Research on Characterization Parameters for Engine Valve Clearance Based on Vibration Signals

To address mechanical faults caused by abnormal engine valve clearance, this study investigates the extraction methods of vibration signal feature parameters. A six-cylinder diesel engine test bench was established to collect cylinder head vibration signals under varying rotational speeds, torque loads, and valve clearance conditions. Four time–domain parameters, including peak-to-peak values, rectified average values, standard deviations, and root mean square amplitude, were extracted alongside the energy distribution of frequency bands obtained through wavelet packet decomposition. The correlations between these parameters and valve clearance states were systematically analyzed. The results demonstrated that all parameters exhibited significant increasing trends with elevated rotational speeds, whereas torque variations exerted minimal influence. Abnormal valve clearance induces significant amplitude increases in time–domain parameters. The energy within frequency bands 5–8 (6–12 kHz) derived from wavelet packet decomposition demonstrates marked sensitivity to valve clearance variations. These findings provide effective technical support for online monitoring of engine valve clearance and early fault warning systems.