基于EMD的Hilbert变换的柴油机气缸套磨损故障诊断

为准确地对柴油机进行状态监测和故障诊断,运用基于经验模态分解(Empirical Mode Decomposition,EMD)的Hilbert变换,对船用低速柴油机缸盖振动加速度信号进行分析.改变柴油机的转矩和转速,测量各工况下缸盖的振动信号,并对其进行分析;将柴油机的6个缸分为3组,分别采用3种不同性能的气缸油,让柴油机在额定工况下连续运转,测取活塞环与气缸套在不同间隙时的缸盖振动信号.结果表明,当转矩和转速同时增大时,振动加速度信号特征频带变化不大,但能量相应增加;当转矩和转速保持不变,活塞环与气缸套间隙增大时,振动加速度信号特征频带发生改变,并且能量增大.与快速傅里叶变换(Fast Fourier Transform,FFT)法及功率密度谱法的对比表明,基于EMD的Hilbert变换可以有效处理缸盖的振动信号,并可以运用到气缸套磨损故障的诊断中.

Fault diagnosis of diesel engine cylinder liner wear using Hilbert transform based on EMD

In order to make condition monitoring and fault diagnosis for diesel engines accurately, the Hilbert transform based on Empirical Mode Decomposition (EMD) is used to analyze the cylinder head vibration acceleration signals of low speed marine diesel engines. The cylinder head vibration signals under different working conditions with different torque and speed are measured and analyzed. 6 cylinders of a diesel engine are divided into 3 groups, which are respectively oiled with cylinder oils of 3 different performances; the diesel engine is continuously operated under a rated condition, and then the cylinder head vibration signals are measured for different clearances between the piston ring and the cylinder liner. The results show that, when the torque and the speed increase at the same time, the characteristic frequency band of the vibration acceleration signal changes a little, while its energy increases correspondingly; when the torque and the speed keep constant, and the clearance between the piston and the cylinder liner increases, the characteristic frequency band of the vibration acceleration signal changes, and its energy increases. Compared with Fast Fourier Transform (FFT) and the power density spectrum method, the Hilbert transform based on EMD is effective for the vibration signal processing and can be applied to cylinder liner wear fault diagnosis.