frequency components greater than 1 GHz is calculated. If the number of path reflections is greater than a preset threshold, the attenuation is calculated using a geometric optics model to obtain a set of high-frequency attenuation coefficients. Based on the propagation path set, the electromagnetic wave propagation loss of low-frequency components less than 1 GHz is analyzed. If the path diffraction angle is greater than a preset threshold, the uniform geometric diffraction theory is used to calculate the loss to obtain the low-frequency propagation loss set; Extract spectrum degradation features from the set of high-frequency attenuation coefficients and the set of low-frequency propagation losses, and use the fast Fourier transform algorithm to convert the features into the frequency domain to obtain spectrum degradation data; By combining spectrum degradation data with spatiotemporal distribution characteristics, a quantitative data model is constructed, and the spectrum degradation trend is fitted using the least squares method to obtain a quantitative data set. For the quantized data set, data verification is performed. If the data point deviates from the fitting curve by more than a preset threshold, the outlier is removed to obtain the final spectrum degradation quantized data set.

[0015] In the above technical solution, the focus is on how to deeply analyze the multipath propagation effects of electromagnetic waves in complex mountainous environments based on spatiotemporal distribution data sets, and accurately calculate the attenuation and loss of electromagnetic waves in different frequency bands, and finally generate a reliable preliminary quantitative data set of spectrum degradation.

[0016] After acquiring the spatiotemporal distribution dataset, a preprocessing method was used to clean the data, remove noise points and outliers, and generate a standardized dataset. This preprocessing effectively improves data quality and ensures the accuracy and reliability of the input data. Ray tracing was used to simulate the electromagnetic wave propagation paths within the standardized dataset, determining the reflection, refraction, and diffraction paths involved in the multipath effect and generating a set of propagation paths. Ray tracing can simulate multiple electromagnetic wave propagation paths based on the topography and atmospheric parameters of complex mountainous environments, visualizing the abstract electromagnetic wave propagation process and providing a clear set of objects for subsequent electromagnetic wave attenuation and loss calculations for different paths. For this set of propagation paths, the attenuation coefficients of electromagnetic waves with high-frequency components greater than 1 GHz were calculated. If the number of reflections on a path exceeds a preset threshold, a geometric optics model was used to calculate the attenuation, resulting in a set of high-frequency attenuation coefficients. The calculation methods for different scenarios were clearly differentiated. When the number of reflections exceeds the threshold, the geometric optics model more accurately describes the attenuation of high-frequency electromagnetic waves during multiple reflections, making the calculation results more consistent with actual physical conditions. Based on a set of propagation paths, the propagation loss of electromagnetic waves with low-frequency components below 1 GHz is analyzed. If the path diffraction angle is greater than a preset threshold, the uniform geometric diffraction theory is used to calculate the loss, resulting in a low-frequency propagation loss set. Similarly, to address the unique propagation characteristics of low-frequency electromagnetic waves, the uniform geometric diffraction theory is introduced for loss calculation, fully accounting for the diffraction effects of low-frequency electromagnetic waves in complex mountainous terrain. This ensures that the low-frequency propagation loss set accurately reflects the actual propagation loss of low-frequency electromagnetic waves. Spectral degradation features are extracted from the set of high-frequency attenuation coefficients and the set of low-frequency propagation losses. These features are then converted to the frequency domain using the Fast Fourier Transform (FFT) algorithm to generate spectral degradation data. The FFT algorithm converts spectral degradation features from the time domain to the frequency domain, presenting them in a more intuitive and suitable form for further analysis. This provides a convenient data foundation for the subsequent construction of a quantitative data model. Using the spectral degradation data and its spatial and temporal distribution characteristics, a quantitative data model is constructed. The spectral degradation trend is fitted using the least squares method to generate a quantitative data set. The extracted spectrum degradation data is integrated with the spatiotemporal distribution characteristics, and the overall trend of spectrum degradation is fitted using the least squares method. The spectrum degradation phenomenon is quantified in the form of a mathematical model, making the description of spectrum degradation more systematic and regular, and improving the usability and interpretability of the data. Data verification is performed on the quantitative dataset. If the data point deviates from the fitting curve by more than a preset threshold, the outlier is removed to obtain the final spectrum degradation quantification dataset. This data verification process can promptly detect and remove anomalous data points that do not conform to the overall pattern due to various factors, thereby further improving the accuracy and credibility of the spectrum degradation quantification dataset and ensuring that it can truly and reliably reflect the spectrum degradation of electromagnetic waves in complex mountainous environments.

[0017] The above steps revolve around the core goal of accurately analyzing the multipath propagation effects of electromagnetic waves in complex mountainous environments. By applying multiple algorithms to process the attenuation and loss of electromagnetic waves in different frequency bands, the resulting spectral degradation quantification dataset is highly accurate and reliable.

[0018] In some embodiments, if the attenuation coefficient of the high-frequency component in the preliminary quantified spectrum degradation data set exceeds an attenuation threshold, the propagation path parameters are optimized by a genetic algorithm to generate a modified spectrum degradation model, including: Obtain high-frequency components from the spectrum degradation dataset and calculate the attenuation coefficient; if the attenuation coefficient exceeds the attenuation threshold, initialize the propagation path parameters through a genetic algorithm to generate an initial parameter set; Based on terrain data and vegetation coverage data, a propagation environment model is constructed, and the reflection angle and refractive index are adjusted to obtain an optimized parameter set. The parameter set is iteratively optimized using a genetic algorithm, and the propagation environment model is integrated to generate a set of candidate paths. Obtain the optimal propagation path from the candidate path set, calculate the corrected high-frequency component attenuation coefficient, and obtain updated spectrum degradation data; The updated spectrum degradation data is used to construct a revised spectrum degradation model and determine the model parameters. The high-frequency component attenuation coefficient is verified through the revised spectrum degradation model to determine the model accuracy.

[0019] The above technical solution focuses on the specific problem of how to use genetic algorithms to optimize propagation path parameters to generate a corrected spectrum degradation model when the attenuation coefficient of the high-frequency component in the preliminary quantification data set of spectrum degradation exceeds the standard, and provides a complete optimization process.

[0020] High-frequency components are extracted from the spectrum degradation dataset, their attenuation coefficients are calculated, and compared with an attenuation threshold. If the attenuation coefficient exceeds the threshold, a genetic algorithm is used to initialize the propagation path parameters to generate an initial parameter set. A propagation environment model is then constructed based on terrain relief and vegetation cover data, and the reflection angle and refractive index are adjusted to obtain an optimized parameter set. The genetic algorithm is introduced here. The generation of the initial parameter set provides a diverse starting point for subsequent optimization. The construction of the propagation environment model fully considers the impact of complex mountain environments on electromagnetic wave propagation, making the optimization process more realistic. The parameter set is iteratively optimized using the genetic algorithm and integrated with the propagation environment model to generate a set of candidate paths. This iterative optimization process continuously selects better parameter combinations, gradually approaching the optimal solution. The integrated propagation environment model ensures the rationality and feasibility of the optimization parameters, resulting in a set of candidate paths that encompasses a wide range of possible propagation paths, providing a richer selection for subsequent optimal path selection. The optimal propagation path is then obtained from the candidate path set, and the revised high-frequency component attenuation coefficient is calculated to obtain updated spectrum degradation data. This step corrects the spectral degradation data by selecting the optimal path and recalculating the attenuation coefficient, ensuring that the data more accurately reflects actual electromagnetic wave propagation conditions. Using the updated spectral degradation data, a revised spectral degradation model is constructed, model parameters are determined, and the high-frequency component attenuation coefficient is verified against this model to determine model accuracy. Rebuilding the model using the corrected data and verifying its accuracy effectively evaluates the effectiveness of the optimization operation and ensures the revised spectral degradation model has higher accuracy and reliability.

[0021] In some embodiments, based on the modified spectrum degradation model, the atmospheric humidity change rate and rainfall intensity change rate in the dynamic meteorological data are extracted, the nonlinear attenuation coefficient of the high-frequency signal is calculated, and a frequency band differential impact data set is generated, including: Obtain the atmospheric humidity change rate and rainfall intensity change rate from dynamic meteorological data, and use time series analysis method to obtain the time series data of the change rate; Based on the acquired time series data, the Fourier transform method is used to extract the high-frequency signal component and determine the initial amplitude of the high-frequency signal; According to the exponential decay formula ,in is the amplitude after attenuation, is the initial amplitude, is the attenuation coefficient, For the propagation distance, the nonlinear attenuation coefficient is calculated in combination with the preset propagation distance; If the calculated nonlinear attenuation coefficient exceeds the preset range, the attenuation coefficient is optimized using the least square method to obtain an optimized attenuation coefficient; From the optimized attenuation coefficient, the attenuation characteristics corresponding to different frequency ranges are extracted to generate a frequency band differential impact data set; By performing cluster analysis on the frequency band differential impact data set and adopting the K-means algorithm, the attenuation pattern of each frequency range is determined; according to the determined attenuation pattern, a frequency band differential impact data set containing the attenuation coefficient and frequency range is generated.

[0022] In the above technical solution, the solution aims to extract key change rates from dynamic meteorological data based on the modified spectrum degradation model, and then accurately calculate the nonlinear attenuation coefficient of high-frequency signals, and generate a frequency band differentiated impact data set with important reference value.

