NASCIMENTO, Manoel Henrique Reis2024-08-132024-08-132022-07-05LOPES, Gisele de Freitas. Aplicação do modelo Autoregressivo não linear com entradas exógenas para previsão do nível do rio no Amazonas. 2022. 69 Folhas. Dissertação do programa de pós-graduação em Engenharia, Gestão de Processos, Sistemas e Ambiental (EGPSA), Instituto de Tecnologia e Educação Galileo da Amazônia (ITEGAM), Manaus, 2022.https://rigalileo.itegam.org.br/handle/123456789/55River level variation is a constant problem, the prediction of river level variation brings a possibility of planning in all areas of action, which provides for a reduction in the impact caused by floods and ebbs. The present work is justified by three basic lines that involve the problem of the theme, which are the use of Artificial Intelligence, the problem of floods in the Amazon and the issue of technology in favor of decision making. The environmental impacts caused by economic and social factors are problems portrayed in scenarios such as floods and ebbs of rivers, bringing up situations such as an increase in diseases, reduction of agricultural production in places that depend on precise geological control, in addition to the increase in erosive processes in risky locations. Thus, the use of AI to predict the river level, which consequently can minimize the problems arising from floods that cause environmental impact, is highly possible, because when it is known in advance that an event is about to happen, decisions can be made that impacts are smaller. This work models and applies NARX to predict the river level in the Amazon with variables that are easy to access and implement through the MATLAB software, in order to contribute with a forecast model capable of predicting a possible flood from the river level. For the application of the methodology, the input variables of the National Institute of Meteorology were used, and the output variable of the Port of Manaus website, Rio Negro station of the port of Manaus, 01/31/2020 to 06/30 /2021. The performance of the models was compared with 5, 10, 15, 20 and 25 steps forward, considering months as the forecast horizon. The NARX model obtained a better response in the predictions, among which the 15-step horizon stood out.pdf.Previsão hidrológicaEnchentes na AmazôniaModelo NARXMATLAB para ModelagemDissertação3.01.04.02-5 Hidrologia