论文部分内容阅读
针对柴油机缸盖表面振动信号的非平稳时变特点，提出了基于小波包分析的柴油机振动诊断方 法。给出了小波包变换的一种改进算法，通过移频处理，克服了原算法中的频率混叠现象，使分解序列 的排列顺序与频带的线性划分顺序—一对应。通过对完整工作循环内的缸盖振动信号进行小波包分 解，实现了整循环诊断特征向量的快速提取。试验结果表明，该方法在柴油机振动诊断中是有效可行 的，对其它复杂机械的振动诊断同样具有参考价值。
Aiming at the non-stationary and time-varying characteristics of diesel engine cylinder head surface vibration signals, a method of diesel engine vibration diagnosis based on wavelet packet analysis is proposed. An improved algorithm of wavelet packet transform is given. By frequency shift processing, the frequency aliasing phenomenon in the original algorithm is overcome, so that the sequence of the decomposition sequence and the linear division sequence of the frequency band are one to one. Through the wavelet packet decomposition of the vibration signal of the cylinder head in the complete working cycle, the feature vector of the whole cycle diagnosis can be quickly extracted. The experimental results show that this method is effective and feasible in the diagnosis of diesel engine vibration. It is also of reference value for the diagnosis of vibration of other complex machines.