摘要
以燃料电池轿车(fuel cell vehicle,FCV)为研究对象,采集其怠速工况不同位置的声音信号作为试验样本,采用成对比较法对其中的24个样本信号进行了主观评价试验,同时计算了可以描述其声音特性的6个客观评价参量,并引入BP神经网络建立了FCV声品质预测模型。通过所建立的模型计算FCV声品质客观评价参量对主观评价结果的影响权重,首次提出使用BP神经网络的方法来确定声品质评价中客观评价参量对主观评价结果的影响权重,研究结果表明,FCV声品质主要受响度、料糙度和A声级三个客观参量的影响。此次分析,不仪适用于燃料电池轿车对其它领域的声品质分析与评价都起到了指导性的意义。
Here, a fuel cell vehicle （fuel cell vehicle, FCV） was focused on as the study object and the sound signals collected in its different locations at an idle condition were as the experimental samples. The paired comparison method was used to perform subjective evaluation tests for 24 signal samples stated above, meanwhile, the six parameters of objective evaluation were calculated to deserib the sound characteristics and a BP neural network was adopted to establish a FCV sound quality prediction model, it could be used to calculate the fluence weight of objective evaluation of sound quality parameters on the results of subjective evaluation. The method of BP neural network was proposed the first time to determine the impact weight measuring objective evaluation parameters contributing to the results of subjective evaluation in the process of sound quality evaluation, then the main objective parameters determining sound quality good or not were obtained. This analysis played a significant guiding role in both fuel-cell vehicle and other areas for sound quality evaluation and analysis.
出处
《振动与冲击》
EI
CSCD
北大核心
2011年第1期91-94,120,共5页
Journal of Vibration and Shock
基金
国家自然科学基金(51075302)
关键词
燃料电池轿车
声品质
BP神经网络
权重
fuel cell vehicle
sound quality
BP neural network
predict model
weight
