Machine - Learning Based Prediction of Stability of Smart Grid

The term “smart grid” refers to a sophisticated power system concept that combines communication within a system network with electricity. It gives real-time access to diverse data to customers, operators, and manufacturers. The need to distribute this electric power to the consumer domains - houses, industries, businesses and smart cities, efficiently is quite big. In order to meet the exact power needs, a smart grid with more stable systems is being used. Stability predictions for smart grids remain difficult. Participation from producers and customers is one of the many factors that determine stability of grid. The smart grid's stability may result from participant identification. This study utilizes a machine learning technique to identify the smart grid's stability. In various studies, popular classifier models such as Support Vector Machine, Logistic Regression, Decision Tree and Bagging Classifier are used to compare the results of the proposed model. With an accuracy of 96%, the proposed model of Light Gradient Boosting Machine (LGBM) performs better than the others.