An Agent-Based Electrical Power Market

This demonstration shows an agent-based model for the electricity power market, in which the optimal power flow is determined in a bottom-up fashion. Here, each agent controls a single electrical node consisting of several power generators, loads (consumer demand), and is connected to neighbouring nodes through transmission lines. Furthermore, each of the components has associated physical constraints, such as the line and generators’ capacities. Through a process resembling tatonnement in markets, the optimal system solution which maximises social welfare is reached within a few iterations. The demonstrator visualises this process and also shows how the various constraints affect the system behaviour and how this changes with different settings.

Authors

• Jaime Cerda Jacobo
• David De Roure
• Enrico Gerding University of Southampton

Keywords

• Electrical Power Markets
• Distributed Constrained Optimization
• Quadratic Separable Programming