PhD Final Dissertation Defense – Tyler Pilet

Date/Time:                                   Thursday, March 3rd 12pm

In-Person Location:                Price Gilbert Library Room 4222 

Virtual Meeting Information:  Join on your computer or mobile app

       Click here to join the meeting

       Or call in (audio only)

       +1 470-705-2566,,970891804#   United States, Atlanta

Committee:

Dr. Tarek Rakha (rakha@design.gatech.edu) Assistant Professor of Architecture (Committee Chair)

Dr. Jason Brown (jbrittonb@gmail.com) Research Engineer, Raven Research

Dr. Russell Gentry (russell.gentry@design.gatech.edu) Associate Professor of Architecture and Civil Engineering (by courtesy)

Cheryn Metzger (cheryn.metzger@pnnl.gov) Residential Program Manager, Pacific Northwest National Laboratory

Dr. Juan Ordóñez (ordonez@eng.famu.fsu.edu) Professor, Mechanical Engineering, Florida State University

Title:

Towards Understanding an Imperfect Built Environment: A Methodology for In-Situ Characterization of Building Envelope Thermal Performance

Abstract:

As buildings age, retrofits are becoming an increasingly important topic for the ever-growing and aging existing building stock.  Following construction, a building's energy footprint typically remains relatively stagnant, effectively locking-in that building's energy usage for its lifetime.  With 50% of America’s building stock built before 1980 and only 0.5–1% of existing buildings retrofitted annually, it is essential to reduce guesswork and make building energy retrofits more accessible to reduce the energy footprint of the building sector.  Building retrofits are plagued by a lack of original design documentation and general uncertainty regarding the building's envelope composition and integrity.  The goal is this work is to utilize the power of transient heat transfer modeling to non-intrusively characterize the thermal properties of a building's envelope to inform energy modeling, facade design, and project appraisal.  This thesis presents a literature survey of the state-of-the-art in in-situ thermal testing, a thermal characterization methodology to non-destructively identify representative thermal properties for existing building envelopes, a simulation-based study to verify the thermal characterization method, two physical experiments to validate the thermal characterization method, and a proof-of-concept machine learning approach to classify in-service assemblies via the proposed thermal characterization methodology.  This dissertation is designed to bridge the gap between the discrete procedures of building audits and building energy modeling processes to enable a better understanding of existing building envelopes and reduce guesswork from envelope retrofits.