Volume 14: Low Carbon Cities and Urban Energy Systems: Part III

Zero energy potential analysis of photovoltaic direct-driven air conditioners based on thermal comfort using machine learning methods Sihui Lia, Jinqing Penga, Bojia Li, Chujie Lu, Yimo Lu, Tao Ma

Abstract

Usually the energy matching between building load and PV generation is rigid for photovoltaic direct-driven air conditioners (PVAC). The utilization of thermal comfort can improve the flexibility of building loads to increase the real-time energy matching for PVACs. This study aims to propose a dynamic zero energy evaluation method considering the thermal comfort temperatures for PVAC. The interaction between the flexible building load and rigid PV generation is investigated using different machine learning models with an one-minute time resolution. The indoor temperatures conditioned by PVAC are simulated under actual operations. Indicators such as hourly self-consumption (SC), hourly self-sufficiency (SS), hourly zero energy time (ZET), and real-time zero energy ratio (RZER) are used to evaluate the dynamic energy performance of PVAC in different seasons. With fixed indoor setting temperature selected from standard, the RZER is only 27.87% in summer for hot-summer and cold-winter zone. While taking the thermal comfort temperature into account, the corresponding RZER for PVACs reaches 51.31% with 100% SC of PV generation. Moreover, zero energy points always appear at times of large cooling demand, which can reduce the burden on the utility grid. An optimization for PV capacity is also conducted and is found that an increase of PV capacity helps to raise RZER but results in the excessive energy output. The real-time zero energy evaluation method with indoor comfort taking into account is useful for evaluating the zero-energy potential and designing more flexible PVAC systems.

Keywords zero energy building, photovoltaic direct-driven air conditioners, machine learning, energy matching