Aristotle University Of Thessaloniki (AUTH)
Ανάλυση χειραγώγησης αγορών ηλεκτρικής ενέργειας με ευφυείς πράκτορες
Abstract
dc:descriptionThe goal of the present doctoral thesis is the study of generators' behavior in a deregulated power market and the analysis of market power exercise phenomena, like capacity withholding and tacit collusion which can emerge in the market. The generator's profit maximization problem is approached through the scope of reinforcement learning theory. The generator's behavior is modeled with Q-learning algorithm, while the policy used is based as the metropolis criterion of simulated annealing. The power market is described in terms of game theory and is modeled as a repeated game. The players' behavior is modeled the reinforcement learning algorithm. Furthermore, a brief description of the process followed during simulations is given, and the model limitations are cited. Finally, two examples are presented: one two-node system with two and eight generators and one five-node systems with five generators.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2007
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Tellidou, Athina
- Τελλίδου, Αθηνά
Subjects
dc:subject × 18- Απελευθερωμένη αγορά ηλεκτρικής ενέργειας
- Παρακράτηση δυναμικού
- Σιωπηρή συμπαιγνία
- Ευφυείς πράκτορες
- Ενισχυτική μάθηση
- Αλγόριθμος μάθησης Q-learning
- Επαναλαμβανόμενα παίγνια
- Deregulated electricity market
- Capacity withholding
- Tacit collusion
- Intelligent agents
- Reinforcement learning
- Q-learning algorithm
- Repeated games
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Engineering and Technology
- Electrical Engineering, Electronic Engineering, Information Engineering
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/19211