DEVELOPMENT OF A RECURRENT SIGMA-PI NEURAL-NETWORK RAINFALL FORECASTING SYSTEM IN HONG-KONG

At the moment, weather forecasting is still an art the experience and intuition of forecasters play a significant role in determining the qu ality of forecasting. This paper describes the development of a new ap proach to rainfall forecasting using neural networks. It deals with th e extraction of information from radar images and an evaluation of pas t rain gauge records to provide short-term rainfall forecasting. All o f the meteorological data were provided by the Royal Observatory of Ho ng Kong (ROHK). Pre-processing procedures were essential for this neur al network rainfall forecasting. The forecast of the rainfall was perf ormed every half an hour so that a storm warning signal can be deliver ed to the public in advance. The network architecture is based on a re current Sigma-Pi network. The results are very promising, and this neu ral-based rainfall forecasting system is capable of providing a rain s torm warning signal to the Hong Kong public one hour ahead.