Session R1: Industrial Applications IV: Marine Hydrokinetic Energy Conversion

Performance and Near-Wake Flow field of A Marine Hydrokinetic Turbine Operating in Free surface Proximity

The current experimental investigation aims at understanding the effect of free surface proximity and associated blockage on near-wake flow-field and performance of a three bladed horizontal axis marine hydrokinetic turbine. Experiments were conducted on a 0.14m radius, three bladed constant chord turbine in a 0.61m$\times $0.61m test section water channel. The turbine was subjected to various rotational speeds, flow speeds and depths of immersion. Experimental data was acquired through a submerged in-line thrust-torque sensor that was corrected to an unblocked dataset with a blockage correction using measured thrust data. A detailed comparison is presented between blocked and unblocked datasets to identify influence of Reynolds number and free surface proximity on blockage effects. The percent change in $C_{\mathrm{p}}$ was found to be dependent on flow velocity, rotational speed and free surface to blade tip clearance. Further, flow visualization using a stereoscopic particle image velocimetry was carried out in the near-wake region of turbine to understand the mechanism responsible for variation of $C_{\mathrm{p}}$ with rotational speed and free surface proximity. Results revealed presence of slower wake at higher rotational velocities and increased asymmetry in the wake at high free surface proximity.   

The influence of surface waves on performance characteristics and wake measurements of a horizontal axis marine current turbine

Performance characteristics and wake flow field results are presented for a 1/25 scale, 0.8 m diameter two bladed horizontal axis marine current turbine. The performance data and 2D PIV measurements were obtained in the 380 ft tow tank at the United States Naval Academy. The turbine was towed at a constant carriage speed of Utow $=$ 1.68 m/s with turbine loading resulting in a nominal tip speed ratio of 7. Conditions with two regular waves were investigated. The first wave had a 2.3 second period and 0.18 m wave height, while the second wave had a 2.0 second period and a 0.20 m wave height. The waves were selected to have the same energy. Flow field measurements were obtained with an underwater PIV system comprised of two submersible housings. The forward looking submersible contained the laser sheet forming optics and the side looking submersible included a camera and remote focus/aperture electronics. Planar wake measurements were obtained 2 diameters downstream of the rotor plane. Flow field structures, as well as wave phase averaged mean velocities turbulence statistics will be presented and compared to the baseline case without surface waves.   

The effect of wall proximity on energy harvesting using a pitching and heaving hydrofoil

Measurements of energy harvesting using a heaving and pitching hydrofoil with an aspect ratio 4.5 were taken in three different situations: unconfined, in close proximity to one wall and in close proximity to two walls. Measured lift and torque forces were used with the hydrofoil positions to calculate the efficiency of energy extracted from the flow. There was a modest increase in efficiency with one-wall proximity, while a pronounced increase in efficiency (up to 50{\%}) was realized when the hydrofoil operated between two walls with strong confinement. The lift coefficient of the hydrofoil experienced a noticeable increase in two-wall proximity cases with the strong confinement, which directly contributed to the increase in efficiency of energy harvesting. In the case of two-wall confinement, we found that the optimal frequency and pitch amplitude were higher than those for both the free stream and the one-wall proximity cases. The power extracted from the heaving motion was greatly enhanced by two-wall proximity at high frequencies and high pitch amplitudes and these gains exceeded the additional power required to execute the pitching motion, resulting in the net increase in energy harvesting effectiveness.   

Effect of Free Stream Turbulence on the Performance of a Marine Hydrokinetic Turbine

The effects of controlled and elevated levels of free stream turbulence on the performance characteristics of a three bladed, constant chord, untwisted marine hydrokinetic turbine is tested experimentally. Controlled homogeneous free stream turbulence levels ranging from 3{\%} to $\sim$20{\%} are achieved by employing an active grid turbulence generator that is placed at the entrance of the water channel test section and is equipped with motor controlled winglet shafts. In addition to free stream turbulence, various (turbine) operating conditions such as the free stream velocity and rotational speed are varied. A comparison of performance characteristics that includes the mean and standard deviations of the power coefficient ($C_{P})$, and thrust coefficient ($C_{T})$ will be presented and compared to the case of a laminar free stream with FST levels \textless 1{\%}.   