[0023] The atmospheric humidity change rate and rainfall intensity change rate are obtained from dynamic meteorological data, and their time series data are obtained by time series analysis method. Through time series analysis, the dynamic change trend of atmospheric humidity and rainfall intensity over time can be clearly shown. Based on the acquired time series data, the Fourier transform method is used to extract the high-frequency signal component and determine the initial amplitude of the high-frequency signal. The Fourier transform converts the time series data from the time domain to the frequency domain, so that the high-frequency signal component can be highlighted, and then its initial amplitude can be accurately determined. According to the exponential decay formula , combined with the preset propagation distance, the nonlinear attenuation coefficient is calculated. By substituting known parameters such as the initial amplitude and propagation distance into this formula, the nonlinear attenuation coefficient can be quantitatively calculated, thereby quantitatively assessing the attenuation degree of high-frequency signals during propagation. If the calculated nonlinear attenuation coefficient exceeds the preset range, the least squares method is used to optimize the attenuation coefficient to obtain the optimized attenuation coefficient. When the initial calculation result does not meet expectations or exceeds the reasonable range, the least squares method is used to optimize the attenuation coefficient to ensure that the result is more consistent with actual physical laws and expected model requirements, thereby improving the reliability and usability of the data. From the optimized attenuation coefficient, the attenuation characteristics corresponding to different frequency ranges are extracted to generate a frequency band differential impact dataset. This step, by deeply exploring the optimized attenuation coefficient, classifies and organizes the attenuation characteristics within different frequency ranges, forming a dataset that intuitively reflects the differences in signal attenuation across frequency bands. Cluster analysis of the frequency band differential impact dataset is performed, and the K-means algorithm is used to determine the attenuation pattern of each frequency range. This dataset, which contains the attenuation coefficients and frequency ranges, is then generated. The application of cluster analysis here enables the systematic summary and classification of the attenuation characteristics of different frequency ranges. The K-means algorithm can classify each frequency range into different attenuation pattern categories based on the inherent similarities of the data. The resulting frequency band differential impact dataset not only contains detailed attenuation coefficient information but also clearly defines the attenuation pattern corresponding to each frequency range.

[0024] The above steps start with the processing of dynamic meteorological data. Through a series of methods such as Fourier transform, exponential decay formula calculation, and least squares optimization, they gradually and deeply explore the nonlinear attenuation characteristics of high-frequency signals in different frequency ranges, and finally generate a frequency band differentiated impact data set containing rich information.

[0025] In some embodiments, based on the frequency band differential impact dataset and combined with signal waveform characteristics, lightning types are classified, cloud-to-ground lightning characteristic distributions are distinguished, and a lightning type identification dataset and confidence level are generated, including: The peak amplitude and rise time are separated from the initial feature set to obtain a quantized feature vector. If the peak amplitude of the quantized feature vector exceeds a preset range, the frequency band difference is analyzed using a fast Fourier transform to obtain the frequency domain feature distribution. Based on the frequency domain feature distribution and quantized feature vectors, the input data set of the support vector machine model was constructed to obtain training data. The support vector machine algorithm was used to classify the training data to distinguish between cloud-to-ground lightning and lightning flashes, and the classification results were obtained. The confidence value was calculated from the classification results through cross-validation to obtain the confidence level of the lightning type. According to the classification results and confidence values, a lightning type classification dataset containing cloud-to-ground lightning is generated to obtain the final dataset.

[0026] The aforementioned technical solution focuses on accurately classifying lightning types based on frequency band-differentiated impact datasets and signal waveform characteristics, distinguishing the characteristic distributions of cloud-to-ground lightning, and generating a lightning type identification dataset and confidence level. Through a multi-step process of feature extraction, model building, and validation, the goal is to improve the accuracy and reliability of lightning type identification, providing critical support for lightning disaster warnings.

[0027] The peak amplitude and rise time are separated from the initial feature set to form a quantized feature vector. If the peak amplitude exceeds the preset range, a fast Fourier transform is used to analyze the frequency band differences and obtain the frequency domain feature distribution. The fast Fourier transform is used to convert the time domain features into frequency domain features, fully capturing the frequency characteristics of the lightning signal and making the feature vector more discriminative. Based on the frequency domain feature distribution and quantized feature vector, the input dataset for the support vector machine model is constructed. The support vector machine algorithm is used to classify the training data, distinguishing between cloud-to-ground lightning and lightning flashes, and obtain classification results. The support vector machine algorithm finds the optimal classification hyperplane in a high-dimensional feature space and can effectively handle complex nonlinear classification problems, ensuring the accuracy of the classification results. A confidence value is calculated from the classification results using cross-validation to obtain the confidence level of the lightning type. Cross-validation fully utilizes the dataset, splitting the training and test sets multiple times for model evaluation to obtain stable confidence estimates and enhance the quantitative confidence of the classification results. Based on the classification results and confidence values, a lightning type classification dataset containing both cloud-to-ground lightning flashes is generated. The final dataset not only contains the classification results of lightning types, but also the corresponding confidence values, providing comprehensive and reliable information for subsequent lightning disaster warning and analysis.

[0028] The above steps start with feature extraction, combine frequency domain analysis and time series analysis, build a support vector machine model for classification, and evaluate the confidence through cross-validation, finally generating a detailed lightning type recognition dataset.

[0029] In some embodiments, if the confidence level of the lightning type identification dataset is lower than a confidence threshold, real-time meteorological transient characteristic data is integrated, and the support vector machine kernel function and penalty coefficient are adjusted to generate an optimized lightning type identification model, including: If the confidence level of the lightning type identification dataset is lower than the confidence threshold, the humidity mutation rate and rainfall intensity mutation values ​​are obtained from the real-time meteorological data to generate a fusion feature dataset; According to the fusion feature data set, the principal component analysis method is used to reduce the dimension of the humidity mutation rate and rainfall intensity mutation value to obtain the reduced dimension feature set; If the feature dimension of the dimensionality reduction feature set meets the preset requirements, the kernel function type and penalty coefficient of the support vector machine are adjusted through the grid search method to obtain the optimized parameter combination; According to the optimized parameter combination, the support vector machine algorithm is used to train the reduced dimension feature set to generate the initial lightning type recognition model; Obtain new humidity mutation rate and rainfall intensity mutation values ​​from real-time meteorological data, make predictions using the initial lightning type recognition model, and obtain prediction confidence; If the prediction confidence is lower than the confidence threshold, the initial lightning type recognition model is iteratively optimized by the gradient boosting method to obtain the optimized lightning type recognition model.

[0030] In this technical solution, when the confidence level of a lightning type recognition dataset falls below a threshold, a method is proposed to optimize the support vector machine model by integrating real-time meteorological transient characteristic data. Through feature extraction, dimensionality reduction, parameter optimization, and model iteration, an optimized lightning type recognition model is generated to improve the accuracy and confidence of lightning type recognition.

[0031] When the confidence level of the lightning type identification dataset falls below the confidence threshold, the humidity mutation rate and rainfall intensity mutation values ​​are extracted from real-time meteorological data to generate a fused feature dataset. Real-time meteorological data can reflect environmental changes during lightning events. These mutation features are closely related to lightning type, and when fused, they provide richer information for the model. Principal component analysis is used to reduce the dimensionality of the humidity mutation rate and rainfall intensity mutation values, generating a reduced-dimensionality feature set. Dimensionality reduction removes redundant information and noise, retains key features, and improves model training efficiency and performance, allowing the model to focus more on key features. If the feature dimensions of the reduced-dimensionality feature set meet the preset requirements, a grid search method is used to adjust the kernel function type and penalty coefficient of the support vector machine (SVM) to obtain the optimized parameter combination. This grid search method comprehensively searches the parameter space to find the parameter combination that achieves the best performance on the cross-validation set, thereby improving the model's classification performance. Based on the optimized parameter combination, the reduced-dimensionality feature set is trained using the support vector machine algorithm to generate an initial lightning type identification model. Using the optimized parameters to train the model allows the model to better learn patterns and regularities in the data, improving the performance of the initial model. New humidity mutation rates and rainfall intensity mutation values ​​are obtained from real-time meteorological data. Predictions are then made using the initial lightning type recognition model to determine the prediction confidence. The model's confidence is evaluated by its prediction performance on new data, which can be used to determine the model's reliability in practical applications. If the prediction confidence falls below the confidence threshold, the initial lightning type recognition model is iteratively optimized using the gradient boosting method to obtain an optimized lightning type recognition model. Gradient boosting gradually corrects model errors, improving the model's predictive power and confidence, and further enhancing the accuracy of lightning type recognition.

[0032] The above steps effectively solve the problem of low confidence in the lightning type recognition dataset by integrating real-time meteorological transient characteristic data and combining principal component analysis dimensionality reduction, grid search optimization parameters and gradient boosting iterative optimization methods, thereby improving the performance and reliability of the lightning type recognition model.

