Electricity price forecasting model based on MLP-mixer

Abstract

Accurate electricity price forecasting is of great importance to all participants in the market, which can provide powerful support for them to make wise decisions in the unpredictable market. In this paper, we propose to use the MLP-Mixer model as a new technique for electricity price forecasting. This model considers the various factors affecting electricity price fluctuations and can use the MLP-Mixer model for concise and practical feature extraction and information interaction, ultimately achieving more accurate electricity price forecasting. The effectiveness of the proposed model is verified by using data from ERCOT, Texas electricity market. Empirical results show that the proposed model can significantly improve the forecasting accuracy, and achieves the optimal results on four evaluation metrics, which fully demonstrates the model's effectiveness for electricity price forecasting.