Adaptive optimization model based on supply function equilibrium in modern power markets

In this paper, an adaptive optimization model based on a closed-loop control system is developed to regulate the strategic bidding process of generation companies (GenCOs) in day-ahead electricity markets. Each day, the bidding problem of each GenCO is submitted in the form of a supply function consisting of 24 sub-problems, one for each hour of the next day. The hourly market clearing price and the total demand of the next day are the unknown values in the bidding problem that should be estimated by the concerned GenCO. The GenCOs, as the main players in the market, receive feedback signals for market clearing price and demand for each hour of the previous day, based on which they set their bidding for the next day. In the optimization model, the limitations on the production level and production change rate are considered in terms of the minimum and maximum quantities constraints. To better adapt to the market demand and price dynamics beforehand, we also used an adaptive forecasting algorithm for the next day's demand and clearing price. Using this adaptive dynamic model, the network operator can clear the market based on the bids received from the GenCOs and the consumers. As we concentrated on the GenCO side, as the most influential player of electricity markets, the bids from the demand side are considered here as a whole and modeled by a linear function. Finally, the real market data from the day-ahead Nordic electricity market (Nord Pool) are used as the case study to verify the effectiveness of the proposed model and its adaptive algorithm. The results show that the GenCO that uses the proposed model can gain more profit in comparison to those that take non-strategic behavior (naive strategy) in the market.