In this work, the acoustic and fluid-dynamic performances of a commercial three-chamber perforated muffler were simulated with a three-dimensional boundary element method and also a one-dimensional approach. The inner insulating material (wool) was taken into account in the performed analyses, together with the presence of a mean flow across the muffler in order to predict both the transmission loss and the pressure drop Dp. Three-dimensional analyses were experimentally validated in a wide frequency range and in the absence of mean flow and were utilized to build a more precise one-dimensional representation of the device. In this way, better agreement between the one-dimensional results and the experimental data was realized, at least in the frequency range characterized by planar wave propagation (below 800 Hz). Once validated, the one-dimensional model was coupled to an external optimizer to perform acoustic and fluid-dynamic optimizations of the considered muffler. Initially, a genetic algorithm was employed to modify the internal muffler geometry and to improve the transmission loss, in the absence of mean flow, in the 100-800 Hz frequency range. A second optimization was also performed to identify the trade-off between the acoustic performance and the fluid-dynamic performance, in terms of the transmission loss and Dp, in the 100-400 Hz frequency range.
Acoustic and fluid-dynamic optimization of an automotive muffler
Daniela SIANO;
2013
Abstract
In this work, the acoustic and fluid-dynamic performances of a commercial three-chamber perforated muffler were simulated with a three-dimensional boundary element method and also a one-dimensional approach. The inner insulating material (wool) was taken into account in the performed analyses, together with the presence of a mean flow across the muffler in order to predict both the transmission loss and the pressure drop Dp. Three-dimensional analyses were experimentally validated in a wide frequency range and in the absence of mean flow and were utilized to build a more precise one-dimensional representation of the device. In this way, better agreement between the one-dimensional results and the experimental data was realized, at least in the frequency range characterized by planar wave propagation (below 800 Hz). Once validated, the one-dimensional model was coupled to an external optimizer to perform acoustic and fluid-dynamic optimizations of the considered muffler. Initially, a genetic algorithm was employed to modify the internal muffler geometry and to improve the transmission loss, in the absence of mean flow, in the 100-800 Hz frequency range. A second optimization was also performed to identify the trade-off between the acoustic performance and the fluid-dynamic performance, in terms of the transmission loss and Dp, in the 100-400 Hz frequency range.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.