A Marginal Power Generation Cost Estimation Method Based on Probability Interval Accumulation Considering Bounded Rationality

Estimating the marginal power generation cost is a fundamental work in electricity markets, which facilitates market power regulation and many market-related fields. Traditional marginal cost estimation methods mainly rely on the perfect rationality assumption, which usually deviates from reality. With the increasing attention given to bounded rational bidding behaviors in real power markets, there is an urgent need to develop a method that can estimate the marginal cost from actual market data rather than based on the theoretical assumption of optimal bidding. Thus, we proposed a data-driven method that can estimate individual generators’ most likely cost parameters from historical bidding data. This method does not require the generator’s bids to be perfectly optimal. As long as the actual bidding is between perfect rationality and the marginal cost, it can effectively contribute to the estimation method. Case analysis examples show the feasibility of the method.