Suppression and analysis on annoyance of motor vehicle noise using water sound injection

流水声注入对机动车噪声烦恼感抑制的影响因素

¹ School of Marine Science and Technology, Northwestern Polytechnical University, Xi’an 710072, China
² State Key Laboratory of Power Grid Environmental Protection, Wuhan 430000, China

Received: 25 September 2021

Abstract

Different from the traditional noise control approaches based on sound energy reduction, this study focuses on the annoyance suppression of motor vehicle noise based on audio injection. Firstly, eight motor vehicle noises were selected as the target sounds and three different controllable sounds were used to overlap with different signal-to-noise ratios. Then absolute threshold and combined sound annoyance subjective evaluation experiment were carried out to obtain absolute threshold range and the existence zone of "destructive effect" for combined noise annoyance. Next the influencing factors of the destructive effect are analyzed. The results of the study found that the identifiability of the target sound and the controllable sound is important to the selection of the controllable sound. For most target sounds with high identifiability, the controllable sound with high identifiability and high SNR should be selected. For most target sounds with low identifiability, the controllable sound with low identifiability and low SNR should be selected. Moreover, after the addition of optimal controllable sound, the descend value of the annoyance degree of the target sound with high identifiability is lower than that of the target sound with low identifiability. The results will also provide guidance for the research on audio injection effects, physical mechanisms, the controllable sound selection and optimization design.

摘要

与传统的基于声能量降低的噪声控制技术不同, 研究的是基于音频注入法的机动车噪声烦恼感抑制问题。以8种机动车噪声为目标声, 分别设计3种不同调控声以不同信噪比与之叠加, 开展绝对阈限和叠加声烦恼感主观评价实验, 得到了绝对阈限范围和叠加声烦恼感的"相消效应"存在条件。对相消效应的影响因素进行了分析。研究结果表明目标声和调控声的可识别性是选择调控声的重要依据。对大多数可识别性高的目标声, 应选择可识别性高、信噪比大的调控声; 对大多数可识别性低的目标声, 应选择可识别性低、信噪比小的调控声。加入最优调控声后, 可识别性高的目标声烦恼度下降值低于可识别性低的目标声。研究将为音频注入效应、物理机制及调控声选择与优化设计提供参考。