Rain Prediction Model Using Artificial Neural Networks

Abstract

Precipitation is crucial for environmental sustainability and life. Accurate forecasts can mitigate floods, environmental disasters, and agricultural losses. This study presents an hourly rain prediction model using Artificial Neural Networks (ANNs) based on meteorological data from the National Institute of Meteorology (INMET) in Manaus/AM. Various network architectures were tested to determine the optimal configuration. The results demonstrated that the model effectively reproduced precipitation behavior, particularly on rainy days. However, for convective events, the model failed to capture intensity accurately, suggesting the need for additional atmospheric variables.