Title: Noise source determination for vehicle engine compartment based on a corrected fast independent component analysis

Authors: Liang Hou; Enlai Zhang; Ya Guo; Jing Ye; Weiping Yang

Addresses: Department of Mechanical and Electrical Engineering, Xiamen University, No. 422, Siming South Road, Xiamen, China ' Department of Mechanical and Electrical Engineering, Xiamen University, No. 422, Siming South Road, Xiamen, China ' Department of Mechanical and Electrical Engineering, Xiamen University, No. 422, Siming South Road, Xiamen, China ' Department of Electronic Science, Xiamen University, No. 422, Siming South Road, Xiamen, China ' Department of Mechanical and Electrical Engineering, Xiamen University, No. 422, Siming South Road, Xiamen, China

Abstract: Accurately determining the unknown noise sources from a vehicle is critical to noise control; however, it is very difficult in the practical application and an efficient method for locating such noise sources is lacking. In this work, a method for noise sources determination is proposed based on a corrected fast independent component analysis (FastICA). In addition, to correct the uncertainty of FastICA estimation in order, phase and amplitude, spectrum maximum similarity (SMS) is presented. The simulation analysis and engineering application clearly demonstrate the proposed method is valid, which is successfully used to determine the main noise sources from the engine compartment of an excavator under three different working conditions.

Keywords: noise source determination; vehicle engine compartment; fast ICA; independent component analysis; FastICA; spectrum maximum similarity; SMS; vehicle noise; noise control; simulation.

DOI: 10.1504/IJVNV.2016.080136

International Journal of Vehicle Noise and Vibration, 2016 Vol.12 No.3, pp.199 - 216

Available online: 03 Nov 2016 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article