Experimental Study on Control-oriented Simulation Models for Building Control and Energy Management

Heuristic rule-based control (RBC) is widely used in the building industry for building automation systems (BASs). Recently, researchers demonstrated that model-based control has the potential to improve the building energy efficiency. Model Predictive Control (MPC) is one of the promising approaches that caught much attention of the research community. The system dynamics model play an important role in model-based control, but constructing the model can be challenging, especially under realistic engineering conditions. In this paper, we experiment a hybrid approach to construct building thermal dynamics model for integrated, advanced control, under the guide of control theory. The approach for developing this control-oriented model is different from that for a first principle simulation model, such as a Trnsys model. Our model development method is tightly coupled with the controller development, and is intended to reduce development efforts. We designed an MPC controller using a technique called feedback linearization. We tested the design of our model and controller using a Trnsys simulation model for Intelligent Workspace (IW) at Carnegie Mellon University (CMU).