AUTOMOTIVE EXTERIOR NOISE OPTIMIZATION USING GREY RELATIONAL ANALYSIS COUPLED WITH PRINCIPAL COMPONENT ANALYSIS
Shuming Chen,Dengfeng Wang,Bo Liu
DOI: https://doi.org/10.1142/s021947751350017x
2013-01-01
Fluctuation and Noise Letters
Abstract:This paper investigates optimization design of the thickness of the sound package performed on a passenger automobile. The major characteristics indexes for performance selected to evaluate the processes are the SPL of the exterior noise and the weight of the sound package, and the corresponding parameters of the sound package are the thickness of the glass wool with aluminum foil for the first layer, the thickness of the glass fiber for the second layer, and the thickness of the PE foam for the third layer. In this paper, the process is fundamentally with multiple performances, thus, the grey relational analysis that utilizes grey relational grade as performance index is especially employed to determine the optimal combination of the thickness of the different layers for the designed sound package. Additionally, in order to evaluate the weighting values corresponding to various performance characteristics, the principal component analysis is used to show their relative importance properly and objectively. The results of the confirmation experiments uncover that grey relational analysis coupled with principal analysis methods can successfully be applied to find the optimal combination of the thickness for each layer of the sound package material. Therefore, the presented method can be an effective tool to improve the vehicle exterior noise and lower the weight of the sound package. In addition, it will also be helpful for other applications in the automotive industry, such as the First Automobile Works in China, Changan Automobile in China, etc.