Computationally Less Intensive Alternative to Model Predictive Control for Energy Efficient Building HVAC System

A framework for the design and simulation of a building envelope and an HVAC system is presented. Building models are first captured in Modelica to leverage its rich building component library and then imported into Simulink to exploit its strong control design environment that enables efficient control design and implementation. Four controllers with different computational intensity are considered and compared: a proportional (P) controller with time varying temperature bounds, a tracking LQR controller with time varying tuning parameters, a tracking d-LQR controller with time varying tuning parameters which incorporates the predictive disturbance information in control derivation and a model predictive controller (MPC). We assess the performance of these controllers using two defined criteria, i.e. energy and discomfort indices. We show that the d-LQR and MPC compared to the P control, manage to reduce the energy index by 41.2% and 46% respectively, and the discomfort index from 3.8 to 0. While d-LQR and MPC have similar performance concerning energy and discomfort index, simulation time in the case of d-LQR is significantly less than that of MPC.