基于MEEMD的内燃机机体活塞敲击激励与燃烧爆发激励分离研究

提出了一种改进的集总平均经验模态分解（MEEMD）方法，并阐述了其实现过程。通过仿真实验，验证了MEEMD是一种更为优秀的自适应信号分解方法，不仅抑制了经验模态分解（EMD）的模态混叠问题，而且消除了集总平均经验模态分解（EEMD）中的，非标准IMF分量、模态分裂和白噪声残余等问题。将MEEMD应用到内燃机振声信号处理中，以某柴油机为研究对象，对其机体上的振动信号进行MEEMD分解，分离其中的活塞敲击引起的机械激励成分与燃烧爆发激励成份，诊断主要振动源。

Separation of piston-slap and combustion shock excitations via MEEMD method

A modified ensemble empirical mode decomposition(MEEMD) was proposed here,and its basic principles were discussed.Numerical simulation showed that MEEMD is a better adaptive signal decomposed method,not only suppresses the mode mixing problem of EMD,but also overcomes non-IMFs,mode splitting and noise residue problems of EEMD.MEEMD was then used to analyze vibration signals of a diesel engine.The analysis results showed that the mechanical excitation component induced by piston-slap can be successfully separated from the combustion shock excitation component,and the main vibration sources can be obtained.