A model-predictive controller for air handling units

ABSTRACT Heating and cooling for thermal comfort are the main consumers of energy in buildings, and there is a growing need to improve the energy efficiency (and thereby reduce CO 2 emissions) of these building services. The regular increase in energy tariffs only exacerbates the problem. Building owners are seldom willing to invest in a deep retrofit that may lower their energy consumption, but are instead willing to replace their outdated HVAC systems. Indeed, off-the-shelf controllers are often based on (only) the outdoor temperature, and occasionally take into account the indoor temperature. In particular, practically no commercial systems take into account weather forecasts. Consequently, these control systems lead to poor comfort and sub-optimal energy efficiency. In this paper, a novel model-predictive control (MPC) algorithm for fan coil units (FCU) is presented, which aims at reducing the operational costs while guaranteeing thermal comfort. It is planned to be deployed on a test site in Greece within the second half of 2015. The simulation results are presented and compared to a standard PI controller. For the MPC based controller, the trade-off between the user comfort and the energy consumption of the will be presented and commented. Simulations have demonstrated energy savings of up to 57% compared with the reference controller. Results from field tests are expected by the end of 2015.