Considering the Green Environmental Value Attribute of Electricity: A Deep Learning Model for Green Electricity Price Forecasting

Against the backdrop of the "dual carbon" goals, the green electricity trading market is gradually developing. Currently, the formation mechanism of green electricity prices is complex, environmental premiums are not unified, and the impact paths are not clear, making the prediction of green electricity prices extremely challenging. Therefore, this paper proposes a deep learning model for forecasting green electricity prices that takes into account the green environmental value attributes of electricity. Firstly, the mechanism of green electricity price formation is studied to explore the impact paths of its influencing factors; then, a quantification index system for green electricity environmental value is constructed; subsequently, the LSTM neural network parameters are optimized using the Whale Optimization Algorithm (WOA) to build the forecasting model; finally, a simulation experiment is conducted in a green electricity market in North China, and the results show that the proposed model has a prediction accuracy of 91.7%, achieving accurate forecasting of green electricity prices, which can provide reference for market trading entities to formulate strategies.