Cross-flow turbines have complex, periodic, and unsteady hydrodynamics. As the turbine rotates, the blades alternate between power production and stall, leading to design and control challenges. While traditionally operated under constant speed or constant torque control, previous works have demonstrated that varying the rotation rate within a single rotation, known as intracycle control, can greatly increase the power output. However, this advanced control strategy comes at the expense of increased rotor loads and high peak-to-average power. Power-maximizing intracycle control has been observed to occur under conditions when the phase and amplitude of the blade angular velocity align with the phase of peak torque, however, the sensitivity of these optima and tradeoffs between power and thrust are poorly understood. We present, the interplay between the performance of sinusoidal intracycle angular velocity control and blade-level hydrodynamic features, as we move toward a mechanistic understanding of generalized cross-flow turbine control schemes that can satisfy a range of objectives.
The performance of one- and two-bladed turbines is studied in conjunction with in-rotor flowfield measurements using particle image velocimetry (PIV). While intracycle control can take many functional forms, we varied the angular velocity sinusoidally. This control strategy is represented non-dimensionally as λcontrol = λ0 + Aλsin(2θ + ϕ) where θ is the angular position, Aλ is the amplitude, and ϕ is the phase shift. Performance experiments were undertaken across a two-dimensional parameter sweep bounded by ϕ = [0, 2π] and Aλ = [0, 0.64λ0], about a mean tip-speed-ratio (λ0= 2), that was identified as power-optimal for constant speed control of the two-bladed system. From the collected data, we identify the sensitivity of performance to swept parameters, the range of power-suboptimal operating conditions, and optima under other metrics such as the power to thrust ratio. Three kinematic conditions were visualized using PIV to identify the hydrodynamic mechanism driving changes in performance. The three chosen kinematics correspond to conditions of optimal and suboptimal control for maximum power output, as well as the maximum of total power to peak in-plane rotor force.
Intracycle control improved efficiency by up to 15% for the two-bladed turbine as compared to constant speed control. Performance enhancements existed across a broad region, with little sensitivity to changes in amplitude but greater sensitivity to the phase, particularly at higher amplitudes. Control kinematics for optimal power to thrust ratio were found and show performance can be improved by 3% while reducing the forces on the turbine by 6% relative to the constant speed case, indicating that intracycle control has utility outside of power maximization. Conditions, where loading is decreased without a loss in performance, would allow less costly support structures when turbines are deployed. Intracycle power improvements are expected to be correlated with x,y, and z in the flowfield, and how this varies with control conditions will be explored in this talk.