Browse Publications Technical Papers 2013-01-1005

Efficient and Accurate Evaluation and Improvement of Engine Radiated Noise 2013-01-1005

In engine radiated noise analysis, the conventional boundary element method (BEM) is unsuitable for conducting large-scale acoustic simulations because of low calculation efficiency. The Fast Multipole boundary element method (FMBEM), although greatly improved calculation efficiency for large-scale acoustic simulations, has the problem of low efficiency at the low frequency region and low accuracy at the high frequency region. Those methods cannot meet the requirements of design engineers to get the results in a couple of days. In order to solve the powertrain radiated noise problems accurately and efficiently, some simplified methods and scripts were developed in big-three auto makers and suppliers, such as DIRA. Those scripts are very fast and can meet the timing requirement, but generally, without clear physical meaning.
A simplified method and a Matlab program was developed in this paper based on surface vibration velocity to calculate the radiated sound power, which can be compared with other conventional methods and commercial software packages. The complete analysis procedure applied for powertrain radiated noise analysis is elaborated and presented in this paper. Several numerical examples including engine block and engine powertrain system were used to compare accuracy and efficiency among BEM, FMBEM, and the surface vibration velocity method. Results show that for the engine block and engine powertrain analysis problems, the Matlab program developed in this paper could reduce the calculation time to 1/133 or 1/104 comparing to FMBEM, and 1/68 or 1/124 comparing to BEM, while keeping the high calculation accuracy.


Subscribers can view annotate, and download all of SAE's content. Learn More »


Members save up to 43% off list price.
Login to see discount.
Special Offer: With TechSelect, you decide what SAE Technical Papers you need, when you need them, and how much you want to pay.