Model Predictive Control for Individual Room Control

This paper deals with model predictive control (MPC) for individual room control (IRC). Two models of a selected room were assembled, each with different complexity. The first of them is used as an exact replacement of the real room for a controller verification, the second one is used for a state observer and MPC controller design. Both the models were created using only a building documentation, what can be suitable especially for large buildings with hundreds of rooms. It was shown that MPC controller can be advantageously used for room temperature control. In the next work weather forecast and occupancy estimation could be incorporated into MPC controller design, what can bring energy savings.