Hydrokinetic Power Harnessing Utilizing Vortex Induced Vibrations through a Virtual c-k VIVACE Model.

Abstract

An apparatus, VCK, is designed and built to replace the physical damper and springs of the VIVACE (Vortex Induced Vibrations for Aquatic Clean Energy) Converter with a motor-controller system. VIVACE harnesses hydrokinetic energy of water currents by converting it to mechanical energy using VIV. Next, it converts the mechanical energy of cylinders in VIV into electricity. VCK enables conducting high number of model tests quickly as damping and springs are set by software rather than hardware. The controller provides a damper-force and spring-force feedback based on displacement and velocity measurements, thus, introducing no additional artificial force-displacement phase lag, which would bias energy conversion. The damping of even such a simple spring-damper-mass system is strongly nonlinear, even in air, particularly away from the system’s natural frequency and strongly affects modeling near the ends of the VIV synchronization range. System identification in air reveals nonlinear viscous damping, static friction, and kinetic friction. Hysteresis, which occurs in the zero velocity limit, is successfully modeled by a proposed nonlinear dynamic damping model LARNOS (Linear Autoregression combined with NOnlinear Static model). To obtain the optimal VIVACE power at a given current speed, extensive VIV tests are performed with the VCK VIVACE apparatus for Reynolds number 40,000<Re<120,000 and damping 0<ζ<0.16 in the Low Turbulence Free Surface Water Channel of the Marine Renewable Energy Laboratory at the University of Michigan. Scarce VIV data exist in that parametric subspace of Re and ζ. From the VIV tests, the optimal damping for energy harnessing is found for velocity 0.41m/s<U<1.11m/s using spring stiffness 400N/m<k<1800N/m. Thus, the VIVACE converter power envelope is developed. The following experimental observations are made: (1) In the high-lift TrSL3 and TrBL0 flow regimes, high-amplitude, high-damping VIV is maintained. (2) VIV strongly depends on Reynolds. (3) The amplitude ratio (A/D) increases with Reynolds number within the upper branch of the VIV synchronization range. (4) In TrSL3/TrBL0, A/D of 1.78 was achieved for a smooth cylinder routinely in low damping. (5) Power density of 98.2 W/m3 at 2 knots is achieved including space between cylinders. This exceeds previous measurements by a factor of five.