Energy Optimal Control of Heating Ventilation and Air Conditioning (HVAC) Systems at CERN
With the ever growing demand for energy and escalating energy prices, the next generation of control solutions must integrate real-time process control with process economic optimization. This talk will demonstrate the advantages of advanced process control techniques in terms of energy savings, safety and stability over conventional control solutions. We will discuss the design of Model Predictive Control (MPC) strategy with a focus on enabling the energy optimal operation of heating ventilation and air conditioning systems (HVAC) at CERN. The main idea behind MPC is to utilize a process model to predict future process behavior and minimize a given performance index subject to different physical and operational constraints. In HVAC systems, the inherent ability of the MPC to account for complex non-linear interactions among different components and the inclusion of weather predictions can significantly improve the performance and energy consumption over conventional PID-based solutions. Some recent results on the development of non-linear and hybrid MPC for air handling units and cooling towers will be discussed. A comparison will be provided with the currently deployed solutions highlighting the potential for energy savings and control performance improvements across a wide range of operating conditions.