Typhoon path random simulation method and system based on geographically weighted regression

Abstract

The invention discloses a typhoon path random simulation method based on geographically weighted regression. The method comprises: selecting an area needing to be simulated; inputting actual measurement data of a historical path; dividing the selected area into grids; obtaining a coefficient of each grid point parameter model by adopting geographically weighted regression; performing spatial interpolation on the obtained coefficient result according to simulation requirements to obtain a spatial distribution surface of each coefficient; and performing random simulation of typhoon in combination with the space distribution surface of each coefficient. According to the simulation method, a geographically weighted regression technology is introduced, and a typhoon path simulation parameter model continuously changing in space is obtained and mainly comprises basic parameters such as the typhoon moving direction, the moving speed, the maximum wind speed radius and the radial air pressure profile index.

Description

technical field [0001] The invention belongs to the technical field of typhoon path simulation, and in particular relates to a typhoon path stochastic simulation method and system based on geographic weighted regression. Background technique [0002] Typhoon (tropical cyclone or hurricane) is a rapidly rotating central low-pressure storm system. When it lands, strong winds, heavy rain, storm surges and even accompanying tornadoes will cause catastrophic damage to coastal cities, especially in the southeastern coastal areas of my country. On average, 7-8 typhoons make landfall, causing immeasurable economic losses. Carrying out typhoon simulations in coastal cities is of great significance for typhoon disaster prevention and mitigation, risk assessment, and guiding the design of wind-sensitive structures. [0003] Since a typhoon is a mesoscale storm system, it is a small probability event that it will directly hit a certain engineering site, but once it lands, it will cause h...

AI Technical Summary

Problems solved by technology

[0003] Since a typhoon is a mesoscale storm system, it is a small probability event that it will directly hit a project site, and once it lands, it will cause heavy losses

The traditional method of wind disaster assessment based on the actual measurement data of weather stations will produce large errors. On the one hand, the typhoon samples of historical observations are often insufficient, and the results obtained by the probability statistics method are not representative. If typhoons and good typhoons are considered at the same time , due to the inhomogeneity of the samples, the impact of strong typhoon data samples on the overall wind speed probability distribution is often amplified. On the other hand, most weather station observation equipment will be damaged during the strong typhoon landing process, and effective strong wind data cannot be captured

Images