多声源共同作用下的混合声剂量值预测方法研究

本文借助单一声样本与人工合成的混合声样本,重点研究了由多个单一声源共同作用形成的混合声的剂量值预测方法.首先,提出了一种基于对作用时长短时化处理的声音样本剂量值确定流程,并利用该方法分别确定了单一声样本与人工合成的混合声样本的剂量值.随后,分析了混合声样本剂量值(亦称总剂量,记为LTotal)与构成混合声样本的每个单一声样本剂量值(亦称组分剂量,记为Li,i=1,2,···,K;K为单一噪声样本的个数亦称组分数量)之间的关系,实现了在已知所有单一声剂量值的前提下成功预测混合声的剂量值,为深入开展复杂声环境下的噪声源控制和噪声总剂量控制、实现高效的环境噪声治理提供了理论依据.

Research on prediction and methods of evaluating sound exposure from a mixture of multiple single sources

In this paper, the prediction and the way for calculating the total exposure level (to be denoted by L_Total) from a mixture of multiple single sources are proposed, on the premise that each single exposure level from every component single source (to be denoted by L_i, where i denotes the number of single sources and i=1, 2, …, K) is known. Firstly, a novel method for sound exposure level evaluation, based on a short-term exposure level in the duration of the sound event, is proposed. Using this method, each single exposure level obtained from all the single sound samples and the total exposure level obtained from every artificially combined sound samples are evaluated. Then, we lay special stress on analyzing the quantitative relationships between L_Total and L_i (i=1, 2, …, K) measured in three types of sound exposure indicators. All analytical results indicate that our anticipative gain of the total exposure level L_Total from a mixture of multiple single sources can be predicted on the premise that each single exposure level from every componential single source is known. To modify the predicted results, we just need to know the number of specific independent components K and the range of the single exposure levels ρ.
Keywords： sound exposure predicting model single source combined source