Investigation of the effects of realistic non-homogenous turbulence on structural loading and the performance of tidal energy devices

Unlike turbulence in the less dense, higher velocity and unbounded flow experienced by wind turbines, structural loading and the performance of tidal stream generators is more heavily influenced by the turbulent characteristics of flow in the operating environment. Whilst the characteristics of mean flow at tidal sites are well understood and can be modelled with some confidence, there is a comparative lack of knowledge regarding turbulence characteristics. Although a range of tidal stream generators (TSG's) are currently being developed and implemented, this lack of certainty in the effects of turbulence serves to hinder progress through less accurate prediction of reliability and performance, thus economic viability, and generally results in very conservative design approaches which are inherently more expensive to implement. Despite cautious design, turbulence has been identified as a possible cause in component failures that have occurred in operational units.
Characterisation of turbulence at potential tidal stream sites has gained increasing focus in the last few years with several papers being published and a field study at several potential sites recently having been completed by Partrac - the results of which are not in the public domain. Despite this focus, few papers specifically investigating the effects of turbulence on tidal steam generators are available in the literature and none have been found regarding the effects turbulence characteristics that are spatially variable over the cross sectional area of the device.
This project proposal was formed after discussion with Dr Venugopal and will be associated with the currently ongoing EPSRC funded FloWTurb project. Measurement data from tidal sites will be used to develop a computational model, implemented with CFD software, of realistic non-homogeneous turbulence in tidal channel flow. This model will then be used to investigate its effects on the structural loading (particularly fatigue loading) of tidal stream generators with the possibility to model actual devices and/or components currently being implemented. The complexity of the model may then be extended to include wave loading, with further work investigating effects on performance of both single individual generators and arrays a possibility.
Experimental validation of results will be sought via scale laboratory testing, preferably using the FloWave facility in conjunction with the FloWTurb project. A component part of this work will include investigation of methods for conditioning flow in the experimental facility to closely match desired turbulence characteristics in a reliable and controlled fashion. Development work for this can be undertaken using the small flume available within the School of Engineering. The results of the project will ultimately be used to provide recommendations for more effective design with regards to turbulent flow characteristics at actual tidal stream sites.