New paper presents method to scale up insights on building operation
© Miha Creative - stock.adobe.com
Our new paper in Energy and Buildings, led by ETHOS PhD student Yufei Zhang, intrdoduces a new framework for inferring building operation characteristics from widely available smart meter data. The framework, OccuEMBED, simulateously infers occupancy levels and analyzes system-level load information for improved flexibility and operation at the urban scale.
Buildings account for a significant share of global energy consumption and emissions, making it critical to operate them efficiently. As electricity grids become more volatile with increasing renewable penetration, buildings must provide flexibility to support grid stability. Building automation plays a key role in enhancing efficiency and flexibility via centralized operations, but its implementation must prioritize occupant-centric strategies to balance energy and comfort targets. However, incorporating occupant information into large-scale, centralized building operations remains challenging due to data limitations.
We investigate the potential of using whole-building smart meter data to infer both occupancy and system operations. Integrating these insights into data-driven building energy analysis may enable more occupant-centric energy-saving and flexibility at scale. Our proposed solution is OccuEMBED, a unified framework for simultaneous occupancy inference and system-level load analysis. It combines two key components: a probabilistic occupancy profile generator, and a controllable and interpretable load disaggregator supported by Kolmogorov-Arnold Networks (KAN). This design embeds prior knowledge of occupancy patterns and load-occupancy-weather relationships into deep learning models. We conducted comprehensive performance evaluations to demonstrate its effectiveness across synthetic and real-world datasets compared to various occupancy inference baselines. OccuEMBED always achieved average F1 scores above 0.8 in discrete occupancy inference and RMSE within 0.1–0.2 for continuous occupancy ratios. We further demonstrate how OccuEMBED integrates with building load monitoring platforms to display occupancy profiles, analyze system-level operations, and inform occupant-responsive strategies.
OccuEMBED lays a robust foundation in scaling intelligent and occupant-centric building management systems to meet the challenges of an evolving energy system.
Funding for this work was provided by the ETH Joint Initiative UrbanTwin and EPFL.