Flow and heat transfer characteristics in adaptive latticework structure designed using shape memory alloys

In this study, numerical simulation and particle image velocimetry experiments were performed to investigate an adaptive flow and heat transfer control structure based on shape memory alloy operating in the range of Reynolds number 100 to 2000. The adaptive structure could operates with low drag coefficient in the flat state, and actively enhance heat transfer with a latticework channel structure in the warping state. Different warping heights and Reynolds numbers were investigated to analyze the flow patterns and heat transfer enhancement mechanism. The warping of the shape memory alloy leads to flow deflection and generates a z-direction velocity component around the gap, which is the main reason for local heat transfer enhancement. The average velocity in the main stream plane (XY plane) close to the heat transfer surface increases sharply with warping. This study found significant increase in normal velocity and in-plane velocity (up to 0.17 and 0.94 times reference inlet velocity uin) in the plane at a distance of 0.1 times hydraulic diameter Dh from the wall due to warping, which strengthens the disturbance near the wall and thus enhances heat transfer. The longitude vortices generated in the structure are divided into the main vortex in the center of the channel and the corner vortex in the sub channel. The corner vortex is mainly caused by the gap flow between the shape memory alloy and the wall surface. In the warping state, the average value of the main stream velocity components near the wall is relatively close to the numerical simulation results (by deviation of 0.04 uin). In numerical results, the structure can achieve a heat transfer regulation ratio of 2.5-3.3 times and a resistance regulation ratio of 5.2-14.6. The performance parameters are not monotonous with the increase of the warping height or the Reynolds number. When the warping height is 0.6 Dh and Reynolds number equals to 400, maximum performance factor (1.34) is achieved.