基于径向基函数模型的某混合动力车型怠速声品质优化  

Optimization of HEV Vehicle’s Idle Sound Quality Based on RBF Prediction Model

作　　者：孔丹丹 郑海生 杜浩 袁懋荣 Kong Dandan;Zheng Haisheng;Du Hao;Yuan Maorong(Automotive Research&Development Center of Guangzhou Automobile Group Co.,Ltd.,Guangzhou 511434)

机构地区：[1]广州汽车集团股份有限公司汽车工程研究院,广州511434

出　　处：《汽车技术》2022年第11期47-53,共7页Automobile Technology

摘　　要：以搭载某自然吸气四缸汽油发动机的混合动力车型为研究对象,选取心理声学客观参数响度、粗糙度、尖锐度进行客观评价研究,并通过径向基函数(RBF)神经网络方法建立该车型怠速声品质预测模型。结果显示:采用RBF方法得到结果的一致性较高,预测精度较高。建立基于RBF的噪声品质客观评价参量的灵敏度模型,计算影响该车型怠速声品质的客观评价参量贡献度,其中粗糙度特征的贡献度影响最大,尖锐度次之。通过分析曲轴系统和正时系统的动态特性,提出优化措施,根据RBF回归模型预测得到的主观评价结果,较优化前的主观评价结果提升显著。With a HEV powered with 4-cylinder normally aspirated gasoline engine as the research object, this paper selected the objective parameters of psychoacoustics such as loudness, roughness, sharpness, for objective evaluation, and Radial Basis Function(RBF) neural network method was used to establish the predictive model of idling sound quality of this HEV. The results show that the results obtained by RBF method have high consistency and high prediction accuracy. The sensitivity model of objective evaluation parameters of sound quality based on RBF was established to calculate quantitatively the contribution of objective evaluation parameters that affect the idling sound quality of this vehicle. It is concluded that contribution of roughness to the sound quality is the largest, followed by sharpness.Through the dynamic characteristics analysis of the crankshaft system and timing system, this paper proposed the optimization measures, and obtained the subjective evaluation result according to the RBF regression model prediction,which indicated that the subjective evaluation results improve remarkably compared with the idling sound quality without optimization.

关 键 词：噪声品质 径向基函数 贡献度分析 曲轴系统 曲轴扭转减振器 正时系统 

分 类 号：U491.91[交通运输工程—交通运输规划与管理]