Intelligent Building Automation and Analytics using Model-Predictive Control (MPC)

Project Key Words:
Model predictive control (MPC), MPC-based methodology for faults analysis and inefficiency diagnosis, building automationsystem
Dr. Wan Man Pun
Organization:Nanyang Technological University
Designation:Associate Professor
Mr. Ridzuan Laili (Schneider)
Project Period:
07/01/2016 To 12/07/2020
Project Description:

Globally, buildings are responsible for around 40% of the final energy use and 30% of CO2 emissions, harbouring enormous potential for energy saving and carbon emission reductions. Conventional building automation and control (BAC) systems that are based on reactive control strategy has limited capability in handling complex building dynamics, multiple service systems as well as the contrasting demands of energy efficiency and occupant’s well-being. The digital revolution offers opportunities to greatly improve the energy efficiency and occupant well-being by enabling the adoption of advanced control techniques and smart solutions for BAC. The project aims at developing a model predictive control (MPC) solution for coordinated control of multiple building service systems, including air conditioning and mechanical ventilation (ACMV), lighting and shading, for optimizing building energy efficiency and occupant’s well-being. MPC can foresee the future response (e.g., room air temperature, RH, thermal comfort) of a building by exploiting a virtual model of the building as well as disturbance information (e.g., outdoor weather, indoor occupancy density, heat release from indoor electric equipment). Based on the foreseeing, MPC solves a cost function to generate future optimal control strategies for building service systems in real-time. The core of the MPC system includes a physics-/machine-learning-based integrated building model, capturing the dynamics of the building, ACMV system, lighting system, shading system, occupant thermal and visual comfort, as the basis for the forward prediction capability. A fast optimisation algorithm is developed to provide real-time integrated control of multiple building service systems with global optimisation. The MPC technology has been implemented in several test buildings, including the BCA SkyLab, lecture theatre and office on NTU campus, as well as office in Ng Teng Feng General Hospital, achieving 20 – 59% energy savings with improved thermal and visual comfort as compared to conventional reactive control in these testbeds. A MPC system that is suitable for commercial deployment is now being developed.