Özet:

For diesel engine misfire fault diagnosis under strong noise, a new method based on variational mode decomposition (VMD) and cross wavelet transform (XWT) is proposed. Firstly, the vibration signal of cylinder head is processed by stages according to the working cycle of diesel engine, and then the vibration signal of each working cycle is re-sampled according to the same angle. Then the vibration signal of cylinder head is decomposed by VMD to denoise adaptively and reconstruct the signal. Next, XWT is used to analyze the time-frequency correlation of any two continuous working cycle signals, and the non-periodic components and random noise in the vibration signals are further eliminated to extract the combustion characteristics of diesel engines. Finally, the misfire fault of diesel engine is diagnosed by calculating the energy proportion of each cylinder in time-frequency space. The effectiveness of the proposed method is verified by simulation and experiment.

Anahtar Kelimeler: