摘要
通过对内燃机的辐射噪声源进行分离和识别,得到的各独立噪声可为其减振降噪和监测诊断提供依据。在内燃机噪声源识别中,燃烧噪声和活塞敲击噪声在时域和频域上均有混叠,很难准确地将其进行分离。在半消音室中进行内燃机振动噪声试验,采集一路内燃机油底壳近场辐射噪声,先对其进行消除趋势项及滑动平均等预处理,减少随机误差成分,然后用集合经验模态分解得到IMF分量,用主分量分析降维,最后用快速独立分量分析进行盲源分离,并结合连续小波时频分析等方法进行识别。研究结果表明:通过采用单通道算法对内燃机油底壳辐射噪声进行分离和识别,可得到各独立分量,分别为内燃机的燃烧噪声和活塞敲击噪声。
The noise source of an internal combustion engine is separated and identified. The independent noisesobtained are used to provide the basis for the noise reduction, monitoring and diagnosis. In the noise source identification ofthe internal combustion engine, the combustion noise and the piston knock noise are found to be overlapped in time domainand frequency domains, so that it is very difficult to accurately separate them. Therefore, the internal combustion enginevibration noise is tested in a semi-anechoic room, and the near-field radiation noise of the oil pan of the internal combustionengine is measured. The measurement data is processed by eliminating the trend item and smoothing the average so as toreduce the random error components. Then, the ensemble empirical mode decomposition is used to get the IMF components,and the principal component analysis is adopted to reduce the dimension. Finally, the fast independent component analysis isused for blind source separation, and the continuous wavelet time-frequency analysis method is adopted for identification. Itis concluded that by using the single channel algorithm to separate and identify the radiation noise of the oil pan of theinternal combustion engine, the independent components obtained are respectively the combustion noise and the pistonknock noise of the internal combustion engine.
出处
《噪声与振动控制》
CSCD
2016年第3期132-136,共5页
Noise and Vibration Control
基金
国家自然科学基金资助项目(51279148)
关键词
声学
内燃机
油底壳
噪声分离
集合经验模态分解
盲源分离
acoustics
internal combustion engine
oil pan
noise separation
ensemble empirical mode decomposition
blind source separation
