Deep learning-based power baseline modelling of a range of electrical loads in smart green building

Abstract

Energy efficiency is important for both commercial and non-commercial applications. Power consumption modeling of electrical loads plays a crucial role in modern-day energy management systems of green buildings. This research focuses on developing a power baseline model using deep learning techniques for a diverse range of electrical loads and the behavior of individual electrical loads will be taken into consideration. The baseline refers to the normal average power consumption of a particular electrical load measured over a designated time frame under specific operating conditions. The power baseline model gives a reference to identify energy saving or loss since it refers to a benchmark for power consumption. It helps identify device abnormalities and implement troubleshooting steps or energy management techniques to optimize energy usage. Energy optimization not only leads to a reduction in environmental impact but also contributes to financial savings. In the context of this research, the baseline for average power consumption of Individual Air Condition (AC) Systems is determined based on a 30-minute or hourly average, depending on which yields the smallest error when averaging according to the type of AC. The baseline is obtained using a deep learning model, which considers weather conditions and other relevant parameters.