A Cycle-to-Cycle Variation Extraction Method for Flow Field Analysis in SI IC Engines Based on Turbulence Scales

To adhere to stringent environmental regulations, SI (spark ignition) engines are required to achieve higher thermal efficiency. In recent years, EGR (exhaust gas recirculation) systems and lean-burn operation has been recognized as key technologies. Under such operating conditions, reducing CCV (cycle-to-cycle variation) in combustion is critical to the enhancement of overall engine performance. Flow-field CCV is one of the considerable factors affecting combustion in engines. Conventionally, in research on flow fields in SI engines, the ensemble average is used to separate the measured velocity field into a mean component and a fluctuation component, the latter of which contains a CCV component and a turbulent component. To extract the CCV of the flow field, previous studies employed spatial filter, temporal filter, and POD (proper orthogonal decomposition) methods. Those studies used a constant- separation filter size for the whole crank angle, although the turbulence scales change rapidly during the intake and compression stroke processes. Hence the definition of filter size has some room to be explored in order to take account of these features. The objective of this research is to improve the method of separating the CCV component using turbulence scales. For this purpose, high-speed PIV measurement was conducted on the symmetrical vertical plane for an optical IC engine at repetition rates of 12 kHz (1 C.A. deg. resolution) for the whole in-cylinder area and 48 kHz (0.25 C.A. deg. resolution) for the plug position. The measured data were separated into CCV and turbulent components by using the proposed filter, whose size was selected adaptively considering the integral time scale of the turbulent flow. The effect of time resolution on the filter size was then elucidated.