David Cheng and Ying Tan
Data-driven optimisation techniques in building management systems
IESTEC/University of Melbourne
The presentation will review how various data-driven optimisation techniques can be used in HVAC applications to minimise energy consumption and improve the performance of a building management system. Cheng and Tan will show how to process large amounts of BMS and external data, with the help of techniques such as machine learning and extremum seeking control.
Three case studies will be highlighted. The first will look at predictive models for cooling demand and their optimisation; the second will discuss clustering for sub-meters and energy profiles; and the third will focus on data-driven PID tuning.
Cheng is heavily involved in many large-scale building projects in Australia and overseas, with more than 15 years of engineering experience in the BMS industry. In 2014, he co-founded IESTEC – a company pioneering the use of data mining technology and AI into automation industries. Currently, he is IESTEC’s R&D leader.
Dr Tan is an Associate Professor and Reader in the Department of Mechanical Engineering at the University of Melbourne. She received her bachelor’s degree from China’s Tianjin University and her PhD from the National University of Singapore. Since 2004, she has been with the University of Melbourne. Her research interests are in intelligent systems, nonlinear control systems, real time optimisation, sampled-data distributed parameter systems and formation control.