Advanced Control for Hammerstein-bilinear HVAC System

ABSTRACT

The COVID-19 pandemic has influenced many aspects of human life, including working environments. Some research finds that there is a tendency to the increase of energy and CO2 emissions of large office buildings in developed countries, such as US and Europe's top five economics, post-pandemic. Therefore, advanced heating, ventilation and air-conditioning (HVAC) technology that can reduce energy consumption in the building sector will yield a significant impact on the total national energy consumption. Many buildings equipped with conventional control in their HVAC control systems, such as PI or PID controls. Such controllers have drawbacks like unable to handle cross-coupling nature and constraints in a HVAC system. Conversely, model predictive control (MPC) - which belongs to advanced control - has the advantages when dealing with the system with constraints and uncertainties as it can take into account them in its optimization control problem formulation. This paper derived mathematically an industrial HVAC system based on Hammerstein-bilinear model - a model consists of a static nonlinearity followed by a dynamic bilinear subsystem. The obtained linear output-error (OE) models are subsequently used as plant models in the MPC design. The MPC controller performance is quite superior and proven to be able to meet the desired control objective (keeping the zone temperature in range of . In addition, the MPC controller gives more economic energy consumption (about save) than the PI one both for temperature and humidity control loop. © 2022 ACM.