Data-Driven Methods for Building Energy Consumption Efficiency under Climate Change

Volume Title: ICASGE2025

Paper ID : 1021-ICASGE-FULL (R1)

¹The American University in Cairo

²McMaster University

Climate change amplifies the frequency and severity of natural disasters which include hydrological, climatological and meteorological hazards. These hazards adversely impact the performance of infrastructure systems. Building heating and cooling activities have been identified as key contributors to greenhouse gas emissions, and subsequently climate change. In this context, finding ways to decrease energy consumed by buildings heating and cooling can dramatically decrease the overall building energy consumption which has been a pressing global need in recent years. Subsequently, this study aims to predict building heating and cooling load requirements based on different design parameters in an effort to optimize these loads. In this respect, simulation of building energy consumption performance was used to obtain heating and cooling load dataset based on such parameters. As such, several data analytics techniques were used to predict heating load requirements for buildings under eight specific design parameters. These techniques include multilinear regression, classification and clustering. This work presents a major step towards achieving more sustainable and energy efficient buildings that contribute to decreasing greenhouse gas emissions without compromising functionality and users’ comfort.

Climate Change; Energy Efficiency; Machine Laerning; Data Analysis; Building Design; Heating Load; Cooling Load

Construction Management (COM)