基于改进局域波分析的柴油机故障诊断

Fault diagnosis of diesel engine based on improved local wave analysis

针对柴油机实测振动信号的非线性非平稳的特点,应用局域波法对柴油机实测振动信号进行分析.对局域波分解存在的模态混叠问题使其难以获得准确的基本模式分量问题,提出一种基于总体平均局域波分解的改进局域波分析,并应用于不同磨损状态下的柴油机机体振动信号的分析.该方法利用总体平均局域波分解获取无模式混淆的基本模式分量,通过局域波边际谱分析信号能量随瞬时频率的变化特征.工程实测分析结果验证了应用该方法进行柴油机缸套磨损故障诊断的有效性.

In respect to the nonlinear and nonstationary character- istics of the surface vibration signals measured from diesel en- gine, local wave analysis is introduced to analyze the diesel cyl- inder block surface vibration signals. However, local wave de- composition（ LWD ） sometimes can not extract intrinsic mode functions accurately because of the mode mixing. Due to this problem, an improved local wave analysis based on ensemble lo- cal wave decomposition （ELWD） is proposed, and the instanta- neous frequency and the energy distribution of signals can be ex- tracted by local wave boundary-spectrum. Results indicate that the proposed method is feasible and effective in fault diagnosis of diesel engine cylinder liner wear.