Low-Cost Digitalization for Sustainable Energy Management: A Hybrid Approach for Predicting Occupant Energy Consumption in Different Indoor Layout Configurations

Abstract

Accurately predicting occupant energy consumption is crucial for optimizing energy management and promoting sustainability in buildings. However, the availability of reliable stochastic data on occupant energy consumption poses significant challenges including cost and privacy considerations. This research proposes a low-cost digitalization approach that integrates Agent-Based Modeling (ABM), System Dynamics (SD), Building Information Modeling (BIM), and Machine Learning (ML) techniques to predict energy consumption in various indoor layout configurations, while prioritizing affordability, privacy, and sustainability. To investigate the effectiveness of the approach, three different geometric shapes representing indoor layouts (rectangular, square, and compound) were selected from a residential building in Chittagong, Bangladesh. An ABM-SD-BIM model was constructed, incorporating occupant data, layout information, and environmental data. The model-generated energy consumption data was calibrated using three months of real data collected from customized sensors installed within the indoor layouts. The calibrated model accurately generated data for the entire summer month period from April to September, which was used to develop the ML prediction model for each layout. Results indicate that Layout 3 (Compound) exhibits consistent and favorable performance across the models, suggesting its suitability for accurate energy consumption prediction. The proposed approach prioritizes affordability, privacy, and sustainability, making it accessible and applicable to various building design, operation, and policy-making scenarios. This research contributes to the broader field of low-cost digitalization, emphasizing the importance of accurate energy prediction for optimizing energy management and promoting sustainable practices.

Presenters

Mohammad Nyme Uddin
Post-Doctoral Fellow, Building and Real Estate (BRE), The Hong Kong Polytechnic University, Hong Kong

Xue Cui

Xuange Zhang

Minhyun Lee

Details

Presentation Type

Paper Presentation in a Themed Session

Theme

Architectonic, Spatial, and Environmental Design

KEYWORDS

OCCUPANT BEHAVIOR,ENERGY CONSUMPTION,LOW-COST DIGITALIZATION,HYBRID-MODEL,INDOOR LAYOUT