Tropical cyclone forecasting method based on depth target detection and numerical weather forecasting

Abstract

The invention discloses a tropical cyclone forecasting method based on depth target detection and numerical weather forecasting, which comprises the following steps of constructing a training data setaccording to a plurality of forecasting elements in a tropical cyclone historical record combined historical numerical weather forecasting product; according to the training data set, training a depth target detection model to obtain model parameters, and taking the model obtained through training as a tropical cyclone recognition model; and according to the tropical cyclone identification model,identifying the central position and the scale of the tropical cyclone from the real-time numerical weather forecast product forecast field, and outputting as the forecast results. According to the invention, the accurate prediction of the path and the scale of the tropical cyclone is realized, and the accurate prediction of the path and the scale of the tropical cyclone has an important effect on prevention and mitigation of related disasters.

Description

technical field [0001] The invention belongs to the technical field of meteorology, and in particular relates to a tropical cyclone forecasting method based on depth target detection and numerical weather forecasting. Background technique [0002] In the modern weather forecasting business, numerical weather prediction is becoming more and more important. Numerical weather prediction is based on the actual situation of the atmosphere, under certain initial value and boundary value conditions, through numerical calculation, to solve the fluid dynamics and thermodynamic equations, and make quantitative and objective forecasts of the future. Various analysis and forecast products obtained by numerical prediction methods are called numerical weather prediction products. With the development of numerical weather prediction, meteorological elements including temperature, pressure, humidity, and wind can obtain forecast conclusions for several hours to tens of hours in the future ...

AI Technical Summary

Problems solved by technology

[0003] In operational applications, meteorological forecasters generally use experience-based subjective forecasting methods and numerical model-based objective forecasting methods for tropical cyclone forecasting. With the development of numerical weather forecasting, tropical cyclone forecasting based on numerical models gradually reflects The greater the advantage, but the forecast of tropical cyclones is not the basic output element of medium and long-term numerical forecasting, but needs to use a specific tropical cyclone model based on the basic elements such as vorticity, pressure or wind field in the basic numerical model elements according to the relevant physical meaning Performing secondary calculations requires a lot of computing resources, and it is difficult for users with limited resources to perform local processing, and more candidate results will be screened out, which needs to be further judged based on live information

Images