Abstract

Homogeneous charge compression ignition (HCCI) is a promising low-temperature combustion technique for low-emission internal combustion engines. Unlike conventional engines, HCCI lacks a direct ignition control mechanism, necessitating closed-loop combustion control. This study proposes a phenomenological-based, cost-effective, and non-intrusive approach using vibration data analysis to determine essential combustion parameters. Experiments were conducted on a single-cylinder research engine with an accelerometer attached to the engine head. The engine operation envelope covered the whole engine’s operating area in naturally aspirated HCCI mode. Wavelet analysis revealed that combustion-related frequencies centered around 500 Hz, independent of operating conditions. The correlation-seeking analysis included peak acceleration amplitude and its crank angle with peak heat release rate (HRR) data. The peak HRR location was accurately identified within one degree when vibration amplitude exceeded the 100 m/s2 threshold. This encompassed 98.5% of the analyzed combustion cycles. The peak HRR prediction accuracy had a maximum error below 21% and was suitable to monitor reaction rates, especially in incomplete combustion and high ringing cycles.