Novel Metrics for Validation of PIV and CFD in IC Engines

In-cylinder flow motion has a significant effect on mixture preparation and combustion. Therefore, it is vital that CFD engine simulations are capable of accurately predicting the in-cylinder velocity fields. High-speed planar Particle Image Velocimetry (PIV) experiments have been performed on a single-cylinder GDI optical engine in order to validate CFD simulations for a range of engine conditions. Novel metrics have been developed to quantify the differences between experimental and simulated velocity fields in both alignment and magnitude. The Weighted Relevance Index (WRI) is a variation of the standard Relevance Index that accounts for the local velocity magnitudes to provide a robust comparison of the alignment between two vector fields. Similarly, the Weighted Magnitude Index (WMI) quantifies the differences in the local magnitudes of the two velocity fields. The WRI and WMI are normalised and combined to produce a combined metric, the Combined Magnitude and Relevance Index (CMRI), that quantifies the differences between two flow fields in both magnitude and alignment simultaneously.

PIV measurements were made every 5°ca in the central tumble plane during the induction and compression strokes. The WRI, WMI and CMRI metrics are used to validate numerical simulations of the motored in-cylinder flow measured with PIV for a range of valve lift profiles and engine speeds.