Characterizing Influence of Transition to Turbulence on the Propulsive Performance of Underwater Gliders

Two models of underwater gliders were tested in a wind tunnel: one corresponding to a legacy shape commonly used in contemporary vehicles and the other a scaled down version of a new design. Performance of the two vehicles was characterized over a range of speeds and angles of attack. Particular attention was paid to the effect of sharp features along the hulls of the two vehicles and how they affect the observed flow regime. It has been shown that the new design, which uses a bow shape designed to encourage natural laminar flow, benefits from a 10% reduction of parasitic drag and 13% increase in lift-to-drag when the hull surface is smooth. The legacy glider, made up of a faired bow and a cylindrical hull, suffers from laminar separation and up to 100% increase in induced drag if the flow over its bow is prevented from transitioning to a turbulent state before encountering adverse pressure gradient at lower Reynolds numbers. This results in lowering of attainable speed at shallow glide path angles, whereas the associated parasitic drag reduction is demonstrated to increase the maximum velocity of the glider when moving at glide slopes greater than approximately 30°. Underwater gliders are autonomous underwater vehicles (AUVs) that rely on using a buoyancy engine to ascend or descend through the water column, and by adjusting their pitch, they can use this vertical motion to develop forward thrust from their hydrofoils. This propulsion method allows them to undertake long-endurance missions, often several months long (Eriksen 2003; Rudnick et al. 2004; Graver 2005). Hydrodynamic performance of an underwater glider is primarily governed by its lift-to-drag (L/D) ratio, which dictates the minimum glide path angle the vehicle may adopt, and drag coefficient, which affects the maximum forward speed the glider may achieve for a fixed amount of vertical force developed (Graver 2005). It is thus important to minimize the drag of the AUV to allow it to perform longer deployments and gather more science data without increasing the size of the engine. Because of the typical Reynolds numbers on the vehicle hulls being less than 106 and of the order of 104–5 on the appendages, laminar and transitional flow regions may occur. Correctly identifying these is critical to achieve a robust performance prediction.