Model predictive control on smart building automation control

Abstract

The world is looking towards environmental sustainability. With energy consumption being a concern to environmental sustainability due to carbon emission and depleting natural resources, energy optimization methods need to be studied. Integrating Model Predictive Control (MPC) had shown effective energy optimization in many studies conducted. Thus, the purpose of the study was to see the effectiveness of energy optimization of MPC when it was integrated with BAC system in a digital twin. IES VE was used to create the digital twins of an office. The model had 2 controllers, a conventional controller built on IES platform and the other built with MPC controller. Comparisons on energy consumption of air-conditioning system and thermal comfort of digital twin model between both controllers were made to see the effect of MPC. Energy optimization depends on the amount of energy saved. Then after, simulations were carried out for different settings of MPC, such as energy bias or thermal comfort bias setting. After simulation of different MPC settings, another simulation of low occupancy density was carried out for comparison. Different settings of MPC or low occupancy density were simulated to see difference in energy consumption and thermal comfort. The most ideal MPC setting based on thermal comfort or energy consumption was selected. Through comparison, MPC achieved energy reduction in range of 11% to 17% compared to the conventional controller. The most optimum MPC setting is weighted narrow PMV range for energy savings with thermal comfort.