Experimental/Numerical Comparison of Turbine Efficiency and Wake Structure in an Array of 3 Scale-Model Marine Hydrokinetic Turbines

Numerical simulations and experiments are conducted for axial-flow Marine Hydro-Kinetic (MHK) turbines operating in a flume. This study aims to understand the influence of coherent structures in high Reynolds number wakes on energy extraction and dynamical rotor control processes. In experiments, rotor torque and rotational position measurements are collected, and the flow field characterized by simultaneous imaging with particle image velocimetry. The performance of 3 turbines are characterized under varying downstream spacing and lateral offsets. To study effects of unsteady hydrodynamics, the turbines are outfitted with open-loop and close-loop feedback controls and compared to the case of uncontrolled rotor. In numerical simulations, different tiers of turbine models are evaluated to discern tradeoffs in fidelity to physics versus cost. Analogous ``actuator methods'' are included from Large-Eddy-Simulations and Reynolds-Averaged-Navier-Stokes, where the models impose body forces upon the flow field in form of disks, lines, or surfaces. An aeroelastic model coupled to LES predicts the dynamical response of rotors to upstream wakes and ambient turbulence. These comparative studies inform how simulations can be scaled up to inform design of utility-scale MHK power plants.   

Flow structure in the near wake of a horizontal axis marine current turbine under steady and unsteady inflow conditions

Near wake flow field results are presented for a 1/25 scale, 0.8 m diameter (D) two bladed horizontal axis tidal turbine. The 2D PIV measurements were obtained in the USNA 380 ft tow tank for two inflow conditions. The first case had steady inflow conditions, i.e. the turbine was towed at a constant carriage speed (Utow $=$ 1.68 m/s) and the second case had a constant carriage speed and incoming regular waves with a period of 2.3 seconds and 0.18 m wave height. The underwater PIV system is comprised of two submersible housings with forward looking submersible containing laser sheet forming optics, and the side looking submersible includes a camera and remote focus/aperture electronics. The resulting individual field of view for this experiment was nominally 30x30 cm$^{\mathrm{2}}$. Near wake mapping is accomplished by ``tiling'' individual fields of view with approximately 5 cm overlap. All measurements were performed at the nominal tip speed ratio (TSR) of 7. The mapping is accomplished in a vertical streamwise plane (x-z plane) centered on the turbine nacelle and the image pair captures were phase locked to two phases: reference blade horizontal and reference blade vertical. Results presented include distribution of mean velocities, Reynolds stresses, 2D turbulent kinetic energy. The discussion will focus on comparisons between steady and unsteady case. Further discussion will include comparisons between the current high resolution PIV measurements and the previous point measurements with the same turbine at different lateral planes in the same flow conditions.   

Wake structure of axial-flow hydrokinetic turbines in tri-frame arrangement

Marine and hydro-kinetic (MHK) energy hold promise for future of sustainable energy generation. Tri-frame of turbines, three turbines mounted on vertices of a triangle, are an effective way to build a power producing array of hydrokinetic turbines in marine environment. Large eddy simulation (LES) is used to simulate the flow past a tri-frame and characterize its wake. Full geometry of all three turbines in the tri-frame is resolved using the Curvilinear Immersed Boundary (CURVIB) method of Kang et al. (2011). High fidelity solution of flow field is obtained owing to the inclusion of detailed geometry of the turbines. Excellent agreement is obtained with the experiments conducted in a flume at Saint Anthony Falls Laboratory (SAFL). The wake evolution of the three turbines is compared to that of an isolated single turbine. The differences in wake dynamics are highlighted to elucidate the importance of turbine wake interaction in an array. The simulations indicate lower levels of TKE and lower levels of momentum deficit in the wake of the upstream turbine of tri-frame compared to the other turbines. Analysis of the far wake recovery is useful for the optimal MHK array design.   

