A Structured Literature Review and Meta-analysis of Forecasting Methods for Energy Consumption in Smart Buildings

The infrastructure, network, and accessibility of any the nation's strength are highly associated with its revenue growth since energy has become a crucial part of daily living in today's world. The largest consumers of electricity in the planet are buildings. Buildings have undergone significant advancements as new technologies have been developed, transforming traditional structures into smart buildings. As a consequence, there has been a tremendous increase in the demand of energy for residential and commercial usage all over the world. These smart buildings use the energy management strategies to control the energy resources within the building for significant energy savings. The ability to alter building operations to use energy more effectively makes the forecast of energy use in buildings a crucial component of these energy plans. The present research attempted to conduct a rigorous and critical assessment of around 81 previous studies in the field of forecasting energy consumption that were published in scholarly journals over a fourteen-year period (2010–2023). Different energy consumption forecasting methods in different forecasting horizons are studied for residential and non-residential buildings. The quantitative analysis of 55 studies is then examined and results show that 84% of the work is done using short-term-based time horizon to forecast building energy which is largest among all time horizons. The study further showed that 75% of research studies focused on (Historical and Calendar) input parameters for commercial building energy forecasting and ignored the occupancy variable in their studies. The research concludes by looking at possible future challenges to forecasting building energy usage.