Nonlinear controller design methods for modern building heating and ventilation systems using inverse dynamics

This paper presents the control of a HVAC system using a nonlinear, multivariable controller design based on a Nonlinear Dynamic Inverse method. The proposed benefits of this design compared to a traditional PID controller are shown through the improved decoupling of the interactions between the heating and ventilation systems, better disturbance rejection and tighter setpoint tracking for humidity and temperature and reduced over-driving of actuator systems resulting in less energy usage. The robustness of the controller when faced with varying degrees of sensor lag is investigated and a method of reducing this sensitivity by dynamically varying the controller’s response is developed using a nonlinear error actuated controller. A model of a HVAC controlled indoor zone is presented and the proposed controller’s performance is compared to that of a PI design. The simulation results demonstrate the improved comfort and energy saving potential of the proposed advanced controller design.