Machine Learning in Smart Building: Opportunities and Future Challenges
This presentation showcase Zixiao’s current and past research outcomes, with an overarching theme envisioning the use of machine learning for building design and operation. Topics such as model reduction, automated fault detection and diagnostics, data cleaning and data synthetization through data driven approaches are covered. While recent development of machine learning provides great opportunities to improve our buildings, wider adoption of ‘smarter’ buildings’ faces many practical challenges, particularly due to the aging building stock and status quo. The presenter would like to take this opportunity to discuss such challenges and their potential solutions with fellow researchers at LBNL.
Zixiao (Shawn) Shi
Research Associate, National Research Council Canada, Ottawa Ontario Adjunct Professor, Carleton University, Ottawa Ontario
Zixiao (Shawn) Shi is a research associate at the National Research Council Canada. He studied Architectural Engineering at Purdue University for his Bachelor and Master. In 2018, he finished his Ph.D. at Carleton University under the supervision of professor Liam O’Brien. His research interest lies in adopting advanced machine learning techniques to improve building design and operation. His experience in engineering consulting also gave him some perspective in real-life applications. He has authored over 25 peer reviewed articles, and is an active member of the ASHRAE and IBPSA.