Development of an predictive model for uniflow scavenging in ultra-long-stroke low-speed marine engines based on Taylor vortex

Abstract

The demand for energy conservation and emission reduction in marine diesel engines is becoming increasingly stringent. Current two-stroke marine diesel engines generally adopt ultra-long stroke uniflow scavenging to achieve this goal. However, there is a lack of a deep understanding of this form of scavenging, which leads to the absence of an accurate scavenging model for ultra-long stroke uniflow scavenging, and the calculation discrepancy for scavenge efficiency would be as high as 16%. Consequently, the design of an efficient and clean marine engines face significant challenges. This study proposes a novel predictive model for uniflow scavenging in ultra-long-stroke low-speed marine engines by integrating the Taylor vortex phenomenon, which was found by experimental and computational analysis. Firstly, high-speed 3D particle image velocimetry measurements on a full-scale optical test rig and transient computational fluid dynamics (CFD) simulations are conducted to characterize velocity field evolution, axial vorticity dynamics, and residual exhaust gas patterns. The results reveal that the uniflow scavenging flow exhibits quasi-steady behavior under operational conditions, and the scavenging port inclined angles (10°–30°) and engine load significantly influences vortex development and residual gas stratification. Based on the finds, a dual-zone scavenging model is developed, partitioning the cylinder into Taylor vortex-dominated peripheral regions and central mixing zones, incorporating vortex intensity, port geometry, and scavenging airflow parameters to predict residual gas distribution with ±3.5% deviation against CFD results.