Flow prediction of the Chira river using artificial neural networks

Abstract

This article proposes an artificial neural network model for predicting the flow of the Chira River, specifically using the LSTM (Long Short-Term Memory) model, designed for the processing of data sequences such as time series. This model is particularly well suited for this task due to its ability to capture long-term dependencies in the data, which is essential when working with variable hydrological information. Based on data recorded and obtained from the national open data platform, we seek to develop a system that allows estimating the flow level of the Chira River with high precision. This system will be very useful not only for residents near the river, who will be able to receive early warnings about possible floods and thus take the necessary measures to protect themselves, but also for engineers specialized in the hydrological branch. Engineers will be able to use these predictions to better plan and manage water resources, conduct environmental impact studies, and design more efficient and safe infrastructure. Furthermore, the implementation of this model will contribute to scientific research in the field of hydrology and environmental sciences, providing an advanced tool for the analysis of water flow patterns and behavior of the Chira River under various climatic and seasonal conditions.