A Practical Procedure to Predict AIS Inlet Noise Using CAE Simulation Tools Chrysler Group LLC

Format:

Conference/Event

Author/Creator:

Zhang, Zhang, author.

Contributor:

Butler, Brian

Likich, Mark

Lynch, Mac

Conference Name:

SAE 2013 World Congress & Exhibition (2013-04-16 : Detroit, Michigan, United States)

Language:

English

Physical Description:

1 online resource

Place of Publication:

Warrendale, PA SAE International 2013

Summary:

The air induction system (AIS), which provides clean air to the engine for combustion, is very important for engine acoustics. A practical CAE procedure to predict AIS inlet noise is presented in this paper. GT-Power, a commercially available software program can be used to simulate the engine performance and predict air induction noise. The accuracy of GT-Power is dependent on many variables, such as: proper duct discretization size, proper number of flow splits to model the air box and the capturing of the correct resonator geometry for tuning frequency. Since GT-Power is based on a 1D assumption, several iterations need be performed to model the complex AIS components, such as, irregular shaped air box, resonator volume, porous ducts and perforated pipes. Because of this, the GT-Power AIS model needs to be correlated to test data using transmission loss data. But in the case when test data is not available for correlation, especially in early design stage when no test parts are available, a virtual test' method needs to be utilized. As an alternative, a 3D Finite Element Method (FEM) or Boundary Element Method (BEM), can be utilized as a virtual test' TL bench for GT-Power model correlation. In this paper, SYSNOISE, which has both a FEM and BEM solver, is utilized to compute the TL of an Air Induction System. The procedure is described as follows. First, a preliminary AIS GT-Power model is built based on the design geometry. Second, the corresponding SYSNOISE model for the same geometry is built and the TL is computed. Then, iterations are taken to update the GT-Power model in order to correlate the TL data with SYSNOISE TL results. Finally, the correlated GT-Power AIS model is attached to the full system GT-Power model and then the software computations are run yielding a predictive inlet noise curve. Several case studies are shown that demonstrate the validity of the proposed procedure

Notes:

Vendor supplied data

Publisher Number:

2013-01-1004

Access Restriction:

Restricted for use by site license