Strategic Bidding for GENCOs using Reinforcement Learning Methodology based on LMP in Electricity Market

The structure of competition among energy suppliers in the production sector of electricity markets has made generation companies search for maximization of their profits by making strategic decisions. In this regard, generation companies try to bid a suitable price higher than their marginal costs to get a larger share of supply of electricity in the power market in their competitions with other generators. The research objective was to propose a method based on identification of optimal strategic interactions of agents in the electricity market to allow for achievement of the Nash equilibrium point. To this end, a heuristic method based on the reinforcement learning algorithm in the pool market structure was used to determine optimal bidding strategies for generation units. On the other hand, considering the effect of the transmission network capacity in a multi-area system, the market clearing process based on local marginal price (LMP) was studied in the single- and multi-generator states. The proposed strategy was tested on the Nord Pool four-area market. Simulations results of this network reflect the capability of the proposed learning algorithm for determining the optimal strategy of generation companies and achieving the market equilibrium point.