Evaluation of wave-turbulence decomposition methods applied to experimental wave and grid-generated turbulence data

Tidal energy turbines are significantly impacted by surface gravity waves and high turbulence levels, which lead to increased blade loads and consequent device fatigue. Accurately describing turbulence quantities is a key factor to improve numerical models and ensure the minimum necessary device longevity. The application of an appropriate wave-turbulence decomposition technique substantially improves estimates and must be debated by the tidal energy community. This work analyses two wave-turbulence decomposition techniques: i) linear wave theory and ii) Synchrosqueezing Wavelet Transform (SWT) based on physical modelling. The experiments were conducted in a towing tank and the physical model used Froude scaling to simulate wave heights and frequencies based on field measurements from a tidal energy candidate site in Australia. Turbulence was generated using a grid with 47 mm openings replicating turbulence intensities of 17%, 23% and 25%. Both decomposition techniques present limitations in wave frequencies above 0.7 Hz. Our results suggest that the SWT technique is more versatile and more efficient under various conditions. Restrictions to the linear wave theory method are discussed with regards to ADCP geometry and possible non-linearity. The variability obtained from the waveturbulence decomposition techniques emphasizes the importance of establishing guidelines for turbulence characterization in tidal energy sites.