Developing Bidding Strategies for Transmission and Wholesale Energy Markets

It is well documented in the literature that inadequacy of the analytical models has been a primary reason for the low economic stability of deregulated power industry. Some of the modeling inadequacies can be attributed to the difficulties of analyzing nonzero sum multiplayer stochastic games that are inherent in deregulated power market operations. This paper presents a two-tier game theoretic modeling framework for obtaining equilibrium bidding strategies in transmission and wholesale energy markets. The two-tier game model works in the following manner. The upper tier game obtains the bilateral and FTR market bidding strategies, which serve as input to the lower tier game that generates bidding strategies for the day ahead and real time markets. A reinforcement-learning (RL) based solution algorithm is developed for the two-tier game model. The algorithm seeks to learn Nash equilibrium bidding strategies for the participants in both transmission and energy (bilateral, DA, and RT) markets. Though the fulfillment of Nash equilibrium criterion by the strategies learned by the RL algorithm is not fully guaranteed, the algorithm has been shown in a previous study to obtain Nash policies in most of the instances of a numerical benchmark (Grid-World) problem. The modeling and the solution methodology presented in this paper can be used by the designers and the participants of the market for evaluating alternative market configurations and developing effective bidding strategies respectively. The model application is demonstrated using a sample power network.