Model predictive control of building HVAC system employing zone thermal energy requests

Control in buildings has been a subject of research interest in the control community for some time. Various control methods have shown a potential for a significant savings in the building operation costs, whereas a large economic gain in the operation of a heating, ventilation and air conditioning (HVAC) system can be obtained by employing information about the building thermal model and the model of actuators, weather conditions, energy demand cost as well as the energy requests in the zones. This paper proposes a model predictive controller for a building chiller that exploits respective information to minimise the cost of cooling in the electricity market with volatile electrical energy prices, while ensuring comfort within the zones and respecting the power demand limitations. Obtained optimal control problem is nonlinear and the minimisation is performed by employing the successive linear programming algorithm within the feasibility region and the gradient algorithm for finding the initial feasible point. A case study HVAC system model is used to validate the performance of the proposed controller in the simulation scenario. Obtained controller minimises the cost of cooling while adhering to the imposed comfort constraints.