Design and Implementation of a New Energy Power Generation Forecasting Model Based on Federated Learning Technology

Accurate power generation forecasting helps grid operators balance supply and demand effectively. This reduces the risks associated with the volatility of new energy power generation. It also provides decision support for market participants, enhancing the efficiency of new energy utilization. In this paper, a new energy power generation forecasting model based on federated learning technology is proposed. In this model, power generation groups act as participants, and the forecasting model is collaboratively trained through federated learning while data privacy is protected. The model’s prediction accuracy has been validated through experiments conducted on a simulation platform.