Research & development - Genk | More than two weeks ago
imec has always been pushing the boundaries of PV cell and module technology development, resulting in ever-increasing performance and reliability figures at decreasing cost. Owing to these developments, we will soon reach a point where PV will become an abundant source of clean energy. However, due to its intermittent and non-dispatchable nature, it will present additional challenges in terms of system stability and grid congestion and will fundamentally challenge the conventional (central) way of operating a power system. Integrating and operating PV in a system-efficient way will be the next big challenge to increase penetration levels to evolve to a (near) 100% renewable energy system.
Dispatchable hybrid PV plants, i.e. large-scale solar paired with energy storage, have the potential to become the future backbone of the 21st-century grid. Smart, innovative energy management and power plant control software that communicates with the grid & market operators is an important part of that future. Such control software requires an accurate forecast of PV generation in order to define optimal storage or curtailment setpoints according to the different markets and ancillary services it is targeting (e.g. peak shaving, frequency control, ramp rate control...). In view of this emerging need, imec’s Energy System team developed and patented a new approach for short-term PV forecasting (“Nowcasting”), combining a sky dome imager with physics-based models and machine learning techniques. In contrast to more conventional approaches (e.g. satellite based), it allows to provide more fine-grain forecasts with higher accuracy under highly dynamic conditions (e.g. scattered clouds).
This thesis will analyse and quantify the added value of this Nowcasting approach for energy management applications in hybrid power plants. Targeting multiple markets and ancillary services, a sensitivity analysis will be performed to assess the impact of PV forecasting accuracy on the technical and economic performance of a hybrid PV plant.
Type of project: Thesis
Duration: 9 months
Required degree: Master of Engineering Technology, Master of Engineering Science
Required background: Electromechanical engineering, Electrotechnics/Electrical Engineering, Energy
Supervising scientist(s): For further information or for application, please contact: Joris Lemmens (Joris.Lemmens@imec.be)
Only for self-supporting students.