AI Techniques for Smart Grid Load Forecasting: An Overview With Focus on External Factors

Accurate load forecasting is essential for managing smart grids effectively. It helps grid operators anticipate demand changes, optimize resources, and prevent overloads. However, achieving high accuracy is challenging because energy use can change a lot, and renewable energy adds more uncertainty. Traditional methods often fall short in addressing these issues, so better tools are needed. This study explores how Artificial Intelligence improves load forecasting in smart grids. Unlike most reviews that focus only on short-term forecasting, this paper covers short-term, mid-term, and long-term horizons. Each type needs different kinds of data, accuracy, and computational efficiency, which AI is well-suited to handle. The paper highlights recent advancements and trends in building reliable and sustainable energy systems. It focuses on studies that adapt to real-time conditions like weather forecasts, social events, and economic indicators. The findings show that AI methods can predict energy demand more accurately, especially when uncontrollable factors are incorporated. This study also shows that hybrid models can make forecasts more reliable. Finally, it points out areas where more research is needed to solve challenges in managing smart grids using AI.