Dynamic performance of restructured wholesale power markets with learning generation companies: an agent-based test bed study
In April 2003, the U.S. Federal Energy Regulatory Commission (FERC) proposed a new market design for U.S. wholesale power markets. Core features of this design include oversight of operations by some form of Independent System Operator (ISO), a two-settlement system consisting of a day-ahead market supported by a parallel real-time market to ensure continual balancing of supply and demand for power, and management of grid congestion by means of locational marginal pricing. Seven U.S. energy regions are now operating under a variant of FERC's market design. This dissertation undertakes the systematic study of core features of FERC's market design by means of intensive simulation experiments.
Specific studied issues include: the effects of generator learning behaviors on market efficiency and supply adequacy; the effects of changes in generator learning parameters, demand-bid price sensitivities, and generator supply-offer price caps on locational marginal price separation and volatility over time; market efficiency implications of ISO net surplus (congestion rent) collections and redistributions; and the effects of generator economic and physical capacity withholding on generator net earnings and market efficiency. To carry out this research, major extensions of the AMES wholesale power market test bed have been developed. To encourage the accumulation of further research findings, these extended versions of AMES have been released as open-source software.