Experimental characterisation of flow effects on marine current turbine behaviour and on its wake properties

Abstract

Experimental results of tests carried out to investigate the hydrodynamics of marine current turbines are presented. The objective is to build an experimental database in order to validate the numerical developments conducted to characterise the flow perturbations induced by marine current turbines. For that purpose, we used a tri-bladed horizontal axis turbine. The work is dedicated to measuring the behaviour of the system and to characterising the wake generated by the turbine. The efficiency of the device is quantified by the measurement of the thrust and the amount of power generated by the rotor for various inflow conditions, whereas the wake is characterised by Laser Doppler Velocimetry. Particular attention is paid to the flow characteristic effects on the performance of a 0.70 m diameter turbine. The load predictions on the structure and the measured performance of the turbine over its working range of currents and rotational speeds are presented. The results showed that this kind of turbine is sensitive to the quality of the incoming flow. The turbulence intensity effects on turbine behaviour and on its wake are also characterised in order to study how the far wake decays downstream and to estimate the effect produced in downstream turbines.

Effects of turbulence on tidal turbines: Implications to performance, blade loads, and condition monitoring

“…However, it was not our intention to create site specific turbulence in this study, but to generate a range of flows with different turbulence intensities and length scales in order to investigate the specific contribution turbulence has to performance and blade loads. It was shown that increasing the turbulence intensity reduced the mean thrust and power ( Figure 24), and is consistent with previous studies that found a 10% reduction in power and thrust with an increase in turbulence intensity from ~5-15% [13]. However, it was found that the opposite occurs for increasing length scales which resulted in the power and thrust increasing (Figure 25).…”

“…Another measure of eddy size may now be defined as the injection length scale and calculated using equation 13.…”

“…Studies using model turbine rotors have also shown significant variations when operated in flows with different turbulence intensities [12]- [14]. It has been found that increasing turbulence intensity from 3-15% typically reduces the power and thrust coefficients of a 3 bladed turbine rotor by roughly 5-10% [13]. Further studies have been performed on a turbine in uniform oscillatory flow using a towing tank with oscillating motion superimposed on the towing carriage [15], [16].…”

“…Here we present an experimental campaign to investigate the effects of turbulence on turbine performance and blade loadings. In previous studies the turbulence was typically characterised as a turbulence intensity and limited control provided by adding, or removing flow straighteners in the flume [13]. It is therefore likely that the scales of turbulence will also differ.…”

Three-dimensional numerical simulations of straight-bladed vertical axis tidal turbines investigating power output, torque ripple and mounting forces

“…Improvements of simulation accuracy at high λ may be possible using newly developed transitional turbulent models that can account for this laminar-turbulent transition behaviour [19]. Over prediction of Cp at high λ may also occur as a result of blockage errors that were not accounted for in EFD results which could artificially increase EFD Cp [26], and may also occur due to differences in the turbulence intensity levels between the CFD models and EFD testing, as high turbulence intensity levels can delay stall [27], leading to increases in Cp especially at high λ [28], with no turbulence intensity measurements recorded during EFD testing to compare to CFD turbulence levels. Significantly both CFD models were able to accurately capture the effect of geometrical changes on maximum Cp, which was simulated to within 14.3% and 6.3% of EFD results for turbines A and B…”