Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

Building Energy Modeling for Green Architecture and Intelligent Dashboard Applications

DeBlois, Justin C. (2014) Building Energy Modeling for Green Architecture and Intelligent Dashboard Applications. Doctoral Dissertation, University of Pittsburgh. (Unpublished)

Primary Text

Download (5MB) | Preview


Buildings are responsible for 40% of the carbon emissions in the United States. Energy efficiency in this sector is key to reducing overall greenhouse gas emissions. This work studied the passive technique called the roof solar chimney for reducing the cooling load in homes architecturally. Three models of the chimney were created: a zonal building energy model, computational fluid dynamics model, and numerical analytic model. The study estimated the error introduced to the building energy model (BEM) through key assumptions, and then used a sensitivity analysis to examine the impact on the model outputs. The conclusion was that the error in the building energy model is small enough to use it for building simulation reliably. Further studies simulated the roof solar chimney in a whole building, integrated into one side of the roof. Comparisons were made between high and low efficiency constructions, and three ventilation strategies. The results showed that in four US climates, the roof solar chimney results in significant cooling load energy savings of up to 90%. After developing this new method for the small scale representation of a passive architecture technique in BEM, the study expanded the scope to address a fundamental issue in modeling - the implementation of the uncertainty from and improvement of occupant behavior. This is believed to be one of the weakest links in both accurate modeling and proper, energy efficient building operation. A calibrated model of the Mascaro Center for Sustainable Innovation’s LEED Gold, 3,400 m2 building was created. Then algorithms were developed for integration to the building’s dashboard application that show the occupant the energy savings for a variety of behaviors in real time. An approach using neural networks to act on real-time building automation system data was found to be the most accurate and efficient way to predict the current energy savings for each scenario. A stochastic study examined the impact of the representation of unpredictable occupancy patterns on model results. Combined, these studies inform modelers and researchers on frameworks for simulating holistically designed architecture and improving the interaction between models and building occupants, in residential and commercial settings.


Social Networking:
Share |


Item Type: University of Pittsburgh ETD
Status: Unpublished
CreatorsEmailPitt UsernameORCID
DeBlois, Justin C.jcd34@pitt.eduJCD34
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Committee ChairSchaefer, Laura A.las149@pitt.eduLAS149
Committee MemberBilec, Melissa M.mbilec@engr.pitt.eduMBILEC
Committee MemberChyu, Minking K.mkchyu@pitt.eduMKCHYU
Committee MemberJones, Alex K.akjones@pitt.eduAKJONES
Committee MemberKimber, Markmlk53@pitt.eduMLK53
Date: 29 January 2014
Date Type: Publication
Defense Date: 11 November 2011
Approval Date: 29 January 2014
Submission Date: 26 November 2013
Access Restriction: No restriction; Release the ETD for access worldwide immediately.
Number of Pages: 152
Institution: University of Pittsburgh
Schools and Programs: Swanson School of Engineering > Mechanical Engineering
Degree: PhD - Doctor of Philosophy
Thesis Type: Doctoral Dissertation
Refereed: Yes
Uncontrolled Keywords: Building Energy Modeling Natural Ventilation Solar Chimney Passive Cooling Energy Efficiency Building Occupancy Building Dashboard Model Calibration Occupant Behavior
Date Deposited: 29 Jan 2014 18:38
Last Modified: 15 Nov 2016 14:15


Monthly Views for the past 3 years

Plum Analytics

Actions (login required)

View Item View Item