Demand Response in PV Dominated Active Buildings

Abstract

This project aims to develop an accurate power forecast of a photovoltaic (PV) system for active buildings, using machine learning (ML) models. The objective is to create a demand response system that regulates energy usage in active buildings by providing recommendations for efficient power usage to the user. In this project, we have collected historical data on environmental factors such as solar irradiance and compared various ML models to select the best one based on its accuracy. The selected ML model is implemented to generate the power forecast of the PV system. To create the demand response system, a mobile application is developed that uses the power forecast to provide recommendations for efficient power usage to the user. Our results show that the ML model accurately predicts the power output of the PV system and that the demand response system based on the power forecast can help to manage energy consumption in active buildings. This project demonstrates the potential of ML-based power forecasting and demand response systems in achieving sustainable and energy-efficient buildings.