Dynamic Temporal Analysis and Modeling of Residential Lighting Consumption for Energy Efficiency and Sustainability

Buildings constitute a significant portion of global energy demand, accounting for approximately 30% of final energy consumption and over 50% of global electricity usage. As per the International Energy Agency, this sector is responsible for 26% of energy-related emissions, with demand projected to increase by 4% in 2024, as per the International Energy Agency. Lighting is a critical component of this consumption and continues to be a major contributor to electricity use. A marked increase in residential lighting energy consumption in South Korea underscores these global trends. Traditional building energy models must address human behavior's complexity, often relying on static schedules. This study leverages statistical and machine learning techniques to analyze lighting energy consumption across varying temporal scales, using real empirical data from South Korean residential buildings. The analysis reveals insights into behavior-driven energy consumption, highlighting the necessity for dynamic energy modeling that integrates occupant behavioral variability for energy use, their intrinsic routines, and temporal patterns. The findings have profound implications for enhancing energy efficiency and optimizing conservation strategies in escalating global energy demands.