A Rain-Stimulated Flood Prediction For Rivers State Using Neural Networks.

Abstract

This project is to study the ability of neural networks to perform short-term prediction of the amount of rainfall of Rumudara River flooding for a specified area given previous rainfall data for a specified period of time. Short-term, in this case, means predicting the total rainfall for the next 24-hr period. The literature provides examples of neural networks being able to generate future averages of weather data if provided with data over a reasonably long range prior to the period being predicted. The dataset used in this study for training and consequently testing the Neural Network was sourced from Weather Underground official web site https://www.wunderground.com. An iterative Methodology was used and implemented in MATLAB. We adopted multi-layer Feedforward Neural Networks. Thus, for this project, the ability of a neural network to predict next day rainfall given a short range of precious days’ data is investigating.