Optimization models and techniques for implementation and pricing of electricity markets

Abstract

Vertically integrated electric power systems extensively use optimization models and solution techniques to guide their optimal operation and planning. The advent of electric power systems re-structuring has created needs for new optimization tools and the revision of the inherited ones from the vertical integration era into the market environment.

This thesis presents further developments on the use of optimization models and techniques for implementation and pricing of primary electricity markets. New models, solution approaches, and price setting alternatives are proposed. Three different modeling groups are studied. The first modeling group considers simplified continuous and discrete models for power pool auctions driven by central-cost minimization. the direct solution of the dual problems, and the use of a Branch-and-Bound algorithm to solve the primal, allows to identify the effects of disequilibrium and different price setting alternatives over the existence of multiple solutions. It is shown that particular pricing rules worsen the conflict of interest that arise when multiple solutions exists under disequilibrium. A price-setting alternative based on dual variables is shown to diminish such conflict. The second modeling group considers the unit commitment problem. An interior-point/cutting-plane method is proposed for the solution of the dual problem. The new method has better convergence characteristics and does not suffer from the parameter tuning drawback as previous methods. The robustness characteristics of the interior-point/cutting-plane method, combined with a non-uniform price setting alternative, show that the conflict of interest is diminished when multiple near optimal solutions exist. The non-uniform price setting alternative is compared to a classic average pricing rule. The last modeling group concerns to a new type of linear network-constrained clearing system models for daily markets for power and spinning reserve. A new model and solution approach is proposed. The model considers bids for supply and demand and bilateral contracts and a direct current model for the transmission network. The use of an interior point method that can take advantage of the special structure of the Newton's system is proposed.

The use of optimization models in a market environment is still facing several challenges; this thesis presents developments that help understand the use of complex optimization models for electricity markets; new models, solution approaches and price setting alternatives are proposed.