11　Development of an Urban Surplus Energy Visualization Tool Linking Machine Learning and GIS

Abstract

Photovoltaic (PV) systems are expected to reduce fossil fuel use and carbon dioxide emissions, and provide distributed energy systems and self-sufficiency. However, installing many PV systems has resulted in the amount of energy generated exceeding the amount of energy demanded, which is being dumped as it is. This has led to the development of V2H (Vehicle to Home) technologies that combine Feed-in Tariff (FIT) systems and PV systems with storage battery systems. PV systems are subject to weather conditions and energy supply risks, and many researchers are actively researching the optimal use of PV systems. This study examines efficient ways to use surplus energy from PV systems. In this study, we built a machine learning model to calculate urban surplus energy in Daimyo, Fukuoka City, as a target area, developed a Geographic Information System (GIS) based urban surplus energy visualization tool, and examined efficient ways to use surplus energy.