| Abstract | In imperfect electricity markets, generation companies (GenCos) could develop bidding
strategies to maximize their profits. A GenCo has to make a decision based on limited
information. For example, a GenCo does not know the actual system market clearing
price (MCP) beforehand since it depends on bidding behaviors of other participants in the
market. Thus, developing an optimal bidding strategy considering the risk of market price
unce1tainty in the competitive environment is a challenging task for a GenCo. And the
price uncertainty could directly impact to GenCo's profit, which requires efficient tools
for the risk management in competitive electricity markets.
For developing a GenCo's bidding strategy with risk consideration, the tradeoff technique
is a conventional method to deal with the multi-objective optimization problem. The two
conflict objectives, the profit maximization and the risk minimization, are combined to be
a single objective. And the optimal solution depends on the weight or risk parameter
selection. Thus, the technique requires a GenCo's decision making on a proper value of
the risk parameter. Moreover, a number of feasible bidding scenarios need to be provided
for selecting a satisfied bidding strategy. This may require many computational efforts in
providing the optimal bidding strategy.
In this dissertation, mean-standard deviation ratio (MSR) derived from mean-variance
portfolio selection theory is used to indicate an optimal risky bidding strategy for a
Genco in a competitive electricity market. The maximum MSR implies the optimal risky
bidding scenario. As non-convex operating cost functions of thermal generation units and
a number of constraints including minimum up/down time, generation limits, and bid
price limits are considered, an efficient optimization technique is required to provide the
optimal bidding solution. Here, self-organizing hierarchical particle swarm optimization
with time-varying acceleration coefficients (SHPSO-TVAC) is proposed to solve the
optimal bidding strategy problem with the objective function of the MSR maximization.
And Monte Carlo (MC) simulation is employed to estimate rivals' behaviors in a
competitive environment. The proposed bidding strategy is implemented in a uniform
price spot market with multi-period trading. In addition, various stochastic search
approaches including genetic algorithm (GA), classical PSO (CPSO), PSO with timevarying inertia weight (PSO-TVIW), and PSO with time-varying acceleration coefficients
(PSO-TVAC) are also compared in providing the optimal bidding solution for a GenCo.
Test results indicate that the SHPSO-TV AC approach could provide better bidding
solutions for a GenCo compared with the other stochastic search approaches. Especially,
it could provide a higher MSR solution for the optimal risky bidding strategy. With the
optimal risk attitude, the proposed MSR maximization bidding strategy could facilitate a
GenCo in managing the risk of profit variation in a spot electricity market since it does
not require any risk parameter specification. And it could efficiently provide the optimal
risky bidding strategy for a GenCo in a competitive electricity market.
lV
There are a number of institutions supporting me for the doctoral study and research.
First, I would like to thank my affiliation, Rajamangala University of Technology
Rattanakosin (RMUTR), for offering the opportunity of the doctoral study at AIT. And I
would like to thank Energy Policy and Planning Office (EPPO), Ministry of Energy,
Thailand, for the study grant of the HM Queen Sirikit Scholarships (Queen HRD). Also, I
thank AIT for its fellowship fund. A significant research grant is from the Office of the
Higher Education Commission, under the Ministry of Education, Thailand, where I would
like to extend my thankfulness. Finally, I would like to thank a number of people who are
in charge of paper works for the grant process and a number of my RMUTR colleagues
who work hard for my students during my leaving period.
My special thank is for Ms. Prow Choompradit for her wonderful relationship. She
significantly encourages me to finish concrete tasks and shares me a fantastic life.
Lastly, my warmest gratefulness is extended to my mother and sister for their never ending encouragement and supports. And I would like to dedicate my entire productive
works and merit to them. |