[0033] In some embodiments, based on the lightning type and characteristic distribution generated by the optimized lightning type recognition model and combined with dynamic meteorological data, a state estimation of the lightning signal propagation path is performed to generate a lightning location accuracy dataset, including: Acquire real-time signal data from lightning monitoring stations, use a pre-established lightning type recognition model to classify signal features, and obtain lightning types and characteristic distributions; Based on the characteristic distribution and combined with dynamic meteorological data, the spatiotemporal distribution matrix of lightning signals is constructed to determine the initial state of the signal propagation path; The Kalman filter algorithm is used to perform state estimation on the signal propagation path to obtain a path estimation result. If the deviation between the path estimation result and the wind field distribution of the dynamic meteorological data exceeds the deviation threshold, the path estimation result is corrected to obtain a corrected propagation path. The location coordinates of the lightning signal are calculated using the corrected propagation path to obtain a position error dataset. Based on the position error dataset, the signal arrival time difference is analyzed to generate a time resolution dataset. For the time resolution dataset, the least squares method is used to optimize the positioning coordinates to obtain the final lightning location accuracy dataset.

[0034] The above technical solution aims to accurately estimate the state of lightning signal propagation paths based on an optimized lightning type identification model and dynamic meteorological data, thereby generating a high-precision lightning location accuracy dataset. Through multi-step signal processing, state estimation, path correction, and data optimization, the accuracy and reliability of lightning location are ensured, providing key support for lightning disaster warning.

[0035] Real-time signal data from lightning monitoring stations is acquired. A pre-established lightning type recognition model is used to classify signal features and determine the lightning type and characteristic distribution. Real-time signal data from lightning monitoring stations is the fundamental data source for the entire location process. Accurately classifying signal features using the optimized recognition model allows for the precise determination of lightning type (cloud-to-ground lightning) and its characteristic distribution, providing critical prior information for subsequent propagation path estimation. Based on the characteristic distribution and combined with dynamic meteorological data, a spatiotemporal distribution matrix of the lightning signal is constructed to determine the initial state of the signal propagation path. This spatiotemporal distribution matrix comprehensively describes the temporal and spatial distribution of lightning signals. The incorporation of dynamic meteorological data further considers the impact of meteorological conditions on signal propagation, making the initial state determination more accurate to the actual propagation environment. A Kalman filter algorithm is used to estimate the state of the signal propagation path, resulting in a path estimate. If the path estimate deviates from the wind field distribution of the dynamic meteorological data by more than a threshold, the path estimate is corrected to obtain the corrected propagation path. The Kalman filter algorithm updates the path state estimate in real time, effectively addressing noise and uncertainty in signal propagation. When the estimated results deviate significantly from actual meteorological conditions, timely correction can significantly improve the accuracy of the path estimation. Using the corrected propagation path, the lightning signal's location coordinates are calculated to generate a location error dataset. Based on the location error dataset, the signal arrival time difference (TDA) is analyzed to generate a time-resolution dataset. Calculating location coordinates is a key step in determining the lightning's location, and the location error dataset reflects the level of location accuracy. The time-resolution dataset further analyzes the characteristics of the signal arrival time difference (TDA), providing a foundation for subsequent optimization. Least squares optimization is used to optimize the location coordinates within the time-resolution dataset, resulting in the final lightning location accuracy dataset. The least squares method optimizes location coordinates by minimizing the sum of squared errors, effectively improving location accuracy. The resulting lightning location accuracy dataset not only contains the lightning's location information but also reflects the level of location accuracy, providing high-quality data support for lightning disaster warning and analysis.

[0036] The above steps achieve accurate state estimation of the lightning signal propagation path and improve lightning location accuracy by integrating the optimized lightning type identification model and dynamic meteorological data, and using technical means such as Kalman filtering, path correction and least squares optimization.

[0037] In some embodiments, based on the lightning location accuracy dataset, real-time meteorological data and terrain data are updated, and dynamic parameters of the modified spectrum degradation model are adjusted to generate a lightning disaster warning parameter dataset, including: Obtain lightning location accuracy datasets, humidity and rainfall intensity from real-time meteorological data, and undulation and vegetation from terrain data, and fuse them into a unified multidimensional feature dataset; The spectrum degradation model is optimized by the least square method, the attenuation coefficient and path loss are adjusted, and the optimized model parameters are obtained; Based on the optimized model parameters, the propagation characteristics of lightning signals are calculated to generate preliminary warning parameters for lightning disasters. If the signal strength in the preliminary warning parameters exceeds a preset threshold, the potential impact range of the lightning disaster is determined by combining humidity, rainfall intensity and undulation, and vegetation data. A random forest algorithm is used to classify meteorological and topographical features within the potential impact range to determine the intensity level of lightning disasters. Based on the intensity level and impact range, a lightning disaster warning parameter dataset is generated, including the warning range and intensity level. For the generated warning parameter dataset, the spatial interpolation algorithm is applied to optimize the boundary of the warning range and obtain the final lightning disaster warning dataset.

[0038] In the above technical solution, starting from the lightning location accuracy dataset, multi-source data is integrated, and through model optimization, characteristic calculation, disaster analysis and spatial interpolation, an accurate lightning disaster warning parameter dataset is generated, providing key support for disaster prevention and mitigation.

[0039] Integrate lightning location accuracy datasets, real-time meteorological data, and terrain data to construct a multidimensional feature dataset. This fusion of multi-source data captures the physical phenomena associated with lightning disasters, providing a comprehensive perspective for subsequent analysis. Least squares optimization is used to optimize the spectrum degradation model, precisely adjusting parameters such as attenuation coefficient and path loss to improve model performance. Propagation characteristics are calculated based on the optimized model to generate warning parameters. Signal strength is combined with meteorological and terrain data to determine the potential impact range and comprehensively consider disaster-related factors. A random forest algorithm is used to classify features within the potential impact range and determine the disaster intensity level. A spatial interpolation algorithm is used to optimize the warning range boundaries to obtain the final warning dataset. This improves the spatial coherence and accuracy of the warning dataset, enhancing its practical application value.

[0040] In some embodiments, based on the lightning disaster warning parameter dataset, it is transmitted to the warning system to complete the lightning disaster warning in complex mountainous environments, including: Obtain a lightning disaster warning parameter dataset, and obtain a structured dataset through data cleaning and standardization. Based on the structured dataset, use an interpolation algorithm to calculate second-level time resolution data to generate a high-time resolution dataset. For high-temporal-resolution datasets, gridding technology is applied to generate data with a spatial resolution of 100 meters, resulting in a spatiotemporal dataset. If the lightning activity parameters in the spatiotemporal dataset exceed the preset activity threshold, a logistic regression algorithm is used to determine the lightning disaster risk level and generate an early warning signal dataset. Based on the early warning signal dataset, JSON serialization technology is used to generate standard JSON format data to obtain formatted early warning signals. The formatted early warning signals are transmitted to the early warning system through the data transmission protocol to complete signal distribution. In view of the complex mountainous terrain characteristics, a terrain correction algorithm is used to optimize the warning signal to generate a disaster warning signal suitable for complex mountainous areas.

[0041] The above-mentioned technical solution aims to achieve lightning disaster warnings in complex mountainous environments based on a lightning disaster warning parameter dataset through data processing, temporal resolution enhancement, spatial resolution refinement, risk assessment, signal format conversion and transmission, and terrain correction. The entire process closely focuses on precise data processing and efficient transmission, fully considering the impact of complex mountain terrain on warning signals, and ensuring the timeliness, accuracy, and adaptability of warning information.

[0042] A lightning disaster warning parameter dataset was obtained and, through data cleaning and standardization, a structured dataset was generated. Data cleaning removed noise and outliers, while standardization unified data from different sources to the same scale. Based on the structured dataset, an interpolation algorithm was used to calculate data with a temporal resolution of seconds, generating a high-temporal-resolution dataset. The application of the interpolation algorithm made the data more detailed in the temporal dimension, enabling more accurate capture of transient changes in lightning activity, which is crucial for timely detection and early warning of lightning disasters. For the high-temporal-resolution dataset, gridding technology was applied to generate data with a spatial resolution of 100 meters, resulting in a spatiotemporal dataset. This gridding technique divided the study area into 100-meter grid cells, further refining the data in the spatial dimension and enabling more accurate localization of the location and extent of lightning activity. If lightning activity parameters in the spatiotemporal dataset exceeded a preset activity threshold, a logistic regression algorithm was used to determine the lightning disaster risk level and generate a warning signal dataset. The logistic regression algorithm was used to establish a quantitative relationship between lightning activity parameters and disaster risk level, enabling rapid and accurate risk assessment based on real-time data, thus enabling the conversion of data into warning signals. Based on the early warning signal dataset, JSON serialization technology was used to generate data in a standard JSON format, resulting in a formatted early warning signal. This formatted warning signal was then transmitted to the early warning system via a data transmission protocol, completing signal distribution. The JSON format is highly readable and easily parsable, making it suitable as a standard format for early warning signals. This data transmission protocol ensures stable and efficient signal transmission, enabling timely and accurate delivery of warning information to the early warning system, achieving comprehensive coverage in complex mountainous areas. A terrain correction algorithm was used to optimize the warning signal to address the complex terrain characteristics of mountainous areas, generating a disaster warning signal adapted for complex mountainous areas. This terrain correction algorithm fully accounts for the impact of mountainous terrain on warning signal propagation, enabling the warning signal to better adapt to complex terrain and improving the effectiveness and coverage of warning information in mountainous areas.

