WAKE2GRID

This research project aims at determining if a wind farm can be globally controlled, in an accurate and robust way, in order to participate in grid balance, via, in particular, the provision of ancillary services.

​

The complex wake phenomena occurring within the wind farm require control strategies based on a model. This allows the prediction of the wake behavior and the development of robust controllers. We here envisage a physics-based wake model that captures the unsteadiness of the wake, including wake meandering, as accurately as possible and on short-time scales.

This model predicts over a thousand seconds in a space of a second, and provides a simplified yet reasonably faithful representation of the wakes. In this project, at least two control strategies will be considered: an approach based on the optimal control and a faster but potentially less optimal strategy based on the heuristic control. Both strategies will use the predictions given by the wake model.

The knowledge of the fine wake physics can also be leveraged in the prediction of the impact of the control actions on the turbine fatigue loads. We thus want to broaden our perspective as such ancillary services should not involve additional fatigue for the turbines, but rather enforcing a prescribed distribution of wear and tear through the wind farm and possibly minimizing the maintenance costs.

The control strategies developed in this research project will be assessed in a high-fidelity numerical environment, representing the rotor behaviors and the complex wake dynamics in an accurate way.