BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Better early-stage design decisions and advanced building control 
 for climate change mitigation - Dr Hannes Gauch\, University of Cambridge
DTSTART:20221027T110000Z
DTEND:20221027T120000Z
UID:TALK189557@talks.cam.ac.uk
CONTACT:James Walsh
DESCRIPTION:Buildings account for over one-third of global emissions and e
 nergy use. Meeting climate pledges will require conciliating high operatio
 nal energy efficiency and low embodied impacts in new construction. In thi
 s talk I will present some of my past and current work addressing this iss
 ue.\n\nFirst\, I will talk about an early-stage design tool for multi-stor
 ey buildings. It automates the design of structural frames to Eurocode and
  calculates each generated design option’s embodied carbon and cost. Thi
 s permits to compare a wide range of design options in early project stage
 s and to evaluate the impact of design decisions. Two test cases illustrat
 e the potential carbon and cost savings unlocked by well-informed early-st
 age design decisions.\n\nThen I will move on to more recent work which exp
 ands the work on structural frames to include envelopes\, building use\, a
 nd operational efficiency. A global sensitivity analysis is employed to ex
 plore the relative importance of a wide range of design and operation para
 meters in terms of embodied cost and carbon\, as well as heating and cooli
 ng loads.\n\nFinally\, I will talk about some research in progress. We are
  using housing stock databases to automatically create dynamic building en
 ergy models for large parts of the housing stock. We will use these models
  to explore the effect of retrofit options or other energy saving interven
 tions. Further we are investigating the energy and emission savings potent
 ial of employing reinforcement learning to optimise building control.
LOCATION:Teaching Room\, James Dyson Building\, Dept of Engineering\, Univ
 ersity of Cambridge
END:VEVENT
END:VCALENDAR