Development of a towing tank PIV system and a wake survey of a marine current turbine under steady conditions

A submersible particle image velocimetry (PIV) system was designed and built at the U.S. Naval Academy. The system was used to study the wake of a scale-independent horizontal axis marine current turbine. The turbine is a 1/25th scale model of the U.S. National Renewable Energy Laboratory's Reference Model 1 (RM1) tidal turbine. It is a two-bladed turbine measuring 0.8 m in diameter and featuring a NACA 63-618 airfoil cross-section. The wake survey was conducted over an area extending 0.25D forward of the turbine tip path to 2.0D aft to a depth of 1.0D beneath the turbine output shaft in the streamwise plane. Each field of view was approximately 30 cm by 30 cm, and each overlapped the adjacent fields of view by 5 cm. The entire flow field was then reconstructed by registering the resultant vector fields together into a single field of investigation. Results include the field of investigation from a representative case, for the mean velocity field averaged over approximately 1,000 realizations, and turbulent statistics including turbulence intensities, Reynolds shear stresses, and turbulent kinetic energy.   

Experimental and Numerical Study on Performance of Ducted Hydrokinetic Turbines with Pre-Swirl Stator Blades.

Ducts (also called shrouds) have been shown to improve performance of hydrokinetic turbines in some situations, bringing the power coefficient (C$_{p})$ closer to the Betz limit. Here we investigate optimization of the duct design as well as the addition of stator blades upstream of the turbine rotor to introduce pre-swirl in the flow. A small scale three-bladed turbine was tested in a towing tank. Three cases (bare turbine, with duct, and with duct and stators) were tested over a range of flow speeds. Important parameters include duct cross-sectional shape, blade-duct gap, stator cross-sectional shape, and stator angle. For each test, C$_{p}$ was evaluated as a function of tip speed ratio (TSR). Experimental results were compared with numerical simulations. Results indicate that ducts and stators can improve performance at slower flow speeds and lower the stall speed compared to a bare turbine, but may degrade performance at higher speeds. Ongoing efforts to optimize duct and stator configurations will be discussed.   

Design of Bi-Directional Hydrofoils for Tidal Current Turbines

Tidal Current Turbines operate in flows which reverse direction. Bi-directional hydrofoils have rotational symmetry and allow such turbines to operate without the need for pitch or yaw control, decreasing the initial and maintenance costs. A numerical test-bed was developed to automate the simulations of hydrofoils in OpenFOAM and was utilized to simulate the flow over eleven classes of hydrofoils comprising a total of 700 foil shapes at different angles of attack. For promising candidate foil shapes physical models of 75 mm chord and 150 mm span were fabricated and tested in the University of New Hampshire High-Speed Cavitation Tunnel (HiCaT). The experimental results were compared to the simulations for model validation. The numerical test-bed successfully generated simulations for a wide range of foil shapes, although, as expected, the $k- \omega$ - SST turbulence model employed here was not adequate for some of the foils and for large angles of attack at which separation occurred. An optimization algorithm is currently being coupled with the numerical test-bed and additional turbulence models will be implemented in the future.   

Phase Resolved Angular Velocity Control of Cross Flow Turbines

Cross flow turbines have a number of operational advantages for the conversion of kinetic energy in marine or fluvial currents, but they are often less efficient than axial flow devices. Here a control scheme is presented in which the angular velocity of a cross flow turbine with two straight blades is prescribed as a function of azimuthal blade position, altering the time-varying effective angle of attack. Flume experiments conducted with a scale model turbine show approximately an 80\% increase in turbine efficiency versus optimal constant angular velocity and constant resistive torque control schemes. Torque, drag, and lateral forces on one- and two-bladed turbines are analyzed and interpreted with bubble flow visualization to develop a simple model that describes the hydrodynamics responsible for the observed increase in mean efficiency. Challenges associated with implementing this control scheme on commercial-scale devices are discussed. If solutions are found, the performance increase presented here may impact the future development of cross flow turbines.