Planning tidal stream turbine array layouts using a coupled blade element momentum – computational fluid dynamics model

Abstract

A coupled blade element momentum-computational fluid dynamics (BEM-CFD) model is used to conduct simulations of groups of tidal stream turbines. Simulations of single, double and triple turbine arrangements are conducted first to evaluate the effects of turbine spacing and arrangement on flow dynamics and rotor performance. Wake recovery to free-stream conditions was independent of flow velocity. Trends identified include significant improvement of performance for the downstream rotor where longitudinal spacing between a longitudinally aligned pair is maximised, whereas maintaining a lateral spacing between two devices of two diameters or greater increases the potential of benefitting from flow acceleration between them. This could significantly improve the performance of a downstream device, particularly where the longitudinal spacing between the two rows is two diameters or less. Due to the computational efficiency of this modelling approach, particularly when compared to transient computational fluid dynamics simulations of rotating blades, the BEM-CFD model can simulate larger numbers of devices. An example of how an understanding of the hydrodynamics around devices are affected by rotor spacing can be used to optimise the performance of a 14 turbine array is presented. Compared to a regular staggered configuration, the total power output of the array was increased by over 10%. Highlights  for a longitudinally spaced pair of turbines, the downstream device will have a lower performance even at 40 diameters' spacing because wake recovery is slow.  for a low lateral spacing between a pair of turbines (3 diameters or less), flow recovery is slow between them because of wake expansion.  flow acceleration between a pair of laterally spaced turbines can improve the performance of a third device positioned further downstream but laterally between them.  the most benefit from such flow acceleration is gained by minimising the longitudinal spacing between the two rows of devices.  for a small lateral spacing, performance of the downstream device is compromised.  based on the patterns observed for two and three turbine arrangements, the overall performance of a 14 turbine staggered array was improved by over 10%.

A Comparison of Numerical Modelling Techniques for Tidal Stream Turbine Analysis

“…In the comparison between these two models, we are primarily interested in the velocity deficit in the turbine wake. An understanding of turbine wakes is vital if tidal turbines are to be deployed in arrays [24] (as turbine wakes will inevitably impinge on other turbines located downstream), and such arrays are the only way tidal turbines can be economically viable [25].…”

Aspects of tidal stream turbine modelling in the natural environment using a coupled BEM–CFD model

“…(2)) as the source term S i . A complete description of the method and validation against tank data is given in [18], and comparisons of the approach against experimental data for 2 and 3 rotors is given in [19].…”

“…5. As in the previous cases, the turbines are positioned mid depth in the 30 m deep channel and the rotors have 3 blades with a diameter of 10 m. The layout is similar to an experimental setup reported by [20] and a parametric study of three turbines rotating in the same direction was investigated previously in [19]. The reader is directed to [6] for a blade resolved CFD analysis of contrarotating turbines.…”

“…Ref. [19] considers arrays of turbines and reports validation for three turbines against the work of [20] before considering 14 turbine arrays.…”

A Metamodel Simulation Based Optimisation Approach for the Tidal Turbine Location Problem

“…In the first type of approach, highly simplified one dimensional tidal flow models are employed (Bryden & Couch, 2007;Garrett & Cummins, 2008;Vennell, 2010Vennell, , 2011Vennell, , 2012. In the second type of approach, more complex multidimensional tidal flow models are adopted (Lee et al, 2010;Divett et al, 2013;Malki et al, 2014). The simplified model approach possesses the appeal of simplicity; however, this approach cannot capture the complex nonlinear fluid flow interactions between turbines.…”