Title:
Stress-informed model predictive control of hybrid hydropower
Abstract:
Hydropower is the first source of renewable production worldwide and a strategic asset of many countries, including Canada, where it accounts for 60% of the total generation. An application that recently came to prominence in the hydropower industry is the notion of hybridization. It refers to retrofitting an existing plant with a local battery system to increase its response time, decrease the wear and tear of the critical mechanical components, and delay expensive maintenance and refurbishments. When operating a hybrid hydropower plant, a problem of practical and immediate interest is how to design the control setpoints of the hydropower plant and the battery (power setpoint splitting). This seminar discusses a model predictive control (MPC) for a hybrid hydropower plant's power setpoint spitting problem. For the first time in the technical literature, this method includes explicit constraints on the mechanical stress of critical components in the decision-making process. These constraints are derived from equivalent circuit models of the hydropower plant, which are suitably linearized and reformulated into a receding-horizon optimization problem. Simulation results with a medium-head hydropower plant show that the penstock's service life can be increased by a factor of 2 while still providing quick balancing power to the grid. This work is within the EU-funded project XFLEX HYDRO (www.xflexhydro.net).
Bio:
Presented by Dr. Fabrizio Sossan, Associate Professor, MINES ParisTech - PSL Universities, France.
Fabrizio is an associate professor at MINES ParisTech - PSL Universities in France. His research interests are planning, scheduling, and control paradigms for distributed energy resources in behind-the-meter (consumer and power plant levels) and grid (distribution and transmission) applications. His objective is to understand how these heterogeneous applications can be harmonized and coordinated to enable feasible techno-economic operations of future power grids and a transition to renewable resources. Fabrizio got his Ph.D. from the Technical University of Denmark. Before his current appointment, he was a postdoctoral researcher at EPFL (Switzerland), visiting scientist at NREL (USA), and scientist at ETHZ (Switzerland).