Civil Engineering, Aidan Wimshurst Thesis

On the outboard sections of horizontal axis rotor blades (that are not enclosed within a duct or shroud), vorticity is shed into the wake of the rotor. The shed vorticity induces a downwash at the rotor plane and span- wise flow accelerations along the blade, which causes the blade loading to drop off as the tip is approached. These tip flow effects are currently not adequately accounted for by reduced order rotor models (such as the blade element momentum and actuator line methods) which are frequently used to represent wind and tidal turbine rotors in large simulations. Hence, the rotor thrust and torque may be considerably over-predicted by these models if they are not corrected appropriately for tip flow effects. In this thesis, the tip loss mechanism experienced by both wind and tidal turbine rotor blades is examined directly in a series of simulations, using computational fluid dynamics. Two different correction methods that can account for tip flow effects are then presented and critically evaluated. Both methods are shown to lead to a significant improvement in the accuracy of the computed blade loading, which is principally achieved by allowing the sectional force vector to reduce in magnitude and rotate towards the streamwise direction as the tip of the blade is approached.

Tip flow effects also reduce the strength of the suction peak developed on the suction surface of the blade. This reduction has considerable implications for the operation of tidal turbine rotors, since tidal turbine operation may be limited by cavitation inception and cavitation inception is most likely to initiate on the outboard sections of the rotor blade (where the static pressure reaches a minimum). In this thesis, a cavitation analysis of two diff t tidal turbine rotors is carried out. When tip fl w effects are properly accounted for, cavitation inception is shown to be less likely at a given operating condition (tip-speed-ratio and submersion depth). Hence, industry standard cavitation analyses that are based on the blade element momentum method (and do not adequately account for tip flow effects) are shown to be currently overly-conservative.

In a separate study, tidal power extraction is examined in a computational domain where the sea bed slopes in the streamwise direction. When the sea bed slopes downwards in the streamwise direction, the velocity shear across the swept area of the rotor increases, increasing the power available for extraction. However, the increased velocity shear also increases the strength of the suction peak developed on the blade at top dead centre, so the device is more likely to cavitate at a given tip-speed-ratio. Conversely, when the sea bed slopes upwards in the streamwise direction, the incident velocity profile is more uniform, so the device is less likely to cavitate at a given tip-speed-ratio. Power extraction is also found to be more efficient on the upwards facing slope, as the flow through the swept area of the rotor is accelerated by the downstream flow passage constriction. At higher blockage ratios, the strength of the suction peak is further increased by the acceleration of the flow through the swept area of the rotor, so the device is even more likely to cavitate at a given tip-speed-ratio.