Machine Learning Based Smart Energy Management for Residential Application in Grid Connected System

Performance improvement of residential load management is studied with the implementation of machine learning based smart energy manager (SEM). The solar energy sources and electric vehicle technology is incorporated in grid interconnected system which decreases the total power usage in housing application. Here case study of sulur village is considered and energy storage is integrated for improved consistency of the grid system. It is recommended to schedule the peak loads during non-peak hours using machine learning based demand side management (DSM) algorithm. The necessary appliances are switched by the controller, rendering the solar power prediction. This is done in MATLAB, and the simulation model is tested by varying the grid and solar panel restrictions on a regular basis.