Non-linear Model Predictive Control for Smart Heating of Buildings

Abstract

Smart and flexible operation of components in district heating systems can play a crucial role in integrating larger shares of renewable energy sources in energy systems. Buildings are one of the crucial components that will enable flexibility in the district heating by using intelligent operation. Recent work suggests that such improved operation at the same time can increase thermal comfort and lower economic costs. We have digitalised the heating system in a Danish school by adding IoT devices, such as smart thermostats and temperature sensors to demonstrate the possibilities of making buildings smart. Based on experimental data, this paper introduces a non-linear grey-box model of the thermal dynamics of the building. A non-linear model predictive control method is presented for the thermostatic set-point control of the building's radiators. Based on the building model and the control algorithm, simulation studies are carried out to show the flexibility potential of the building. When used for lowering the return temperature the results suggest that operational costs can be lowered by around 10% using predictive control.

Hybrid Model Predictive Control Scheme for Controlling Temperature in Building Under Uncertainties

“…The research focused on employing MPC in BEMS has witnessed substantial activity [5], [6]. The application of MPC for regulating indoor tempeature using both active heating systems and passive solar blinds were investigated [7].…”