Price Spread Direction Prediction Based on an Improved LSTM Model in China’s Electricity Spot Market

With the continuous development of China’s electricity spot markets, the study of bidding strategies is critical for every power generation company. This paper proposes a price spread direction prediction model based on an improved LSTM model for the electricity spot market, aiming to help power generation companies achieve higher revenues. First, we analyze the data and perform feature engineering to enrich the feature information. Second, we design the Multi-Layer LSTM model according to the data characteristics and validate it using actual data from a wind farm in Shandong Province. The results show that our model not only outperforms other models in terms of F1-Score and precision but also demonstrates higher real-world backtesting returns.