[0043] The above steps start with the acquisition and preprocessing of lightning disaster warning parameter datasets. Through steps such as improving temporal resolution, refining spatial resolution, judging risk levels, converting and transmitting signal formats, and terrain correction, lightning disaster warnings in complex mountainous environments are achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0045] Figure 1 The present invention is a flowchart of an embodiment of a lightning disaster early warning method based on meteorological data. DETAILED DESCRIPTION

[0046] The present invention will be described in further detail below with reference to the accompanying drawings and examples. It is particularly noted that the following examples are intended only to illustrate the present invention and are not intended to limit the scope of the present invention. Similarly, the following examples are only some embodiments of the present invention and are not intended to be exhaustive. All other embodiments obtained by those of ordinary skill in the art without creative effort are intended to fall within the scope of protection of the present invention.

[0047] The present invention provides a lightning disaster warning method based on meteorological data. It aims to solve the business scenario problems of complex electromagnetic wave propagation paths, difficult lightning type identification and insufficient positioning accuracy caused by undulating mountain terrain, vegetation cover and dynamic meteorological conditions. By integrating the spatiotemporal distribution characteristics of atmospheric humidity, rainfall intensity, temperature and terrain data in meteorological data, a high-precision data set is generated by weighted superposition and spatiotemporal interpolation. The ray tracing method is used to analyze the multipath propagation effect of electromagnetic waves and calculate the high and low frequency attenuation coefficients. If the high-frequency attenuation exceeds the standard, the path parameters are optimized by genetic algorithm, the lightning type identification is optimized by combining support vector machine and grid search, the path state is estimated by integrating Kalman filter, and a lightning positioning accuracy data set with high spatiotemporal resolution is generated. The spectrum degradation model parameters are dynamically adjusted by least squares method, and finally a lightning disaster warning signal data set with second-level time resolution and hundred-meter-level spatial resolution is generated.

[0048] Example 1 See also Figure 1 , a lightning disaster early warning method based on meteorological data, the method comprising: S1. Integrate meteorological data with terrain data and generate a spatiotemporal distribution dataset of atmospheric parameters and terrain characteristics in a complex mountainous environment through weighted superposition and spatiotemporal interpolation methods; In this embodiment, S1, the meteorological data and the terrain data are integrated to generate a spatiotemporal distribution dataset of atmospheric parameters and terrain characteristics in a complex mountainous environment through weighted superposition and spatiotemporal interpolation methods, including: S11, obtaining atmospheric humidity, rainfall intensity, and temperature from meteorological data and terrain relief and vegetation cover from terrain data, and generating standardized meteorological and terrain data sets through preprocessing; S12. Using the weighted superposition method, different weights are assigned to atmospheric humidity, rainfall intensity, and temperature, and a comprehensive meteorological impact factor is generated through linear combination to obtain a comprehensive meteorological data set; S13, extracting spatiotemporal distribution features from the comprehensive meteorological dataset and terrain data by spatiotemporal interpolation method to generate a continuous meteorological and terrain spatiotemporal distribution dataset; S14. If the rainfall intensity in the comprehensive meteorological dataset exceeds the intensity threshold, Gaussian smoothing is performed on the rainfall intensity to generate a smoothed meteorological spatiotemporal distribution dataset; S15. Based on the smoothed meteorological spatiotemporal distribution dataset and the terrain spatiotemporal distribution dataset, a Kriging interpolation method is used to generate a spatiotemporal distribution dataset of atmospheric parameters and terrain characteristics in a complex mountainous environment.

[0049] For example, an initial dataset is first obtained by fusing meteorological data on atmospheric humidity, rainfall intensity, and temperature with terrain relief and vegetation cover from topographic data. Assume that the atmospheric humidity data is hourly relative humidity (%). For example, at 12:00 PM on July 22, 2025, the humidity at a mountainous site is 75%, the rainfall intensity is 5 mm per hour, and the temperature is 25°C. The terrain data uses a digital elevation model (DEM) with a resolution of 30 meters, a point elevation of 1500 meters, a slope of 20 degrees, and a vegetation cover of 0.6 (normalized value). To fuse these data, a weighted overlay method is used, with weights determined based on the parameters' impact on electromagnetic wave propagation: humidity weight 0.4, rainfall intensity weight 0.3, temperature weight 0.2, terrain relief weight 0.2, and vegetation cover weight 0.1. The calculation formula is: Composite value = 0.4 × humidity + 0.3 × rainfall intensity + 0.2 × temperature + 0.2 × terrain relief + 0.1 × vegetation cover. For example, the normalized data for a given point is humidity 0.75, rainfall intensity 0.5 (normalized to 0-1), temperature 0.625 (upper bound of 40°C), terrain relief 0.4 (upper bound of 50-degree slope), and vegetation cover 0.6. The calculated composite value is 0.4 × 0.75 + 0.3 × 0.5 + 0.2 × 0.625 + 0.2 × 0.4 + 0.1 × 0.6 = 0.615. Next, spatiotemporal interpolation is performed to generate a continuously distributed dataset. The inverse distance weighted (IDW) interpolation algorithm is used, with the formula: ,in For known points Parameter value, For distance, is a power (2). Assuming that the distances from an unknown point to three known points are 100 meters, 200 meters, and 300 meters, and the integrated values ​​are 0.615, 0.58, and 0.55, respectively, the interpolated integrated value is calculated as (0.615 / 100^2 + 0.58 / 200^2 + 0.55 / 300^2) / (1 / 100^2 + 1 / 200^2 + 1 / 300^2) ≈ 0.602. Temporal interpolation uses linear interpolation. Assuming the integrated values ​​at 12:00 and 13:00 are 0.602 and 0.620, respectively, the interpolated value at 12:30 is 0.602 + (0.620 - 0.602) × 0.5 = 0.611. The resulting dataset contains integrated values ​​for each grid point (30-meter resolution) with a temporal resolution of 30 minutes.

[0050] S2. Analyze the multipath propagation effects of electromagnetic waves based on the spatiotemporal distribution dataset, calculate the attenuation coefficient of high-frequency components greater than 1 GHz and the propagation loss of low-frequency components less than 1 GHz, and generate a preliminary quantitative dataset of spectrum degradation; In this embodiment, S2 analyzes the multipath propagation effect of electromagnetic waves based on the spatiotemporal distribution dataset, calculates the attenuation coefficient of high-frequency components greater than 1 GHz and the propagation loss of low-frequency components less than 1 GHz, and generates a preliminary quantization dataset of spectrum degradation, including: S21. Obtain a spatiotemporal distribution data set, clean the data using a preprocessing method, remove noise points and outliers, and obtain a standardized data set; S22. Simulating electromagnetic wave propagation paths for the standardized data set using a ray tracing method, determining reflection, refraction, and diffraction paths in a multipath effect, and obtaining a propagation path set; S23. For the propagation path set, calculate the attenuation coefficient of electromagnetic waves with high-frequency components greater than 1 GHz. If the number of path reflections is greater than a preset threshold, calculate the attenuation using a geometric optics model to obtain a set of high-frequency attenuation coefficients. S24. Analyze the electromagnetic wave propagation loss of the low-frequency component less than 1 GHz based on the propagation path set. If the path diffraction angle is greater than a preset threshold, calculate the loss using the uniform geometric diffraction theory to obtain a low-frequency propagation loss set. S25. Extracting spectrum degradation features from the set of high-frequency attenuation coefficients and the set of low-frequency propagation losses, and performing frequency domain conversion on the features using a fast Fourier transform algorithm to obtain spectrum degradation data; S26. Build a quantitative data model based on the spectrum degradation data and the temporal and spatial distribution characteristics, and use the least squares method to fit the spectrum degradation trend to obtain a quantitative data set. S27. Perform data verification on the quantized data set. If the deviation of a data point from the fitting curve exceeds a preset threshold, the abnormal point is removed to obtain a final spectrum degradation quantized data set.

[0051] For example, based on the spatiotemporal distribution data set, the multipath propagation effect of electromagnetic waves is analyzed. First, a three-dimensional environmental model is constructed. The scene is assumed to be an area of ​​1000m×1000m×100m, containing 50 randomly distributed buildings with a height range of 20-80m. The material is concrete (dielectric constant εr=7, conductivity σ=0.01 S / m). Using the ray tracing method, the emission source is set at (500m, 500m, 10m), the transmission power is 1W, and the frequency covers 0.1GHz to 10GHz. The ray tracing algorithm is based on geometric optics and calculates direct, reflected and diffraction paths. The maximum reflection order is 3, and the diffraction adopts the UTD model. The propagation loss of each path is calculated by the free space loss formula Calculate, where is the path length (m), is the frequency (GHz). For high-frequency components (>1GHz), the attenuation coefficient is calculated by multipath superposition, taking into account reflection loss (each reflection attenuates approximately 3-5dB, depending on the angle of incidence, assuming an average of 4dB) and diffraction loss (based on a knife-edge diffraction model, with a typical value of 6-10dB). Taking 5GHz as an example, assume that there are three valid paths to the receiving point (600m, 600m, 5m): the direct path is 141.42m long, with a loss of 20log10(141.42)+20log10(5)+32.44=86.98dB; the single reflection path is 150m long, with a loss of 88.52dB+4dB=92.52dB; and the diffraction path is 160m long, with a loss of 90.08dB+8dB=98.08dB. The total received power is synthesized using the RSSI formula to obtain -90.12dBm, with an attenuation coefficient of approximately 0.9dB / m. For low-frequency components (<1 GHz), taking 0.5 GHz as an example, the loss is primarily affected by free-space propagation and building penetration. Penetration loss is calculated as 10 dB / wall. Assuming two walls, the loss is 20 dB, and the combined path loss is 20 log 10 (141.42) + 20 log 10 (0.5) + 32.44 + 20 = 66.98 dB. The spectrum degradation dataset is generated through simulation and stored in CSV format, containing fields such as frequency, path loss, and attenuation coefficient.

