Research on Algorithms for Forecasting Renewable Energy Demand

This study aims to improve the accuracy of renewable energy demand forecasting and solve the instability and unpredictability problems faced by traditional methods in dealing with energy demand forecasting. By using the seasonal autoregressive integrated moving average (SARIMA) model, combined with time series forecasting methods and causal models, this study conducted a detailed analysis of solar and wind energy consumption data in a certain region in 2023. The SARIMA model used in the study has been parameter optimized and adjusted to adapt to the seasonal fluctuations and trend changes of the data. The results show that the SARIMA model can effectively predict the monthly consumption of renewable energy, and the forecast error rate is mostly maintained within 3%, accurately capturing the seasonal changes in energy demand.