SEMINAR: How advanced statistics and machine learning can improve energy efficiency
The increasing availability of advanced metering infrastructure in buildings has resulted in a massive amount of data that can be used to provide knowledge for a variety of energy efficiency applications such as more accurate energy savings estimation, fault detection and diagnosis, demand response and HVAC system optimization. This large amount of data has also opened the door to the use of advanced statistical and machine learning methods, which holds promise for providing more accurate building energy data analytics. This talk will provide a review of some innovative data driven methods and tools that have been developed at LBNL to produce more accurate estimations of saved energy in commercial buildings energy efficiency projects. It will also describe the challenges that faces the adoption of these advanced analytics. Furthermore, future research directions in machine learning applied to building energy analysis and management will be discussed.
Senior Scientific Engineering Associate, Commercial Building Systems Group, Whole Building Systems Department, Building Technology & Urban Systems Division
Dr. Samir Touzani is a Senior Scientific Engineering Associate within the Building Technology and Urban Systems Division at Lawrence Berkeley National Laboratory. Dr. Touzani's research interests lie at the intersection of Machine Learning, Statistics and Energy Efficiency. Most of his current work focuses on developing data driven methods to solve energy performance analysis problems. Prior to joining LBNL, Dr. Touzani was a Research Scientist at the French Institute of Petroleum (IFP Energies Nouvelles) where he conducted research in the development of statistical learning and uncertainty quantification methods and tools for oil and gas reservoir engineering applications. He received a Master's degree in Physics from Pierre and Marie Curie University and a Ph.D. in Statistics from Joseph Fourier University.