[0052] S3. If the attenuation coefficient of the high-frequency component in the preliminary quantified data set of spectrum degradation exceeds the attenuation threshold, the propagation path parameters are optimized by a genetic algorithm to generate a modified spectrum degradation model; In this embodiment, S3, if the attenuation coefficient of the high-frequency component in the preliminary quantized spectrum degradation data set exceeds a threshold, optimizing the propagation path parameters by a genetic algorithm to generate a modified spectrum degradation model includes: S31, obtaining high-frequency components from the spectrum degradation data set and calculating the attenuation coefficient; if the attenuation coefficient exceeds the attenuation threshold, initializing the propagation path parameters through a genetic algorithm to generate an initial parameter set; S32. Construct a propagation environment model based on the terrain relief data and the vegetation cover data, adjust the reflection angle and the refractive index, and obtain an optimized parameter set; iteratively optimize the parameter set using a genetic algorithm, integrate the propagation environment model, and generate a candidate path set; S33. Obtain an optimal propagation path from the candidate path set, calculate a corrected high-frequency component attenuation coefficient, and obtain updated spectrum degradation data; S34. Using the updated spectrum degradation data, construct a revised spectrum degradation model and determine the model parameters; verify the high-frequency component attenuation coefficient through the revised spectrum degradation model and determine the model accuracy.

[0053] For example, if the attenuation coefficient of the high-frequency component in the preliminary quantified spectral degradation dataset exceeds 0.5 dB / km, a genetic algorithm is used to optimize the propagation path parameters, integrating terrain and vegetation cover data to generate a modified spectral degradation model. First, assume that a high-frequency component (e.g., 2.4 GHz) propagates in a certain area. Preliminary quantified data indicates an attenuation coefficient of 0.6 dB / km, exceeding the threshold of 0.5 dB / km, triggering the optimization process. Terrain data is collected, and a digital elevation model is used to represent the terrain at a 10-meter resolution, obtaining a range of elevation variations (e.g., 0-200 meters). Vegetation cover data is analyzed using remote sensing imagery to quantify coverage (e.g., 40% densely forested areas, with an initial refractive index of 1.0003). The genetic algorithm initializes the population size to 100, with individuals representing reflectance angles (0°-90°) and refractive indices (1.0001-1.0005). The fitness function is defined as minimizing the difference between the attenuation coefficient and the target value of 0.4 dB / km. During the iteration process, the population was updated with a crossover probability of 0.8 and a mutation probability of 0.01. The propagation path loss was calculated for each transmission path. The Fresnel equation was used to assess the impact of reflection angle, and the path loss was adjusted based on the terrain slope (e.g., 5°). After 50 iterations, the reflection angle was optimized to 45°, the refractive index was adjusted to 1.0004, and the attenuation coefficient was reduced to 0.45 dB / km. After integrating terrain and vegetation data, the modified model simulated the propagation path using a ray tracing algorithm, accounting for multipath effects and vegetation absorption loss (e.g., 0.1 dB / m). Spectral degradation curves were generated, verifying that the attenuation coefficient remained stable between 0.4 and 0.45 dB / km. The optimized path parameters effectively reduced high-frequency attenuation, meeting the communication system design requirements. If terrain complexity increases (e.g., a slope exceeding 10°), the mutation probability can be further adjusted to 0.02 to accelerate convergence.

[0054] S4. Extract the atmospheric humidity change rate and rainfall intensity change rate from the dynamic meteorological data based on the modified spectrum degradation model, calculate the nonlinear attenuation coefficient of the high-frequency signal, and generate a frequency band differential impact dataset; In this embodiment, S4, based on the modified spectrum degradation model, extracting the atmospheric humidity change rate and rainfall intensity change rate from the dynamic meteorological data, calculating the high-frequency signal nonlinear attenuation coefficient, and generating a frequency band differentiated impact data set, includes: S41. Obtaining the atmospheric humidity change rate and the rainfall intensity change rate from the dynamic meteorological data, and using a time series analysis method to obtain time series data of the change rates; S42. Based on the acquired time series data, using a Fourier transform method, extract the high-frequency signal component and determine the initial amplitude of the high-frequency signal; S43, according to the exponential decay formula ,in is the amplitude after attenuation, is the initial amplitude, is the attenuation coefficient, is the propagation distance, and the nonlinear attenuation coefficient is calculated in combination with the preset propagation distance; S44. If the calculated nonlinear attenuation coefficient exceeds a preset range, the attenuation coefficient is optimized using a least squares method to obtain an optimized attenuation coefficient; S45. Extracting attenuation characteristics corresponding to different frequency ranges from the optimized attenuation coefficients to generate a frequency band differential impact data set; S46. Perform cluster analysis on the frequency band differential impact data set and use a K-means algorithm to determine the attenuation pattern of each frequency range; and generate a frequency band differential impact data set including an attenuation coefficient and a frequency range based on the determined attenuation pattern.

[0055] As an example, based on the modified spectral degradation model, the atmospheric humidity change rate and rainfall intensity change rate are extracted from dynamic meteorological data, and the nonlinear attenuation coefficient of the high-frequency signal is calculated to generate a frequency band-differentiated impact dataset. Assume that the dynamic meteorological data comes from a real-time meteorological monitoring station in a certain region with a time resolution of 1 hour and contains atmospheric humidity (%) and rainfall intensity (mm / h). First, the humidity change rate and rainfall intensity change rate are extracted from the meteorological data through time series analysis.

[0056] For example, the humidity data sequence is [60, 62, 65, 63]%, and the time interval is 1 hour. The humidity change rate is calculated using the difference method: (62-60) / 1=2% / h, (65-62) / 1=3% / h, (63-65) / 1=-2% / h, and the change rate sequence is [2, 3, -2]% / h. Similarly, the rainfall intensity sequence [0, 2, 5, 3] mm / h is calculated to get the change rate [2, 3, -2] mm / h. Next, based on the exponential decay formula Calculate the nonlinear attenuation coefficient of high-frequency signals Assuming the initial signal amplitude =100, propagation distance = 10km, the frequency band is divided into low frequency (1-5GHz), medium frequency (5-10GHz), and high frequency (10-20GHz). For the humidity change rate of 2% / h and the rainfall intensity change rate of 2 mm / h, assuming It is linearly related to the rate of change of humidity and rainfall intensity: =0.01×|Humidity change rate|+0.02×|Rainfall intensity change rate|, substitute the numerical value =0.01×2+0.02×2=0.06. Then the amplitude after attenuation is =100e^(-0.06×10)=54.88. For different frequency bands, adjust The impact factors, such as high frequency band If it increases by 20%, =0.072, =100e^(-0.072×10)=48.66. The generated data set includes frequency bands, Value and attenuation amplitude: [(1-5GHz, 0.06, 54.88), (5-10GHz, 0.066, 51.61), (10-20GHz,0.072, 48.66)].

[0057] S5. Based on the frequency band differential impact dataset and the signal waveform characteristics, lightning types are classified, the characteristic distribution of cloud-to-ground lightning is distinguished, and a lightning type identification dataset and confidence level are generated. In this embodiment, S5, based on the frequency band differential impact dataset and combined with the signal waveform characteristics, classifies the lightning type, distinguishes the characteristic distribution of cloud-to-ground lightning, and generates a type identification dataset and confidence level, including: S51, separating the peak amplitude and rise time from the initial feature set to obtain a quantized feature vector; if the peak amplitude of the quantized feature vector exceeds a preset range, using a fast Fourier transform to analyze the frequency band difference to obtain a frequency domain feature distribution; S52. Constructing an input data set for a support vector machine model based on the frequency domain feature distribution and the quantized feature vector to obtain training data; classifying the training data using a support vector machine algorithm to distinguish between cloud-to-ground lightning and ground-to-ground lightning to obtain a classification result; and calculating a confidence value from the classification result using a cross-validation method to obtain a confidence level for the lightning type. S53. Generate a lightning type classification dataset including cloud-to-ground lightning and ground-to-cloud lightning according to the classification results and the confidence value, and obtain a final dataset.

