Prediction of energetic solar particle event dose-time profiles using artificial neural networks

A set of artificial neural networks has been developed which is capable of forecasting the three unknown parameters used to describe dose-time profiles of energetic solar particle events based on the doses obtained during the early stages of the event. Example forecasts are given for three events: a "worst-case" scenario (4/spl times/ August 1972), August 1972 (a two-rise event) and October 1989 (a four-rise event).