Real-Time Short-Term Load Forecasting for Remote Power Grid with Satellite Internet

Traditional power load forecasting methods rely on terrestrial communication networks, which are often unavailable or unreliable in remote power grid. To address this limitation, we introduce satellite internet as a communication backbone, effectively managing the challenges of data loss caused by satellite transmission. This paper presents a novel approach to real-time short-term load forecasting for remote power grid by integrating satellite internet technology. We employ the Prophet algorithm for load forecasting after preprocessing, leveraging its ability to model the periodicity, seasonality, and holiday effects of power load data. The proposed method not only improves the reliability of forecasting under satellite communication conditions but also enhances prediction accuracy. Simulation results using real power load data from a specific region demonstrate the effectiveness of our approach. This work lays the foundation for the broader application of satellite internet in future smart grid systems, particularly in underserved and remote locations.