A physics-based linear parametric model of room temperature in office buildings

This paper presents a physics-based autogregressive moving average (ARMAX) model of room temperature in office buildings. Thermodynamic equations are used to determine the structure and order of the model. Extensive measurements over 109 days are used to develop and validate the model. The model can be used to predict the room temperature variations accurately in both short-term and long-term periods, as long as ten weeks with a mean squared error less than 0.01 and a coefficient of determination larger than 0.99. This model also provides an analytical foundation of the previously proposed spatial and temporal partition strategy of VAVs for fault detection of HVAC systems, and is suitable for real-time fault detection and control applications.

► We study a physics-based autoregressive moving average model of room temperature in office buildings.
► Thermodynamic equations are used to determine the optimal structure of the model.
► The model demonstrates improved prediction compared to existing models in the same domain.
► The model provides an analytical foundation for the previously proposed partition strategy for fault detection of HVAV systems.