Design and analysis of competitive electricity markets

Ndrio, Mariola

Permalink

https://hdl.handle.net/2142/110442 Copy

Description

Title

Design and analysis of competitive electricity markets

Author(s)

Ndrio, Mariola

Issue Date

2021-04-07

Director of Research (if dissertation) or Advisor (if thesis)

Bose, Subhonmesh

Doctoral Committee Chair(s)

Bose, Subhonmesh

Committee Member(s)

• Sauer, Peter
• Hajek, Bruce
• Dominguez-Garcia, Alejandro
• Srikant, Rayadurgam

Department of Study

Electrical & Computer Eng

Discipline

Electrical & Computer Engr

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

game theory, Nash equilibria, electricity markets, supply function competition, AC optimla power flow, locational marginal pricing

Abstract

This thesis focuses on the study of allocation mechanisms and pricing schemes for the design and analysis of competitive electricity markets. Motivated by the increasing demand-side participation in high- and low-voltage power grids, we consider two-sided competition models where a finite group of producers and consumers compete through scalar-parameterized supply offers and demand bids. Acting as a smooth approximation to supply offers used in practice, scalar-parameterized offers greatly facilitate mathematical analysis while preserving the primary determinants and mechanisms by which market power is exercised in electricity markets. In the framework of a pool-based market, characterized by a central dispatch and pricing mechanism, when strategic, capacity-constrained suppliers face strategic, price- responsive consumers, we show that market allocative efficiency loss and price markup at the Nash equilibrium are bounded. We demonstrate analogous efficiency bounds in the study of inter-area electricity markets where we exploit scalar-parameterized offers to model budget-constrained price arbitrageurs that compete against affine inter-area price spreads. Our analysis provides important insights on the type of behavior that may occur at the equilibrium including the pivotal role assumed by certain players, the impacts of aggregate liquidity and uncertainty as well financial positions in other electricity markets. Through the application of reinforcement learn- ing algorithms we demonstrate that players can discover their equilibrium actions even when they know little to nothing about the game setting. The simplicity of scalar-parameterized supply offers that grant market ac- tors’ one-dimensional action spaces while properly constraining their strategic flexibility, render such offer/bid structures an attractive candidate for the expansion of electricity markets to distribution grids. Motivated by the rapid proliferation of distributed energy resources that increasingly hold value for the grid either as power suppliers or flexible demand, we leverage scalar-parameterized supply offers together with appropriate pricing schemes to design a pool-based market for the retail sector. Our goal is complicated by the underlying physics of distribution grids that render the central dispatch problem, in its full generality, non-linear and non-convex. To get around this difficulty, we exploit semidefinite relaxations of the optimal power flow problem and leverage duality theory to define prices for electricity as the optimal Lagrange multipliers of nodal real and reactive power balance constraints. We demonstrate that such prices stand on sound economic principles that together with scalar-parameterized offers/bids, constitute a comprehensive mechanism for the expansion of markets to the low-voltage side of the electric power grid.

Graduation Semester

2021-05

Type of Resource

Thesis

Permalink

http://hdl.handle.net/2142/110442

Copyright and License Information

Copyright 2021 Mariola Ndrio

Owning Collections