SOUND QUALITY ANALYSIS AND PREDICTION OF VEHICLE INTERIOR NOISE BASED ON GREY SYSTEM THEORY

This paper presents a novel effective methodology for sound quality analysis and prediction of vehicle interior noise using gray system theory. Four objective psychoacoustic parameters selected to evaluate acoustic performance of vehicle interior environment are loudness, sharpness, roughness, and fluctuation. Subjective evaluation is presented using paired comparison method, and Bradley–Terry model is also created. The relationship between subjective evaluation and psychoacoustic parameters is determined by using gray relational analysis. Meanwhile, the correlation among psychoacoustic parameters is also revealed. Sound quality prediction models are created based on GM (0, N) and GM (1, N) model. The prediction results show that GM (1, N) model is more capable than GM (0, N) model for prediction of sound quality. Thus, the analysis and prediction results confirm that the proposed method in this study can be a useful tool to analyze and predict sound quality of the vehicle interior noise.