[0058] For example, based on a frequency-band-differentiated impact dataset, lightning signals are first preprocessed. Raw signal waveforms for cloud-to-ground flashes and lightning flashes are collected, assuming a sampling frequency of 10 MHz and a time window of 1 ms. These waveforms include peak amplitude and rise time features. Peak amplitude is calculated by detecting the maximum value of the signal. For example, the peak amplitude of cloud-to-ground flashes ranges from 0.5 to 2.0 V, while that of lightning flashes ranges from 1.5 to 5.0 V. Rise time is defined as the time it takes for the signal to transition from 10% to 90% of its peak value. Typical values ​​for cloud-to-ground flashes are 2 to 5 μs, while those for lightning flashes range from 0.5 to 2 μs. Next, frequency band features are extracted. Fast Fourier Transform (FFT) is used to convert the time domain signal into the frequency domain. The energy distribution in the 0-1 MHz low-frequency band and the 1-5 MHz high-frequency band is analyzed. Low-frequency energy accounts for approximately 70% of cloud-to-ground flashes, while high-frequency energy accounts for approximately 60% of ground-to-ground flashes. To enhance feature differentiation, the ratio of peak amplitude to rise time is calculated as a combined feature. For example, the ratio for cloud-to-ground flashes ranges from 0.1 to 0.4 V / μs, while that for lightning flashes ranges from 0.75 to 10 V / μs. Classification was performed using the Support Vector Machine (SVM) algorithm, using a radial basis function (RBF) kernel, parameters C set to 1.0, and γ set to 0.1, optimized using grid search. The training dataset consisted of 1,000 samples (600 cloud-to-ground flashes and 400 ground-to-ground flashes), and the test set consisted of 200 samples. Using 5-fold cross-validation, the classification accuracy reached 92%. The confidence score was output by the SVM decision function and mapped to the [0,1] interval. For example, the confidence score for cloud-to-ground flashes was 0.85, and for ground-to-ground flashes was 0.95. The resulting identification dataset consisted of feature vectors (peak amplitude, rise time, frequency band energy ratio, and combined features) and classification labels (cloud-to-ground flashes / ground-to-ground flashes), along with an additional confidence column.

[0059] S6. If the confidence level of the lightning type identification dataset is lower than the confidence threshold, the real-time meteorological transient characteristic data is integrated, the kernel function and the penalty coefficient are adjusted, and an optimized lightning type identification model is generated; In this embodiment, S6, if the confidence level of the lightning type identification dataset is lower than the confidence threshold, the real-time meteorological transient characteristic data is integrated, the support vector machine kernel function and the penalty coefficient are adjusted, and an optimized lightning type identification model is generated, including: S61. If the confidence level of the lightning type identification dataset is lower than the confidence threshold, obtain the humidity mutation rate and rainfall intensity mutation value from the real-time meteorological data to generate a fused feature dataset; S62. Based on the fused feature data set, a principal component analysis method is used to perform dimensionality reduction processing on the humidity mutation rate and rainfall intensity mutation values ​​to obtain a dimensionality reduction feature set; S63. If the feature dimension of the dimensionality reduction feature set meets the preset requirements, the kernel function type and penalty coefficient of the support vector machine are adjusted by a grid search method to obtain an optimized parameter combination; S64. Based on the optimized parameter combination, a support vector machine algorithm is used to train the reduced dimension feature set to generate an initial lightning type recognition model; S65. Obtain new humidity mutation rate and rainfall intensity mutation values ​​from real-time meteorological data, perform predictions using the initial lightning type recognition model, and obtain prediction confidence. S66. If the prediction confidence is lower than the confidence threshold, the initial lightning type recognition model is iteratively optimized by a gradient boosting method to obtain an optimized lightning type recognition model.

[0060] For example, after receiving a lightning type identification dataset, the system first evaluates the confidence of each sample. Assuming the dataset contains 1,000 lightning event samples, the confidence is calculated using a softmax function, ranging from 0 to 1. If the sample confidence is lower than 0.9, for example, a sample with a confidence of 0.85, the data fusion process is triggered. The system extracts humidity mutation rate and rainfall intensity mutation values ​​from a real-time meteorological database, assuming a humidity mutation rate of 2.5% per minute and a rainfall intensity mutation value of 3.2 mm per hour. This data is retrieved via an API in JSON format with a 1-minute time resolution to ensure alignment with the lightning event timestamp. The fusion process uses a weighted average method to concatenate the lightning feature vector (including voltage and current characteristics) with the meteorological feature vector (humidity mutation rate and rainfall intensity mutation value), with weights of 0.6 and 0.4, respectively. After normalization, a new feature vector is formed. Next, the system uses a grid search to optimize the support vector machine (SVM) model. Candidate kernel functions include linear, RBF, and polynomial kernels, and the penalty coefficient C is searched within the range of [0.1, 1, 10, 100]. This grid search is based on 5-fold cross-validation, and the evaluation metric is the F1 score. Assuming that the RBF kernel achieves the highest F1 score of 0.92 when C=10, the system generates the optimized SVM model and saves it as a .pkl file. It then classifies and predicts new lightning events, outputting the type (e.g., positive or negative) and the confidence level. If the F1 score falls short of expectations (e.g., below 0.9), the system automatically triggers a second grid search, narrowing the range of C to [5, 15] with a step size of 1. The optimization is repeated until the threshold is met.

[0061] S7. Based on the lightning type and characteristic distribution generated by the optimized lightning type identification model and combined with dynamic meteorological data, the state of the lightning signal propagation path is estimated to generate a lightning location accuracy dataset; In this embodiment, S7, based on the lightning type and characteristic distribution generated by the optimized lightning type identification model, combined with dynamic meteorological data, performs state estimation on the lightning signal propagation path to generate a lightning location accuracy dataset, including: S71. Acquire real-time signal data from a lightning monitoring site, classify signal features using a pre-established lightning type recognition model, and obtain lightning types and feature distributions. S72. Based on the characteristic distribution and combined with dynamic meteorological data, construct a spatiotemporal distribution matrix of the lightning signal and determine the initial state of the signal propagation path; S73. Using a Kalman filter algorithm, perform state estimation on the signal propagation path to obtain a path estimation result; if the deviation between the path estimation result and the wind field distribution of the dynamic meteorological data exceeds a deviation threshold, correct the path estimation result to obtain a corrected propagation path; S74. Calculate the location coordinates of the lightning signal using the corrected propagation path to obtain a location error dataset; analyze the signal arrival time difference based on the location error dataset to generate a time resolution dataset; S75. For the time resolution dataset, the least squares method is used to optimize the positioning coordinates to obtain the final lightning location accuracy dataset.

[0062] For example, based on the optimized lightning type recognition model, a convolutional neural network (CNN) was first used to process lightning electromagnetic signals. The input consisted of 1,000 lightning signal samples, each consisting of a time series (sampling rate 10 MHz, duration 1 ms) and spectral features (0-500 kHz). The model used a Reluctant Unit (ReLU) activation function and Dropout (0.3) regularization. After training, the model achieved an accuracy of 95% for identifying intra-thundercloud lightning, cloud-to-ground lightning, and positive-polarity lightning. Feature distribution analysis revealed that the peak current of cloud-to-ground lightning averaged 30 kA, with spectral energy concentrated in the 50-100 kHz range. Combined with dynamic meteorological data such as wind speed (5 m / s), humidity (80%), and temperature (25°C), a multidimensional linear regression model was established to correlate lightning type with meteorological parameters. The correlation coefficient (R²) was 0.85, indicating a significant influence of humidity on lightning type. Next, a Kalman filter algorithm was used to estimate the state of the lightning signal propagation path. The inputs were an initial position error of 100 m, a velocity vector of 0.1 m / s, a process noise covariance Q of 0.01, and an observation noise covariance R of 0.1. After 50 iterations of the algorithm, the position state estimation converged, and the root mean square error (RMSE) of the output path trajectory was 15 m. This generated a lightning location accuracy dataset consisting of 1000 data sets, each including latitude and longitude (resolution 0.001°), altitude (error ±10 m), and timestamp (resolution 0.1 ms). Analysis showed that the temporal resolution was limited by the signal sampling rate, and the positioning error was related to meteorological conditions, with the error slightly increasing (approximately 5%) with higher humidity. Dataset statistics showed that 90% of the positioning errors were less than 20 m, and the temporal resolution reached 0.12 ms, meeting real-time monitoring requirements. To ensure logical rigor, after the positioning data was fused with the meteorological data, the stability of the path estimation was further verified by the support vector machine (SVM). The classification accuracy reached 92%, indicating that the Kalman filter results were reliable.

[0063] S8. Based on the lightning location accuracy dataset, update the real-time meteorological data and terrain data, adjust the dynamic parameters of the corrected spectrum degradation model, and generate a lightning disaster warning parameter dataset; based on the lightning disaster warning parameter dataset, transmit it to the warning system to complete the lightning disaster warning in complex mountainous environments.

[0064] In this embodiment, S81, based on the lightning location accuracy dataset, updates the real-time meteorological data and terrain data, adjusts the dynamic parameters of the modified spectrum degradation model, and generates a disaster warning parameter dataset, including: S811. Obtain a lightning location accuracy dataset, humidity and rainfall intensity from real-time meteorological data, and terrain data including undulation and vegetation, and fuse them into a unified multidimensional feature dataset. S812. Optimizing the spectrum degradation model by using a least squares method, adjusting the attenuation coefficient and the path loss, and obtaining optimized model parameters; S813. Calculate the propagation characteristics of lightning signals based on the optimized model parameters to generate preliminary warning parameters for lightning disasters. If the signal strength in the preliminary warning parameters exceeds a preset threshold, determine the potential impact range of the lightning disaster by combining humidity, rainfall intensity, undulation, and vegetation data. S814. Use the random forest algorithm to classify meteorological and topographical features within the potential impact range to determine the intensity level of the lightning disaster; generate a lightning disaster warning parameter dataset based on the intensity level and impact range, including the warning range and intensity level; S815. Apply a spatial interpolation algorithm to the generated warning parameter dataset to optimize the boundary of the warning range and obtain a final lightning disaster warning dataset.

