This study propose a method for predicting power generation based on time series data using Artificial Intelligence. Power generation is influenced by various factors such as weather conditions, seasonal changes, and equipment performance, and accurately predicting these changes is crucial for enhancing power supply stability and efficiency. In this study, an LSTM (Long Short-Term Memory) model is employed to learn the time series data of power plants and predict future power generation fluctuations. Experimental results show that the AI model demonstrates superior accuracy in handling sudden power generation fluctuations compared to traditional prediction methods. This research is expected to contribute to the improved operational efficiency of power systems.