A Review on Deep Learning and Hybrid Model for Forecasting Residential and Commercial Buildings Energy Consumption

• Published in: 2024 5th International Conference on Artificial Intelligence and Data Sciences (AiDAS) pp. 93 - 98
• Main Authors: Isa, Siti Sarah Mat, Samat, Ahmad Asri Abd, Shamsudin, Nur Hazahsha, Hussain, Mohd Najib Mohd, Isa, Siti Salwa Mat, Omar, Abdul Malek Saidina
• Format: Conference Proceeding
• Language: English
• Published: IEEE 03-09-2024
• Subjects: Buildings; Deep learning; Energy consumption; Forecasting; hybrid model; predicting; Predictive models; Resource management; Reviews; Soft sensors; Urban areas; Weather forecasting
• DOI: 10.1109/AiDAS63860.2024.10729959

The population growth and urbanization have a significant impact on the current rise in electricity demand. Therefore, it is essential to embrace a proactive approach to determine the future energy requirements to consistently meet user needs. The prediction of energy usage within buildings holds great importance in the realm of effective resource management, as it directly impacts both the economy and the environment. The conventional approach, which relied heavily on historical data and economic indicators, is now being replaced by more advanced methodologies like Machine Learning (ML) and Artificial Intelligence (AI). These modern techniques integrate a wider range of data sources such as weather patterns, occupancy, and seasonal variations to enhance the precision of energy consumption forecasts. This paper offers an extensive review of literature pertaining to energy consumption prediction through Deep Learning (DL) and hybrid model, a combination of various forecasting methods applied in real-world situations. The study considered different types of forecasting approaches, two building categories, time frame, findings, and future recommendation. The insights provided in this review are anticipated to guide future research endeavors and identify potential research gaps, particularly in the domain of energy consumption forecasting for buildings.