[0065] For example, based on the lightning location accuracy dataset, the lightning location is first obtained through a high-precision lightning location system. Assuming the location accuracy is 100 meters, the data contains latitude and longitude coordinates (such as 39.9°N, 116.4°E) and timestamps. Humidity data is extracted from a real-time meteorological database. Assuming the humidity in a certain area is 75%, the humidity is gridded using an interpolation algorithm (linear interpolation) with a grid resolution of 1 km. The calculation formula is: ,in 、 is the humidity value of the neighboring station. The rainfall intensity is obtained through Doppler radar. Assuming that the rainfall intensity at a certain point is 10 mm / h, the Kriging interpolation method is used to generate a continuous rainfall intensity field. The formula is , weight Calculated by variogram. The terrain relief is extracted from the digital elevation model (DEM), assuming that the maximum relief in a certain area is 200 meters. The vegetation coverage is calculated using remote sensing NDVI data, assuming it is 0.6, and then normalized for model input. The spectrum degradation model is adjusted based on the least squares method, with the initial value of the attenuation coefficient α being 0.02 and the path loss is 80 decibels, the model is ,in For distance, is the reference loss, is the path loss exponent (set to 2.5). Minimize the error by the least squares method , iterative update α is 0.025, L is 82 decibels. Generate lightning disaster warning parameters. The warning range is based on the lightning location point and extends to a radius of 5 kilometers. The intensity level is calculated based on the rainfall intensity and attenuation coefficient. The formula is: , assuming I=5.5, it is classified as a medium warning.

[0066] In this embodiment, S82, based on the lightning disaster warning parameter data set, transmits it to the warning system to complete the lightning disaster warning in complex mountainous environment, including: S821. Obtain a lightning disaster warning parameter dataset, and obtain a structured dataset through data cleaning and standardization. Based on the structured dataset, use an interpolation algorithm to calculate second-level time resolution data to generate a high-time resolution dataset. S822. Applying gridding technology to the high temporal resolution dataset to generate data with a spatial resolution of 100 meters to obtain a spatiotemporal dataset. If the lightning activity parameters in the spatiotemporal dataset exceed a preset activity threshold, a logistic regression algorithm is used to determine the lightning disaster risk level and generate an early warning signal dataset. S823. Based on the warning signal dataset, use JSON serialization technology to generate standard JSON format data to obtain a formatted warning signal; transmit the formatted warning signal to the warning system through a data transmission protocol to complete signal distribution; S824. Based on the complex mountainous terrain characteristics, a terrain correction algorithm is used to optimize the warning signal to generate a disaster warning signal suitable for complex mountainous areas.

[0067] For example, generating a warning signal dataset with a temporal resolution of seconds and a spatial resolution of 100 meters based on a lightning disaster warning parameter dataset requires multi-step processing to achieve lightning disaster warnings in complex mountainous environments and transmit it in a standard JSON format. First, assume that the input dataset contains the lightning occurrence time (accurate to milliseconds, such as 2025-07-22 18:39:45.123), longitude and latitude coordinates (such as 114.1234°E, 22.5678°N), lightning intensity (in units of peak current kA, such as 30.5 kA), and terrain height (such as 500.3 meters). To achieve second-level temporal resolution, a temporal interpolation algorithm was used to aggregate millisecond-level data to the second level. This was done by taking the average of all lightning events within each second. For example, if three lightning events within a given second had timestamps of 18:39:45.123, 18:39:45.456, and 18:39:45.789, their average time was calculated as 18:39:45.456, and the intensity was averaged (if the timestamps were 30.5 kA, 32.1 kA, and 31.2 kA, respectively, the average would be 31.27 kA). Spatial resolution was processed using a gridding method, dividing the study area into 100 m × 100 m grids. Lightning events were assigned to corresponding grids based on latitude and longitude. If there were no events within a grid, the intensity was marked as 0. If there were multiple events, the maximum intensity was taken. For example, if two lightning events within a grid had intensities of 31.27 kA and 33.4 kA, 33.4 kA would be recorded. To adapt to complex mountainous environments, a terrain correction factor is introduced. Slope is calculated based on a digital elevation model (DEM) (for example, slope arctan (Δh / Δd), where Δh is a 50-meter elevation difference, Δd is a 100-meter horizontal distance, and a slope of 26.57°). The lightning risk factor for grids with slopes greater than 15° is increased by 20%, for example, from 33.4 kA to 40.08 kA. Warning signals are generated using a threshold method, with lightning intensities greater than 25 kA and greater than 30 kA after terrain adjustment considered high risk. Warning levels are then generated (high, medium, or low), with 40.08 kA corresponding to high risk. The final dataset is encapsulated in JSON format, for example, {“timestamp”:“2025-07-22 18:39:45”, “grid_id”:“114.1234_22.5678”, “intensity”:40.08, “risk_level”:“high”}, and transmitted to the early warning system through the API interface. The system parses the JSON and triggers a regional alarm.

[0068] The above descriptions are only some embodiments of the present invention and do not limit the scope of protection of the present invention. Any equivalent device or equivalent process transformation made by using the contents of the description and drawings of the present invention, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A lightning disaster early warning method based on meteorological data, characterized in that: The method comprises: By integrating meteorological data with topographic data and using weighted overlay and spatiotemporal interpolation methods, we can generate a spatiotemporal distribution dataset of atmospheric parameters and topographic characteristics in complex mountainous environments. Based on the spatiotemporal distribution dataset, the multipath propagation effects of electromagnetic waves were analyzed, the attenuation coefficients of high-frequency components greater than 1 GHz and the propagation losses of low-frequency components less than 1 GHz were calculated, and a preliminary quantitative dataset of spectral degradation was generated. If the attenuation coefficient of the high-frequency component in the preliminary quantified data set of spectrum degradation exceeds the attenuation threshold, the propagation path parameters are optimized by genetic algorithm to generate a modified spectrum degradation model; Based on the modified spectrum degradation model, the atmospheric humidity change rate and rainfall intensity change rate in the dynamic meteorological data are extracted, the nonlinear attenuation coefficient of the high-frequency signal is calculated, and a frequency band differential impact data set is generated; Based on the frequency band differential impact dataset and signal waveform characteristics, lightning types are classified, the characteristic distribution of cloud-to-ground lightning is distinguished, and a lightning type identification dataset and confidence level are generated. If the confidence level of the lightning type recognition dataset is lower than the confidence threshold, the real-time meteorological transient characteristic data is integrated, the support vector machine kernel function and penalty coefficient are adjusted, and an optimized lightning type recognition model is generated; Based on the lightning type and characteristic distribution generated by the optimized lightning type identification model and combined with dynamic meteorological data, the state of the lightning signal propagation path is estimated to generate a lightning location accuracy dataset; Based on the lightning location accuracy dataset, real-time meteorological data and terrain data are updated, the dynamic parameters of the corrected spectrum degradation model are adjusted, and a lightning disaster warning parameter dataset is generated. Based on the lightning disaster warning parameter dataset, it is transmitted to the warning system to complete lightning disaster warning in complex mountainous environments.

2. The lightning disaster early warning method based on meteorological data according to claim 1, characterized in that: By integrating meteorological data with terrain data and using weighted overlay and spatiotemporal interpolation methods, we can generate a spatiotemporal distribution dataset of atmospheric parameters and terrain characteristics in complex mountainous environments, including: Obtain atmospheric humidity, rainfall intensity, temperature from meteorological data, and terrain relief and vegetation cover from terrain data, and generate standardized meteorological and terrain datasets through preprocessing; A weighted superposition method was used to assign different weights to atmospheric humidity, rainfall intensity, and temperature, and a comprehensive meteorological impact factor was generated through linear combination to obtain a comprehensive meteorological dataset. By using the spatiotemporal interpolation method, the spatiotemporal distribution characteristics are extracted from the comprehensive meteorological dataset and terrain data to generate a continuous meteorological and terrain spatiotemporal distribution dataset; If the rainfall intensity in the comprehensive meteorological dataset exceeds the intensity threshold, the rainfall intensity is Gaussian smoothed to generate a smoothed meteorological spatiotemporal distribution dataset; Based on the smoothed meteorological and terrain spatiotemporal distribution datasets, the Kriging interpolation method is used to generate the spatiotemporal distribution datasets of atmospheric parameters and terrain characteristics in complex mountainous environments.

