Lightning early warning method and system

Abstract

The invention discloses a thunder and lightning early warning method and system, and relates to the technical field of thunder and lightning early warning. Aiming at the characteristics of large electric field data volume and multiple data dimensions, the sparse auto-encoder is utilized to compress the dimension characteristics, and compared with the traditional manual selection characteristics (such as Fourier decomposition, wavelet transform and empirical mode decomposition), the original characteristics of the data can be reserved to the greatest extent, and the accuracy of the data is improved. The problem that electric field time sequence features are difficult to extract in the field of thunder and lightning early warning is well solved, and follow-up processing is facilitated. According to the invention, by using the characteristics of distribution and high efficiency of the LightGBM, the problems of difficulty in modeling, low accuracy and insufficient timeliness of lightning early warning are well solved. According to the invention, adaptive training and debugging can be carried out according to conditions of different regions, and the method has universality. And as the operation time of the model is prolonged and the data volume is increased, updating iteration can be carried out to ensure that the forecasting performance is continuously improved.

Description

technical field [0001] The invention relates to the technical field of lightning early warning, in particular to a lightning early warning method and system. Background technique [0002] Lightning is a discharge phenomenon that occurs between clouds or between clouds and the ground in nature. With the continuous development of economy and the rapid progress of society, the personal safety and property losses caused by lightning have attracted people's attention. As an important measure in active lightning protection, lightning warning is of great significance for reducing the damage caused by lightning. [0003] During the development of thunderstorms, particles such as ice crystals and graupel in the clouds are constantly frictionally electrified, causing them to be charged with a strong charge, thereby forming a strong electric field between the atmosphere and the ground, so the atmospheric electric field can be used to calculate the formation of thunderstorms, process o...

AI Technical Summary

Problems solved by technology

[0004]Traditional lightning warning methods based on the characteristics of the atmospheric electric field are mostly based on artificially setting the threshold of the atmospheric electric field and the difference threshold of the atmospheric electric field, or based on Fourier decomposition and wavelet transform Methods such as empirical mode decomposition and other methods to process electric field time-series signals do not dig deep into the characteristics of atmospheric electric field oscillations, and some characteristics will be lost, which has the problems of poor applicability and low early warning accuracy

If the data sample is large and the amount of calculation is large, it is difficult to meet the needs of actual thunderstorm warning

Images