Graduation Year

2015

Document Type

Dissertation

Degree

Ph.D.

Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department

Industrial and Management Systems Engineering

Major Professor

Bo Zeng, Ph.D.

Committee Member

Tapas Das, Ph.D.

Committee Member

Alex Savachkin, Ph.D.

Committee Member

Yao Liu, Ph.D.

Committee Member

Balaji Padmanabhan, Ph.D.

Committee Member

Tongxin Zheng, Ph.D.

Keywords

Defender-attacker-defender Model, Distribution Network Planning, Mixed Integer Programming, Network Topology Control, Robust Optimization

Abstract

In this dissertation, we introduce and study robust optimization models and decomposition algorithms in order to deal with the uncertainties such as terrorist attacks, natural

disasters, and uncertain demand that are becoming more and more signicant in power systems operation and planning. An optimal power grid hardening problem is presented as a

defender-attacker-defender (DAD) sequential game and solved by an exact decomposition

algorithm. Network topology control, which is an eective corrective measure in power systems, is then incorporated into the defender-attacker-defender model as a recourse operation

for the power system operator after a terrorist attack. Computational results validate the

cost-eectiveness of the novel model. In addition, a resilient distribution network planning

problem (RDNP) is proposed in order to coordinate the hardening and distributed generation resource placement with the objective of minimizing the distribution system damage

under uncertain natural disaster events. A multi-stage and multi-zone based uncertainty set

is designed to capture the spatial and temporal dynamics of a natural disaster as an extension

to the N-K worst-case network interdiction approach. Finally, a power market day-ahead

solution algorithm for the RUC model. generation scheduling problem, i.e., robust unit commitment (RUC) problem, that takes account of uncertain demand is analyzed. Improvements have been made in achieving a fast

Share

COinS