基于小波包分解的整循环征兆提取与故障识别

Whole cycle symptom abstract and fault diagnosis based on the wavelet packet decomposition

为了检测内燃机气阀漏气的气密性故障,利用小波包分解改进算法,通过对柴油机完整工作循环内的缸盖振动信号进行小波包分解,从小波包分解系数中提取柴油机振动诊断的整循环征兆。由整循环特征向量图表明,正常状态时柴油机气缸盖振动信号中低频部分能量相对较大,高频部分能量相对较小;漏气状况时振动信号中的低频部分能量减小,而高频部分能量增加,由此实现了故障的识别。这说明基于小波包分解的整循环征兆提取与故障识别方法有效、可行。

To detect the gas leaking fault of the diesel engine valve, with the ameliorated arithmetic of the wavelet packet decomposition, the vibration signals of the diesel engine cylinder lid in a whole working cycle was analyzed and the whole cycle symptom of vibration diagnosis from the decomposition coefficients picked up. It is shown from the chart of the whole cycle eigenvectors that the energy of the low frequency parts is weaker but that of the high frequency parts is stronger in gas-leaking state than in normal state. Therefore, the fault identification is realized, and the effectiveness and feasibility of this method is clearly manifested.