The emergence of tidal energy as a key renewable energy source requires the development of computational design models for accurate prediction of turbine performance and wake effects whilst also being computationally efficient. In this paper, we develop and validate a three-dimensional CFD model for vertical axis turbines, which achieves high accuracy. We also investigate the limitations of two-dimensional models and present a blockage correction for improved prediction. The two-dimensional blockage correction model is potentially attractive for preliminary design studies due to its computational advantage over three-dimensional models.

Tidal energy researchers and developers use experimental testing of scaled devices as a method of evaluating device performance. Much of the focus to date has been on horizontal axis turbines. This study is focused on a novel vertical axis turbine which incorporates variable-pitch blades and a flow accelerator. The research involves laboratory testing of scale model devices in a recirculating flume. Computational fluid dynamic modelling is used to reproduce the measured flow data to investigate disparities in experimental data. The results show that the device is capable of achieving localised flow acceleration of up to a factor of 2 above the freestream velocity and achieved a mechanical power efficiency of 40%.

No abstract

This paper presents the numerical modelling of a novel vertical axis tidal turbine that incorporates localised flow acceleration and variable-pitch blades. The focus is to develop a computational fluid dynamics model of a 1:20 scale model of the device using ANSYS® Fluent®. A nested sliding mesh technique is presented, using an outer sliding mesh to model the turbine and additional inner sliding meshes used for each of the six blades. The turbine sliding mesh is embedded in an outer static domain which includes the flow accelerating bluff body. Modelled power performance and velocity data are compared with experimental results obtained from scale model tests in a recirculating flume. The predicted power curves show general agreement with the measured data; the relative difference in maximum performance coefficient for example, is just 5.7 %. The model also accurately reproduces measured flows downstream of the turbine. The verified and experimentally validated model is subsequently used to investigate the effects of the variable-pitching and number of blades on device performance.