3. The lightning disaster early warning method based on meteorological data according to claim 1, characterized in that: Based on the spatiotemporal distribution dataset, the multipath propagation effects of electromagnetic waves are analyzed, the attenuation coefficients of high-frequency components greater than 1 GHz and the propagation losses of low-frequency components less than 1 GHz are calculated, and a preliminary quantitative dataset of spectrum degradation is generated, including: Obtain a spatiotemporal distribution data set, use preprocessing methods to clean the data, remove noise points and outliers, and obtain a standardized data set; The electromagnetic wave propagation path is simulated on the standardized data set by ray tracing method, and the reflection, refraction and diffraction paths in the multipath effect are determined to obtain the propagation path set; For a set of propagation paths, the attenuation coefficient of electromagnetic waves with high-frequency components greater than 1 GHz is calculated. If the number of path reflections is greater than a preset threshold, the attenuation is calculated using a geometric optics model to obtain a set of high-frequency attenuation coefficients. Based on the propagation path set, the electromagnetic wave propagation loss of low-frequency components less than 1 GHz is analyzed. If the path diffraction angle is greater than a preset threshold, the uniform geometric diffraction theory is used to calculate the loss to obtain the low-frequency propagation loss set; Extract spectrum degradation features from the set of high-frequency attenuation coefficients and the set of low-frequency propagation losses, and use the fast Fourier transform algorithm to convert the features into the frequency domain to obtain spectrum degradation data; By combining spectrum degradation data with spatiotemporal distribution characteristics, a quantitative data model is constructed, and the spectrum degradation trend is fitted using the least squares method to obtain a quantitative data set. For the quantized data set, data verification is performed. If the data point deviates from the fitting curve by more than a preset threshold, the outlier is removed to obtain the final spectrum degradation quantized data set.

4. The lightning disaster early warning method based on meteorological data according to claim 1, characterized in that: If the attenuation coefficient of the high-frequency component in the preliminary quantified spectral degradation data set exceeds the attenuation threshold, the propagation path parameters are optimized using a genetic algorithm to generate a modified spectral degradation model, including: Obtain high-frequency components from the spectrum degradation dataset and calculate the attenuation coefficient; if the attenuation coefficient exceeds the attenuation threshold, initialize the propagation path parameters through a genetic algorithm to generate an initial parameter set; Based on terrain data and vegetation coverage data, a propagation environment model is constructed, and the reflection angle and refractive index are adjusted to obtain an optimized parameter set. The parameter set is iteratively optimized using a genetic algorithm, and the propagation environment model is integrated to generate a set of candidate paths. Obtain the optimal propagation path from the candidate path set, calculate the corrected high-frequency component attenuation coefficient, and obtain updated spectrum degradation data; The updated spectrum degradation data is used to construct a revised spectrum degradation model and determine the model parameters. The high-frequency component attenuation coefficient is verified through the revised spectrum degradation model to determine the model accuracy.

5. The lightning disaster early warning method based on meteorological data according to claim 1, characterized in that: Based on the modified spectrum degradation model, the atmospheric humidity change rate and rainfall intensity change rate in the dynamic meteorological data are extracted, the nonlinear attenuation coefficient of the high-frequency signal is calculated, and a frequency band differential impact data set is generated, including: Obtain the atmospheric humidity change rate and rainfall intensity change rate from dynamic meteorological data, and use time series analysis method to obtain the time series data of the change rate; Based on the acquired time series data, the Fourier transform method is used to extract the high-frequency signal component and determine the initial amplitude of the high-frequency signal; According to the exponential decay formula ,in is the amplitude after attenuation, is the initial amplitude, is the attenuation coefficient, For the propagation distance, the nonlinear attenuation coefficient is calculated in combination with the preset propagation distance; If the calculated nonlinear attenuation coefficient exceeds the preset range, the attenuation coefficient is optimized using the least square method to obtain an optimized attenuation coefficient; From the optimized attenuation coefficient, the attenuation characteristics corresponding to different frequency ranges are extracted to generate a frequency band differential impact data set; By performing cluster analysis on the frequency band differential impact data set and adopting the K-means algorithm, the attenuation pattern of each frequency range is determined; according to the determined attenuation pattern, a frequency band differential impact data set containing the attenuation coefficient and frequency range is generated.

6. The lightning disaster early warning method based on meteorological data according to claim 1, characterized in that: Based on the frequency band differential impact dataset and signal waveform characteristics, lightning types are classified, cloud-to-ground lightning distribution characteristics are distinguished, and a lightning type identification dataset and confidence level are generated, including: The peak amplitude and rise time are separated from the initial feature set to obtain a quantized feature vector. If the peak amplitude of the quantized feature vector exceeds a preset range, the frequency band difference is analyzed using a fast Fourier transform to obtain the frequency domain feature distribution. Based on the frequency domain feature distribution and quantized feature vectors, the input data set of the support vector machine model was constructed to obtain training data. The support vector machine algorithm was used to classify the training data to distinguish between cloud-to-ground lightning and lightning flashes, and the classification results were obtained. The confidence value was calculated from the classification results through cross-validation to obtain the confidence level of the lightning type. According to the classification results and confidence values, a lightning type classification dataset containing cloud-to-ground lightning is generated to obtain the final dataset.

7. The lightning disaster early warning method based on meteorological data according to claim 1, characterized in that: If the confidence level of the lightning type identification dataset is lower than the confidence threshold, the real-time meteorological transient characteristic data is integrated, the support vector machine kernel function and penalty coefficient are adjusted, and an optimized lightning type identification model is generated, including: If the confidence level of the lightning type identification dataset is lower than the confidence threshold, the humidity mutation rate and rainfall intensity mutation values ​​are obtained from the real-time meteorological data to generate a fusion feature dataset; According to the fusion feature data set, the principal component analysis method is used to reduce the dimension of the humidity mutation rate and rainfall intensity mutation value to obtain the reduced dimension feature set; If the feature dimension of the dimensionality reduction feature set meets the preset requirements, the kernel function type and penalty coefficient of the support vector machine are adjusted through the grid search method to obtain the optimized parameter combination; According to the optimized parameter combination, the support vector machine algorithm is used to train the reduced dimension feature set to generate the initial lightning type recognition model; Obtain new humidity mutation rate and rainfall intensity mutation values ​​from real-time meteorological data, make predictions using the initial lightning type recognition model, and obtain prediction confidence; If the prediction confidence is lower than the confidence threshold, the initial lightning type recognition model is iteratively optimized by the gradient boosting method to obtain the optimized lightning type recognition model.

8. The lightning disaster early warning method based on meteorological data according to claim 1, characterized in that: Based on the lightning types and characteristic distribution generated by the optimized lightning type identification model and combined with dynamic meteorological data, the state of the lightning signal propagation path is estimated to generate a lightning location accuracy dataset, including: Acquire real-time signal data from lightning monitoring stations, use a pre-established lightning type recognition model to classify signal features, and obtain lightning types and characteristic distributions; Based on the characteristic distribution and combined with dynamic meteorological data, the spatiotemporal distribution matrix of lightning signals is constructed to determine the initial state of the signal propagation path; The Kalman filter algorithm is used to perform state estimation on the signal propagation path to obtain a path estimation result. If the deviation between the path estimation result and the wind field distribution of the dynamic meteorological data exceeds the deviation threshold, the path estimation result is corrected to obtain a corrected propagation path. The location coordinates of the lightning signal are calculated using the corrected propagation path to obtain a position error dataset. Based on the position error dataset, the signal arrival time difference is analyzed to generate a time resolution dataset. For the time resolution dataset, the least squares method is used to optimize the positioning coordinates to obtain the final lightning location accuracy dataset.

9. The lightning disaster early warning method based on meteorological data according to claim 1, characterized in that: Based on the lightning location accuracy dataset, real-time meteorological and terrain data are updated, and the dynamic parameters of the corrected spectrum degradation model are adjusted to generate a lightning disaster warning parameter dataset, including: Obtain lightning location accuracy datasets, humidity and rainfall intensity from real-time meteorological data, and undulation and vegetation from terrain data, and fuse them into a unified multidimensional feature dataset; The spectrum degradation model is optimized by the least square method, the attenuation coefficient and path loss are adjusted, and the optimized model parameters are obtained; Based on the optimized model parameters, the propagation characteristics of lightning signals are calculated to generate preliminary warning parameters for lightning disasters. If the signal strength in the preliminary warning parameters exceeds a preset threshold, the potential impact range of the lightning disaster is determined by combining humidity, rainfall intensity and undulation, and vegetation data. A random forest algorithm is used to classify meteorological and topographical features within the potential impact range to determine the intensity level of lightning disasters. Based on the intensity level and impact range, a lightning disaster warning parameter dataset is generated, including the warning range and intensity level. For the generated warning parameter dataset, the spatial interpolation algorithm is applied to optimize the boundary of the warning range and obtain the final lightning disaster warning dataset.

10. The lightning disaster early warning method based on meteorological data according to claim 1, characterized in that: Based on the lightning disaster warning parameter data set, it is transmitted to the warning system to complete the lightning disaster warning in complex mountainous environments, including: Obtain a lightning disaster warning parameter dataset, and obtain a structured dataset through data cleaning and standardization. Based on the structured dataset, use an interpolation algorithm to calculate second-level time resolution data to generate a high-time resolution dataset. For high-temporal-resolution datasets, gridding technology is applied to generate data with a spatial resolution of 100 meters, resulting in a spatiotemporal dataset. If the lightning activity parameters in the spatiotemporal dataset exceed the preset activity threshold, a logistic regression algorithm is used to determine the lightning disaster risk level and generate an early warning signal dataset. Based on the early warning signal dataset, JSON serialization technology is used to generate standard JSON format data to obtain formatted early warning signals. The formatted early warning signals are transmitted to the early warning system through the data transmission protocol to complete signal distribution. In view of the complex mountainous terrain characteristics, a terrain correction algorithm is used to optimize the warning signal to generate a disaster warning signal suitable for complex mountainous areas.

Images