This study aims to optimize battery energy storage systems by using artificial intelligence techniques to predict energy consumption and production from renewable sources. We propose a strategy for identifying appropriate intervals for battery storage. Then, we employed two artificial intelligence techniques, namely Extreme Gradient Boosting (XGBoost) and Long Short-Term Memory (LSTM), to make predictions on the data set. The XGBoost technique demonstrated better performance. In addition, by identifying the best and worst days for predicting energy consumption and energy generation from renewable sources, the risk associated with battery storage can be mitigated.