The electricity spot market is a barometer of the supply and demand situation

making the accurate spot price forecasting crucial for market participants. As the spot market in China transitions from pilot to full-scale operation

the government is improving market rules and clarifying the transition mechanism between the medium- and long-term market and the spot market

significantly impacting the spot price. However

existing research has overlooked the influence of medium- and long-term market factors on the spot price

and the forecasting accuracy needs to be improved. In order to further capture the key factors affecting electricity price

this paper extracts the coupling relationship between the electricity quantity/price of the medium- and long-term market and the spot price

and proposes a new feature

namely the net spot market bidding capacity

for forecasting of the day-ahead electricity price. First

the annual

monthly

and ten-day trends of day-ahead prices are extracted by ensemble empirical mode decomposition (EEMD). Then

considering market rules

the approximate electricity quantity in the medium- and long-term market is decomposed from load forecast data

and the quantity is eliminated from the original market bidding capacity to derive a net spot market bidding capacity with higher correlation to the day-ahead price. Finally

the new feature is input into the subsequent machine learning model with other features to better capture underlying patterns and output the final forecasting results. Validation on Shanxi spot market data shows that the proposed method effectively improves the accuracy of the day-ahead price forecasting.