基于EEMD与广义S变换的内燃机噪声源识别研究

Application of EEMD and GST Method in Noise Characteristics Analysis of Internal Combustion Engine

介绍了一种新颖的自适应信号处理方法——集总平均经验模态分解(EEMD)的基本原理,通过仿真试验,验证了EEMD在一定程度上克服了传统经验模态分解(EMD)在模态混叠问题上的缺陷,能够对复杂的非稳态信号进行较为准确的自适应模态分解。以某4缸四冲程柴油机为研究对象,对其气缸顶部的噪声信号进行EEMD分解,并采用广义S变换(GST)对其中各个本征模态函数(IMF)分量进行时频分析。研究结果表明:分解得到的各个IMF分量均具有独立和真实的物理意义,能够准确地反映内燃机各个噪声源的分布及其大小。

A novel adaptive signal decomposition method——ensemble empirical mode decomposition（EEMD） was introduced.Simulation experiment demonstrates that EEMD method overcomes to a certain extent the mode mixing problem of traditional EMD method.With the EEMD method,the components with truly physical meanings can be adaptively extracted from complicated nonstationary signals.Taking a four cylinder four stroke diesel engine as the object,the EEMD method was used to analyze its noise signals at the top of the cylinder head cover,and the time-frequency representations of each intrinsic mode function（IMF） component were performed by using generalized S transform（GST） method.The results indicate that each IMF component has truly and independent physical meaning.The dominant noise sources of the engine can be identified correctly,and their amplitudes obtained too.