Stratégies de contrôle de ferme d’éoliennes basées sur une modélisation dynamique simplifiée des sillages pour la gestion d’un compromis entre production d’énergie et limitation de la fatigue structurelle

IFP Energies nouvelles (IFPEN) is a major research and training player in the fields of energy, transport and the environment. From research to industry, technological innovation is central to all its activities, structured around three strategic priorities: sustainable mobility, new energies and responsible oil and gas.
In the field of wind energy, operators are now focusing on using wind turbines located in wind farms in the best possible way, to either produce the maximum energy possible, either produce the right amount of energy at the right time, such that they provide services for the electrical grid, while limiting the wind turbine mechanical stress, in order to eventually minimize the cost of energy. It is possible to limit the interactions between a turbine wake and the downwind turbines by controlling its yaw angle and power produced, and thus alleviating the production losses and mechanical fatigue. In this context, our central question will be “How to robustly minimize a wind farm cost of energy via a control algorithm using a dynamic wind farm flow model, and how to implement it in real world?”. Indeed, the “implementation in real world”, “dynamic wind farm flow model” and “cost of energy minimization” aspect of things are very important. 
On one hand, most of current works focus on cases where the farm is operated in normal conditions, whereas it is of primal importance to detect and manage cases where the farm is in abnormal operating conditions, for a robust real-world implementation. 
On the other hand, most of the works are using steady state models for wind farm control. Therefore, the capacity to derive an optimal control problem, relying on innovative dynamic wind farm flow models, constitutes an important contribution of the thesis. 
Eventually, some recent works had as an objective to either maximize energy production, either regulate power production with wind turbines load alleviation as secondary objective. However, very few contributions focused on the explicit minimization of energy cost, which is one of the main drivers for operators. We are thus facing a sound scientific, and economic, challenge.
The PhD results will contribute to three majors advances:
1.    Contribute to the implementation and deployment of enhanced, robust and efficient wind farm control algorithms on commercial wind farms;
2.    Emphasize and quantify the added value of using dynamic wind farm flow models in place of steady ones for wind farm flow control;
3.    Define a cost criterion allowing to efficiently minimize the cost of energy over long time horizons.
The candidate must be graduated of a master's degree in mathematics or mechanical engineering with a preferred specialization in automatic, optimization or signal processing. Knowledge in machine learning would be appreciated.