The invention relates to the technical field of
electric power markets, in particular to an AI-based power curve optimization method for a
new energy field
station oriented to a spot market, and aims to construct a day-ahead spot market optimal power curve
declaration model based on spot rules in an
electric power trading market. The invention provides an AI-based power curve optimization method for a
new energy field
station oriented to a spot market, and the method comprises the steps: obtaining data, building a day-ahead clearing
electricity price prediction model, calculating a historical
price difference through a historical day-ahead clearing
electricity price and a real-time clearing
electricity price, building a plurality of models for predicting the
price difference, and carrying out the prediction of the
price difference. And calculating a price difference direction and an adjustment probability by combining a statistical relationship between the clearing
electricity price and the price difference direction in the historical day, correcting the adjustment proportion based on the adjustment probability and the power predicted value before adjustment, and calculating the adjusted power predicted value. In conclusion, after the adjustment direction and the adjustment proportion of the power curve are determined, the profit-making optimal power prediction curve of the
new energy station in the day-ahead market is determined.