Martin Densing: Katalogdaten im Frühjahrssemester 2019

NameHerr Dr. Martin Densing
Adresse
Paul Scherrer Institut
Büro OHSA/E09
5232 Villigen PSI
SWITZERLAND
Telefon0563102598
E-Maildensingm@ethz.ch
DepartementInformationstechnologie und Elektrotechnik
BeziehungDozent

NummerTitelECTSUmfangDozierende
227-0530-00LOptimization in Energy Systems6 KP4GG. Hug, H. Abgottspon, M. Densing
KurzbeschreibungThe course covers various aspects of optimization with a focus on applications to energy networks and scheduling of hydro power. Throughout the course, concepts from optimization theory are introduced followed by practical applications of the discussed approaches.
LernzielAfter this class, the students should have a good handle on how to approach a research question which involves optimization and implement and solve the resulting optimization problem by choosing appropriate tools.
InhaltIn our everyday’s life, we always try to take the decision which results in the best outcome. But how do we know what the best outcome will be? What are the actions leading to this optimal outcome? What are the constraints? These questions also have to be answered when controlling a system such as energy systems. Optimization theory provides the opportunity to find the answers by using mathematical formulation and solution of an optimization problem.
The course covers various aspects of optimization with a focus on applications to energy networks. Throughout the course, concepts from optimization theory are introduced followed by practical applications of the discussed approaches. The applications are focused on 1) the Optimal Power Flow problem which is formulated and solved to find optimal device settings in the electric power grid and 2) the scheduling problem of hydro power plants which in many countries, including Switzerland, dominate the electric power generation. On the theoretical side, the formulation and solving of unconstrained and constrained optimization problems, multi-time step optimization, stochastic optimization including probabilistic constraints and decomposed optimization (Lagrangian and Benders decomposition) are discussed.