A computational approach for optimizing vehicles' interior noise and vibration

Abstract

This paper proposes a Genetic Algorithm (GA) to optimise vehicles' interior noise and vibration caused by powertrain, tire-road surface interaction and type of car. Toward this end, an experimental design was carried out to obtain the noise and vibration data of three local compact-sized cars at stationary and running conditions and varying engine speeds. The acquired data were analysed to obtain sound quality parameters such as loudness and sharpness, sound pressure level and vibration exposures in the interior cabin. Besides that, a K-means clustering algorithm was utilised to cluster the noise and vibration to determine the comfort level in the vehicle's interior cabin. The overall findings indicate that the comfort level is influenced by the types of road surface, powertrain and vehicle design. The results also indicate that the proposed GA approach is reliable and can be utilised by automotive researchers to identify the optimal Noise, Vibration and Harshness (NVH) values for vehicle refinement and noise control.