Comparisons of building system modeling approaches for control system design

To design and evaluate advanced controls for buildings , building system models that can show detailed dynamics of feedback control loops are required. The models should also be computationally efficient if they are used for model-based control in real time. However, most building energy simulation programs apply idealized feedback control and steady-state model for HVAC equipment. TRNSYS and Modelica may be applicable to study and design local controllers such as ON-OFF sequencing controller and proportionalintegral controller, which are most commonly employed for feedback control of set-points in local controllers of HVAC equipment. In this paper, results computed by different models for a case study were compared with respect to overall energy consumption, peak power demand, computing time, and short-term dynamics, which are necessary for controls design and verification. To compare the computing time between high and low fidelity models, this study also included reduced order models for the envelope of the same building.