基于EEMD-ICA-CWT的装载机室内噪声盲源分离和识别

发布时间：2018-06-13 13:40

本文选题：装载机司机位置处噪声 + 盲源分离　；参考：《天津大学学报(自然科学与工程技术版)》2015年09期

【摘要】：为了分离装载机的噪声源,采用集合经验模态分解(EEMD)、独立分量分析(ICA)和连续小波变换(CWT)技术相结合的方法,对装载机司机位置处噪声信号进行了盲源分离和声源识别研究.针对单一通道噪声信号进行盲源分离,将其分解成一系列独立分量.在削弱了传统经验模态分解(EMD)算法处理噪声信号时产生的模态混叠现象的同时,克服了独立分量分析方法要求传感器数目必须大于等于分离出分量数目的限制;借助连续小波变换良好的时频定位特性,对ICA分离结果进行时频分析.结合时频分析结果和各噪声源信号的频谱结构,确定了各独立分量与装载机不同噪声源的对应关系.结果表明,这些独立分量分别对应着装载机的燃烧噪声、冷却风扇辐射噪声及排气噪声等噪声源.
[Abstract]:In order to separate the noise source of the loader, blind source separation and sound source recognition are studied by means of the combination of empirical mode decomposition, independent component analysis (ICA) and continuous wavelet transform (CWT). Blind source separation is performed for single channel noise signal, which is decomposed into a series of independent components. The traditional empirical mode decomposition (EMD) algorithm not only weakens the modal aliasing phenomenon in the processing of noise signals, but also overcomes the limitation that the number of sensors must be greater than the number of separated components by independent component analysis (ICA) method. With the help of the good time-frequency localization characteristics of continuous wavelet transform, the separation results of ICA are analyzed. Combined with the results of time-frequency analysis and the spectrum structure of each noise source, the corresponding relationship between the independent components and the different noise sources of the loader is determined. The results show that these independent components correspond to the noise sources such as the combustion noise of the loader, the radiation noise of the cooling fan and the exhaust noise respectively.
【作者单位】：天津大学内燃机燃烧学国家重点实验室;天津大学机械工程学院;
【分类号】：TH243

【参考文献】