BackgroundLarge heating, ventilation, and air conditioning (HVAC) systems in commercial buildings consume tremendous amounts of energy and have become a prime target for control strategies aimed at energy cost minimization in recent years. Model predictive control (MPC) has been suggested as a promising solution for HVAC control. Many MPC-based control architectures have been proposed for handling large systems.
However, it is difficult to evaluate the novel ideas and decompositions without having a common system against which to benchmark performance. Here, we present a case study problem definition loosely based on an industrial application at Stanford University (depicted above). This case study is made publicly available for other researchers in the HVAC community to provide a standardized problem for the research community to design and test the performance and viability various control systems. |