噪声与振动控制2018，Vol.38Issue(3)：55-59,5.DOI:10.3969/j.issn.1006-1355.2018.03.010
基于BP神经网络的车内噪声时变综合烦躁度评价模型
Evaluation Model of Time-varying Comprehensive Irritability of Vehicle Interior Noise based on BP Neural Network
摘要
Abstract
:The sound quality of vehicle interior noise under non-steady condition is evaluated. Considering the time variability of the interior noise, the noise database under non-steady condition is established and the noise-related time-varying values of the psychoacoustic parameters are calculated. The subjective evaluation based on collected noise data is conducted. The"time-varying comprehensive irritability"is taken as the subjective evaluation index of the sound quality and the Semantic differential method is used as the subjective evaluation method of the sound quality. The evaluator can slide the slider in evaluation software according to his own subjective perception to the sound so as to obtain the continuous time-varying noise evaluation value. Finally, an objective evaluation model of BP neural network is established according to the time-varying value of noise and the real-time subjective evaluation value. Results of 4-fold cross-validating experiments verify that this objective evaluation model is effective in predicting vehicle interior non-steady noises.关键词
声学/非平稳车内噪声/时变综合烦躁度/心理声学参数/BP神经网络Key words
acoustics/vehicle interior non-steady noise/time-varying comprehensive irritability/psychoacoustic parameters/BP neural network分类
交通工程引用本文复制引用
陈卉,孙跃东,王岩松,周萍,冯天培..基于BP神经网络的车内噪声时变综合烦躁度评价模型[J].噪声与振动控制,2018,38(3):55-59,5.基金项目
国家自然科学基金资助项目(51675324) （51675324）
