Machine Learning Approaches To Predict The Stability of Smart Grid

Abstract

The smart grid is a modern solution to generate, distribute, and use energy effectively and efficiently. Ensuring the stability of the smart grid is critical to guarantee safe and consistent operation. This study proposes a machine learning model to identify smart grid stability more efficiently. To accomplish this, we collected the electrical grid stability dataset from a public machine learning repository and analyzed the correlation of individual features. Different feature selection and transformation techniques were employed to generate subsequent feature sets. Then, various classification techniques were applied to the primary and its subsets to identify the best machine-learning model for predicting smart grid stability more efficiently. The results indicated that MLP showed the highest accuracy (97.06%) for the Z-score transformed data set using Grid Search Optimization. This study demonstrates the potential of using machine learning to predict smart grid stability accurately and efficiently, providing a critical tool for improving the reliability, efficiency, and sustainability of the energy industry.