Determinação das ações atuantes nos silos através de redes neurais artificiais

Abstract

Storage of bulk products is performed in buildings called silos, which are typically used in industry, agriculture, steel and ports, among others. These silos can be built with several types of materials such as concrete, steel mill, wood, mortar, etc.. Most of the silos in the world does not present ideal conditions of operation due to poor knowledge of the pressures, flow and also due to the large number of variables that affect the behavior of the products stored in silos. This explains the large number of accidents and collapse. Objective with this research was to determine and evaluate the methods of calculation of the pressures acting in silos and the flow of product through an artificial intelligence technique known as Artificial Neural Networks. This is a powerful computational tool used in different areas of science and is inspired by the neural structure of intelligent organisms, acquiring knowledge through experience. Using this tool provides alternative methods for analysis of these actions, providing reliable results in accordance with key international regulatory codes.