Sound quality prediction for engine-radiated noise. (May 2015)

• Record Type: Journal Article
• Authors: Liu, Hai; Zhang, Junhong; Guo, Peng; Bi, Fengrong; Yu, Hanzhengnan; Ni, Guangjian
• Is Part Of: Mechanical systems and signal processing. Volume 56/57(2015)
• Page Start: 277
• Page End: 287
• Publication Date: 2015-05
• Subjects: Sound quality -- Subjective assessment -- Psychoacoustic -- GA-SVM -- Engine-radiated noise; Structural dynamics -- Periodicals; Vibration -- Periodicals; Constructions -- Dynamique -- Périodiques; Vibration -- Périodiques; Structural dynamics; Vibration; Periodicals; 621
• DOI: 10.1016/j.ymssp.2014.10.005
• Languages: English
• ISSNs: 0888-3270

Abstract

Abstract: Diesel engine-radiated noise quality prediction is an important topic because engine noise has a significant impact on the overall vehicle noise. Sound quality prediction is based on subjective and objective evaluation of engine noise. The integrated satisfaction index (ISI) is proposed as a criterion for differentiate noise quality in the subjective evaluation, and five psychoacoustic parameters are selected for characterizing and analyzing the noise quality of the diesel engine objectively. The combination of support vector machines (SVM) and genetic algorithm (GA) is proposed in order to establish a model for predicting the diesel engine-radiated noise quality for all operation conditions. The performance of the GA-SVM model is compared with the BP neural network model, and the results show that the mean relative error of the GA-SVM model is smaller than the BP neural network model. The importance rank of the sound quality metrics to the ISI is indicated by the non-parametric correlation analysis. This study suggests that the GA-SVM model is very useful for accurately predicting the diesel engine-radiated noise quality. Highlights: Psychoacoustic evaluation of engine-radiated noise varying with speeds and loads. The Genetic Algorithm-Support Vector Machine (GA-SVM) algorithm is applied to predict the engine-radiated noise quality. Correlations analysis is used to rank the contributions of sound quality metrics.