Thunder and lightning early warning and forecasting method and system

Abstract

The invention provides a thunder and lightning early warning and forecasting method, which comprises the following steps of S1, determining a forecasting starting moment, and acquiring thunder and lightning mode product data and thunder and lightning observation data in a to-be-forecasted time period after the forecasting starting moment, and meteorological observation data before the forecasting starting moment, S2, repeating the step S1, and respectively forming a thunder and lightning mode product data sample set, a thunder and lightning activity data sample set and a meteorological observation data sample set, S3, performing sample training to form an artificial intelligence model, and S4, obtaining thunder and lightning mode product data in a to-be-predicted time period and meteorological observation data before a forecasting starting moment, inputting the data into the artificial intelligence model in the step S3, and obtaining thunder and lightning activity information in the to-be-predicted time period. The invention further provides a thunder and lightning early warning and forecasting system for implementing the method. By adopting the method and the system provided by the invention, the inconvenience of split operation of lightning approaching early warning and short-time forecasting caused by adopting different methods and technical routes in the prior art can be overcome, and lightning approaching early warning and short-time forecasting are integrally realized.

Description

technical field [0001] The present invention relates to the field of weather forecasting, in particular to a lightning warning and forecasting method and system. Background technique [0002] Because of its high voltage, high current, strong electromagnetic radiation and other characteristics, lightning activities are usually very destructive, causing heavy casualties and property losses. In order to reduce the losses caused by such lightning disasters, an important means is to carry out short-term forecast (2-12 hours) and near-warning (0-2 hours) of lightning activities. [0003] The existing short-term forecast of lightning activities is basically based on numerical forecasting models, mainly including two methods. The first method can be called a diagnostic method, which uses model parameters closely related to lightning activities to forecast lightning activities. For example, parametric formulas for lightning activity and vertical airflow or ice-phase content in the c...

AI Technical Summary

Problems solved by technology

Limited by the current level of theoretical research, there are still deficiencies in the application of these methods. For example, for lightning activity forecasting, the short-term lightning forecast based on numerical models has an essential dependence on the accuracy of the model itself, especially extreme Rely on the accuracy of cloud simulation, and in the numerical model of normal operation, there is still a lot of room for improvement in the accuracy of cloud simulation; for lightning impending warning, or a single data is not enough to provide enough information to indicate lightning activities, Or the multi-data fusion method is too simple, and it is difficult to mine the deep-level relationship between different data and lightning activities

Ultimately, the current lightning early warning and forecasting has problems such as low short-term forecast accuracy and small improvement space for impending warning.

At the same time, due to different methods, the current lightning nowcasting and short-term forecasting are usually two independent systems, which cannot meet the requirements of meteorological seamless forecasting

Images