Análise numérico-analítica para o dimensionamento de um conjunto de discos atenuadores em dutos visando o controle de ruído industrial

Abstract

Many researches, through articles, doctoral theses and master's dissertations address different models of perforated sheet, however, there is no content in the literature that helps engineers such as its correct application for a set of attenuating discs in a case of study. Thus, the proposed work suggests an optimal methodology for applying a set of perforated plate inside pipelines. The analysis developed in this research presents a unique numerical-analytical methodology, for the application of multiple perforated plates aiming at the attenuation of industrial noise in duct flows. Through the application of the Transfer Matrix Method (TMM), combined with the Finite Element Method (FEM) and the Finite Volume Method (FVM), a procedure is presented for dimensioning multiple perforated plates in duct flows, in an optimal approach. The methodology consists of applying the Finite Element Method to identify, first, the best location for installing the perforated discs in a given flow to be treated acoustically. Subsequently, it follows the computational fluid dynamics (CFD) simulation to obtain the flow velocity inside the holes of a certain perforated plate mesh and, finally, the application of the Transfer Matrix Method for calculating the acoustic performance of the disc through Transmission Loss (TL) parameter. All the analytical models analyzed showed the need to take into account the flow speed during drilling. With that, the CFD proved to be a useful tool in the calculation of this local speed, this step was validated using experimental data from a FAPEMIG survey carried out at the UFMG School of Engineering, together with ISOBRASIL. Through the methodology performed, it was possible to insert the perforated discs aiming at maximum attenuation, transmission loss calculation, as well as obtaining the fluidic parameters through the CFD simulation and validating the transfer matrix method and computational simulations performed through of experimental data.