| dc.description.abstract | Renewable sources of energy such as wind turbine generators and solar panels
have attracted much attention because they are environmentally friendly, do not
consume fossil fuels, and can enhance a nation’s energy security. As a result, recently
more significant amounts of renewable energy are being integrated into conventional
power grids. The research reported in this dissertation primarily investigates the
reliability-constrained planning and operations of electric power systems including
renewable sources of energy by accounting for uncertainty. The major sources of
uncertainty in these systems include equipment failures and stochastic variations in
time-dependent power sources.
Different energy sources have different characteristics in terms of cost, power
dispatchability, and environmental impact. For instance, the intermittency of some
renewable energy sources may compromise the system reliability when they are integrated into the traditional power grids. Thus, multiple issues should be considered in
grid interconnection, including system cost, reliability, and pollutant emissions. Furthermore, due to the high complexity and high nonlinearity of such non-traditional
power systems with multiple energy sources, computational intelligence based optimization methods are used to resolve several important and challenging problems in
their operations and planning. Meanwhile, probabilistic methods are used for reliability evaluation in these reliability-constrained planning and design.
The major problems studied in the dissertation include reliability evaluation of
power systems with time-dependent energy sources, multi-objective design of hybrid
generation systems, risk and cost tradeoff in economic dispatch with wind power penetration, optimal placement of distributed generators and protective devices in power
distribution systems, and reliability-based estimation of wind power capacity credit.
These case studies have demonstrated the viability and effectiveness of computational
intelligence based methods in dealing with a set of important problems in this research
arena. | en |
