Abstract In competitive electricity markets, electric power producers compete through energy and ancillary services and therefore, in order to maximize their profit , they need to make decisions on energy and ancillary services bidding, arbitraging, and risk management. In this thesis, new methods are developed to guide power producers in day-ahead market. At first, a new method for forecasting based bidding strategy is presented which applies Bayesian neural networks to forecast electricity prices using published historical data and the bidding strategy index as input. The forecasted price is then used to develop the optimal bidding strategy for two types of participants, i.e. risk averse and risk seeker. Besides, influence of suppliers’ emission of pollutants and transmission lines congestion on market participant bidding behavior is analyzed in the framework of game theory to assess the equilibrium methods for price makers’ bidding strategy. In an attempt to achieve a model that covers different type of requirements for various power producers, a new model for the static self-scheduling based bidding strategy in the joint energy and reserve market is proposed. The proposed model, not only considers all the operational constraints of generating units for supplying energy and reserves, but also incorporates line flow constraints to ensure network security. Moreover, the uncertainties of electricity prices as well as forced outage of generating units and transmission lines are also incorporated to the proposed model. A new multiobjective viewpoint which deals with the objective functions related to emission of pollutants and risk is also proposed which includes both environmental and techno-economical aspects of generation companies. Several case studies with different complexity are simulated on two IEEE test systems to test the proposed models. The results indicate the effectiveness and capability of the proposed techniques. |
