Monitoring the tribological behaviour of piston ring-cylinder liner in a four-cylinder diesel engine by using acoustic emission signals

Internal combustion (IC) engines are the primer power source for different means of transport and power generation systems. Their performance and healthy conditions are decisive for ensuring safe and efficient operations of such important systems. This paper focuses on investigating acoustic emission (AE) signals for achieving an early and reliable online monitoring of engine combustion and tribological behaviour. Based on changes in major nonstationary AE events such as combustion shocks, valve impacts, it is not difficult to discriminate a misfired cylinder from noisy AE signals. However, AE responses from tribological characteristics are much weaker and submerged in the major events. Considering that tribological AE is correlated more with oil viscosity and sliding speed, the quasi-stationary AE responses around the mid-stroke are examined in this study for monitoring changes in lubrication conditions. Especially, a misfired cylinder can operate with a stronger viscous shear effect because the lubricant is with a higher viscosity due to reduced cylinder temperature. To characterise the quasi-stationary AE, an iterative analysis of raw AE signals based on the wavelet packet transform (WPT) and the principle component analysis (PCA) is adopted to suppresses the nonstationarity significantly, which results in a more stationary AE signatures for reflecting the tribological behaviour of ring-liner system. Furthermore, the average amplitudes from theses stationary signatures show an increased AE amplitude with engine speeds and become even higher with misfired conditions. This consistency with tribological AE excitations of engine lubrication conditions confirms that AE measurements can be based for lubrication condition monitoring. Along with monitoring combustion performance and valve conditions, AE can be a more efficient approach for engine condition monitoring