Application Research of Spectral Decomposition-Acoustic Imaging in Engine Noise Source Identification

To effectively enhance the noise source identification performance of critical automotive components, spectral decomposition-beamforming technology based on array measurement theory is applied to measure and identify automotive engine noise. Based on conventional beamforming and matrix function theory, the paper provides a theoretical derivation for function beamforming within typical spectral decomposition methods. Computational analysis was conducted using both the proposed method and conventional approaches for numerical simulations of single, dual, and quadruple sound sources in two-dimensional and three-dimensional scenarios, as well as for actual engine tests across different rotational speeds and narrowband/wideband computational frequency bands. Results demonstrate that spectral decomposition methods exhibit excellent main-beam focusing and side-lobe suppression performance. Under identical test and computational conditions, proposed methods significantly enhance spatial resolution.