FLOOD FORECASTING USING ARTIFICIAL NEURAL NETWORK (ANN) FOR GODAVARI REGION

Abstract

Prolonged heavy seasonal rains cause the rivers to overflow and inundate adjoining areas. So, to reduce the flood risk and to ensure appropriate measures to control flood risk, to manage water resource systems, and mitigate flood hazard during flood its forecasting is very important. Among the various parameters used, Rainfall and present river water level data are most used due to its standardized form and variable timescale. Nowadays flood forecasting using a machine learning model has the upper hand over traditional methods due to its adaptability with nonlinear characters. In this study, various machine learning models are used to predict the river water level (RWL) at different timescales like RWL-3, RWL -6, RWL -9, and RWL -12 of ‘Godavari’, the flood-prone region in the Andhra Pradesh State of India, for the period of 1960 to 2020. Along with Rainfall and present river water level, some other parameters like catchment characteristics, elevation, expansion and contraction of the river, roughness, etc. are also given as an input which shows a significant increase in accuracy. It produced satisfactory results, and it could be used as a rapid tool for decision making.