Search Results

The investigation of critical noise sources on pass-by noise tests is demanding development of the current techniques in order to locate and quantify these sources. One recent approach is to use beamforming techniques to this purpose. The phased array information can be processed using several methods, for example, conventional delay-and-sum algorithms, deconvolution based algorithms, such as DAMAS, and more recently, the generalized inverse beamforming. This later method, presents the advantage of separating coherent sources with better dynamic range than conventional beamforming. Also, recent developments, such as Iteratively Re-Weigthing Least Squares, increases the localization accuracy allowing it to be used in a challenging problem as a fast moving source detection, a non-stationary condition. The work will raise the main advantages and disadvantages on this method using a practical case, a passenger vehicle pass-by test.

Prediction of the drive-by noise level in the early design stage of an automotive vehicle is feasible if the source signatures and source-receiver transfer functions may be determined from simulations based on the available CAD/CAE models. This paper reports on the performance of a drive-by noise synthesis procedure in which the transfer functions are numerically evaluated by employing the Fast Multipole Boundary Element Method (FMBEM). The proposed synthesis procedure first computes the steady-state receiver contributions of the sources as appearing from a number of vehicle positions along the drive path. In a second step, these contributions are then combined into a single transient signal from a moving vehicle for each source-receiver pair by means of a travel time correction.

While CAE methods allow improving nominal product design using virtual prototypes, uncertainty and variability in properties and manufacturing processes lead to scatter in actual performances. Uncertainty must hence be incorporated in the CAE process to guarantee the robustness and reliability of the design. This paper presents an overview of uncertainty-based design in automotive and aerospace engineering. Fuzzy methods take uncertainty into account, whereas reliability analysis and a reliability-based design optimization framework can deal with variability. Key enabling technologies to alleviate the computational burden, such as workflow automation, substructuring and design of experiments, are discussed, and industrial applications are presented.