A Decision Support Framework for Building Energy Efficiency and Productivity Based on a Sensor Network

Abstract

The buildings sector accounts for 40% of global energy consumption and over 30% of carbon emissions, with HVAC (heating, ventilation, and air-conditioning) systems responsible for more than 51% of building energy use. To address the challenge of reducing HVAC energy consumption while maintaining thermal comfort and indoor air quality (IAQ), this study proposes an enhanced HVAC system model for air-water systems, integrating advanced optimization techniques. The model incorporates real-time data from occupancy and IAQ sensors, including occupancy levels, temperature, CO2, and indoor air pollutant concentrations. Using this data, the system dynamically adjusts control variables such as airflow and fan speeds to optimize energy efficiency and maintain IAQ. To manage uncertainties in occupancy and environmental conditions, a robust optimization framework is developed based on a deterministic HVAC model. Adjustable robust optimization enhances system adaptability, ensuring consistent performance under varying real-world conditions. The proposed framework is validated through EnergyPlus, a high-fidelity building energy simulation platform simulating real-world HVAC operations. This research demonstrates a scalable solution for energy-efficient HVAC management. It achieves significant energy savings while maintaining essential comfort and air quality standards, contributing to the broader goal of sustainable building operations.