Êtes-vous un étudiant de l'EPFL à la recherche d'un projet de semestre?
Travaillez avec nous sur des projets en science des données et en visualisation, et déployez votre projet sous forme d'application sur Graph Search.
The European Union's Green Deal aims for a 55% reduction in greenhouse gas emissions by 2030. To reach this goal, a massive integration of Renewable Energy Sources (RES) into the power grid is necessary. As RES become a large part of the electricity generation mix, there is a growing need for grid services. Hydropower Plants (HPPs ) emerge as vital assets for their controllability and crucial for mitigating the intermittency of RES. In this PhD thesis, we propose optimal control techniques for the enhanced provision of ancillary services by HPPs. The thesis is divided into distinct parts, to answer the research question: "How can the integration and control of HPPs with new technologies be optimized to enhance their role in ancillary service provision and ensure asset longevity?" The first part focuses on aFRR provision. In this domain, Pumped Storage Power Plants (PSPs) play a fundamental role due to their ability to absorb, or produce, electricity. However, a challenge arises with PSPs equipped with reversible pump-turbines, as they are machines operated in on-off mode, offering limited power regulation capabilities. To address this challenge, we introduce a control framework for PSPs to optimize the dispatch and aFRR reserve allocation among multiple units. The objective of the optimal dispatching is to maximize efficiency and reduce the number of start and stop of the machines. In the second part of the thesis, we discuss the contribution of Run-of-River (RoR) HPPs to the provision of FCR, presenting a control strategy for optimal discharge management. Moreover, we explore the impact of FCR provision on RoR HPPs equipped with Kaplan turbines, introducing a modeling method for online efficiency estimation. The obtained models are used to compute a new CAM relation. However, despite advancements in monitoring and movement reduction in RoR HPPs, a large portion of servomechanism activity still comes from FCR provision. For this reason, we proceed by focusing on Battery Energy Storage Systems (BESSs) as autonomous FCR providers, laying the basis for the last part of the thesis.Indeed, while HPPs are fundamental assets for FCR provision, they suffer from FCR-related wear, increasing maintenance needs. Conversely, BESSs offer quick response but are limited by their energy and power capacity. In the last part of this thesis, we explore HPP and BESS integration, introducing an optimal control framework, based on a double layer MPC. Moreover, we propose an alternative approach: re-purposing aging Kaplan turbines into variable-speed propellers by employing full-size frequency converters. On top of providing optimal control strategies, it becomes necessary to have experimental facilities able to test and validate the proposed solutions. For this reason, we present an experimental facility where different control techniques can be tested on reduced-scale models of hydraulic machines hybridized with a BESS. On the platform, we validate the MPC-based control for BESS-hybridized RoR HPPs. The results demonstrate the superior performance of the proposed controller, compared to simpler techniques like dead-band control or the standalone RoR scenario. Moreover, the experiments reveal that Kaplan turbines repurposed as VARspeed are a viable alternative to BESS hybridization. In conclusion, this thesis presents a set of strategies and control frameworks tailored for HPP operators, which are implementable at the plant level.
François Maréchal, Julia Granacher