David Flanagan

Project Title: Control-Oriented modelling of Wind Farms using Computationally Efficient Methods

Keywords: Wind Farm Flow Control, Wake Effects, Control-Oriented Modelling, Model Reduction

Based on site conditions and economics, offshore wind turbines are often clustered in farms to best utilise limited site areas. As a result of the constrained spacing of turbines, downstream wind turbines are exposed to reduced wind velocities and increased turbulence intensities. These two phenomena are more broadly categorised as wake effects.

In a wind farm context, wake effects have been shown to reduce power production and increase structural loads on turbines in affected regions. It is clear that effective wind farm flow controls are needed to minimise wake effects and improve overall wind farm performance. 

Conventional control methods are heavily focused on maximising the performance of individual wind turbines. However, studies have estimated that wind farms lose between 10-23% of their potential power due to the interaction effects between individual wind turbines within the wind farm. Therefore, a more holistic approach to wind farm flow control is needed to ensure that these losses are minimised. 

For such a control strategy wind farm models must be developed that are capable of capturing these wake effects. In addition, control oriented wind farm models which are of a lower fidelity, are also necessary to enable real time control. 

My research is focused on developing control oriented wind farm models and controllers using first principles approaches coupled with data driven and machine learning methods. 

This research is supported by the Science Foundation of Ireland

Supervisor:  Dr Breiffni Fitzgerald