Development of multi-zone temperature control algorithm using input-output feedback linearization

An input-output feedback control of multi-zone indoor space is investigated. The system consists of two environmental zones, a circulating fan-coil unit and ductwork arrangement for each zone, a boiler, and two heat pumps. This system is represented by a seventh order bilinear system model of the multi-zone space heating system. The bilinear model of the system is linearized through the use of the input-output feedback linearization decoupling method, where only the dominant states of the system are considered. A feedback control law is also designed based on the dominant states of the system model. It is shown that the input-output equations are only feedback linearizable if the output variables are chosen appropriately. The system is acted upon by constant unknown (diurnal temperature variation) disturbance and step changes in the desired set points of temperature. The control objective is to maintain the two zones, the boiler, and the temperatures of the condenser of the two heat pumps close to some chosen set points. Results are presented showing the robustness, sensitivity analysis and decentralized characteristics of the closed-loop system. A special case study for energy consumption is also illustrated. From the viewpoint of zone air temperature control, the system performance with feedback control is shown to be good and acceptable.