Model predictive control for building automation : an experimental study

• Main Author: Uthaya Kumar, Nithiakuhan
• Other Authors: Wan Man Pun
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2020
• Subjects: Engineering::Mechanical engineering
• Online Access: https://hdl.handle.net/10356/141212
• Institution: Nanyang Technological University

Over the next decade, it is projected that more cities in the ASEAN region will be urbanized. This in turn will cause the number of commercial buildings such as hospitals, factories and commercial spaces to increase. Increasing global temperatures coupled with increasing societal awareness about the need to be energy efficient will drive the need for efficient ACMV systems to be utilized in order to bring comfort to occupants within indoor spaces. An effective way to reducing energy consumption due to ACMV systems is to implement a new control logic to existing building automation systems. Model predictive control is gaining popularity in this aspect as it has forward visibility on thermal comfort parameters and is able to update the myriad of setpoints in an ACMV system in real time. Model predictive control’s optimization algorithm coupled with machine learning of the building’s dynamics is seen as an effective solution to reduce energy consumption whilst still maintaining the thermal comfort of occupants in a building.