Optimization in Electricity Generation

MD23.1 Optimizing Generators Offers in an Electricity Market

• Philip Neame; University of Auckland, Dept. of Engineering Sci., Private Bag 92019, Auckland, , New Zealand; p.neame@auckland.ac.nz
• Andrew Philpott; University of Auckland, Dept. of Eng. Science, Private Bag 92019, Auckland, , New Zealand; a.philpott@auckland.ac.nz
• Geoffrey Pritchard; University of Auckland, Dept. of Stats., Private Bag 92019, Auckland, , New Zealand; pritchar@scitec.auckland.ac.nz

New Zealand has recently implemented an electricity spot market similar to many such markets being introduced around the world. We briefly outline the structure of this market, then consider a generator offering into the market. For a given time period, they must submit an offer stack consisting of a finite number of quantity-price pairs (or tranches)...

MD23.2 Negative Prices & Must-Run Generation Auctions in Electricity Markets

• Geoffrey Pritchard; University of Auckland, Dept. of Stats., Private Bag 92019, Auckland, , New Zealand; pritchar@scitec.auckland.ac.nz
• Andrew Philpott; University of Auckland, Dept. of Eng. Science, Private Bag 92019, Auckland, , New Zealand; a.philpott@auckland.ac.nz
• Philip Neame; University of Auckland, Dept. of Engineering Sci., Private Bag 92019, Auckland, , New Zealand; p.neame@auckland.ac.nz

We consider wholesale spot markets for electricity in which generation is offered in the form of closed bid supply curves that are then dispatched by a central authority to minimize the total revealed cost of delivered power. The dispatch optimization problem is assumed to be convex and approximated by an LP. Markets of this type operate in New Zealand, Australia, England and Wales and parts of North America...

MD23.3 Stochastic Dynamic Programming for Hydro Scheduling using AMPL

• Tom S. Halliburton; Enron Corporation, 15135 Memorial Dr., Apt. 7104, Houston, TX 77079; tom.halliburton@enron.com
• Jan Grygier; Pacific Gas & Electric Co.; jcg8@pge.com

Pacific Gas & Electric have developed a stochastic dynamic programming model as a part of their hydro generation planning process. The AMPL modeling language and its looping features are used to repeatedly solve linear sub problems. A technique proposed by Stedinger et al is used to represent hydrological scenarios.