Acoustic Optimization of a Muffler through the Sherpa Algorithm

Regulations on noise and gas emissions become more and more stringent resulting in noise abatement devices needing further engineering and optimization. Mufflers are installed at the end of powertrains to reduce the acoustic impact of the sound pressure from the engine. Such acoustic reduction is achieved through internal structures that promote destructive interference within the muffler. However, the muffler increases the backpressure downstream of the aftertreatment line, thus decreasing the engine efficiency. In the following work, an optimization workflow is presented to find the best design for a muffler geometry. The optimization is performed with the Sherpa algorithm that uses several optimization algorithms simultaneously to increase robustness and efficiency. Sherpa is implemented in an optimization tool that manages the workflow of two other software tools. The acoustic transmission loss of the muffler is calculated as a function of the geometrical characteristics of the muffler design. The parametrized muffler CAD is changed at each optimization iteration. For each CAD variation, the transmission loss and the backpressure are calculated. A Pareto front is generated and used to analyze the relation between the backpressure and the transmission loss. A first optimization cycle highlights three distinct regions of the Pareto front. From this first general analysis, a more detailed study is performed on a constrained portion of the Pareto front in order to find the best designs. The methodology allowed for an improvement of the backpressure of 4.3% from the baseline design and an increased transmission loss above all at the higher frequencies.