Jie (Jay) Zhao

PhD | Carnegie Mellon University

Design-Build-Operate Energy Information Modeling for Occupant-Oriented Predictive Building Control

Buildings consume 42% of the total energy in the US. Currently, a building energy simulation model is often created during the building design stage only to demonstrate green building standard compliance, such as LEED. Although creating the energy model is costly, it is often discarded after LEED certification is awarded. This project proposes a Design-Build-Operate Energy Information Modeling (DBO-EIM) infrastructure to extend the use of the energy model to the entire building life-cycle. The DBO-EIM infrastructure updates the design stage energy model and integrates it to an occupant-oriented predictive control system during the building operation stage. The control system can use the energy model to run numerous scenarios and generate a set of optimal control commands in real time to reduce energy use and improve occupant thermal comfort. A newly constructed office building - Phipps Center for Sustainable Landscapes is studied as a test-bed for the project. Click here for more information.

The research is partly funded by National Science Foundation (NSF) Emerging Frontiers in Research and Innovation (EFRI) in Science in Energy and Environmental Design (SEED) (Award#: 1038139) and Department of Energy (DOE) Energy Efficient Building Hub (EEB-HUB)(Subtask 6.4).

Pittsburgh PA USA, 09/2010 - 02/2015

    DBO-EIM Infrastructure

Occupant Behavior And Schedule Prediction Based on Office Appliance Energy Consumption Data Mining

43.5% of the total site energy in commercial buildings is consumed by office appliance and service equipment majorly in the form of electricity. This number is expected to increase to 51.3% by 2015. Comparing to lighting and HVAC energy saving strategies, the technical aspect of reducing plug loads is straightforward, simply by turning off the appliances when they are not used. However, the use of appliance is highly dependent on the occupant behavior, which is hard to learn and predict. The need for understanding the occupant behavior and users' collaboration is essential for reducing plug loads. A machine learning study is performed on the office appliance energy consumption data to predict occupant behavior and schedule. All-in-one wireless electrical outlet meters/switches Plugwise® are installed to meter office appliance consumptions. Fitbit® pedometers are used to record occupants "ground truth". Click here for more information.

This study contributes to International Energy Agency (IEA) Annex 66 - Definition and Simulation of Occupant Behavior in Buildings.

The research is partly funded by National Science Foundation (NSF) Emerging Frontiers in Research and Innovation (EFRI) in Science in Energy and Environmental Design (SEED) (Award#: 1038139) and Department of Energy (DOE) Energy Efficient Building Hub (EEB-HUB)(Subtask 6.4).

Pittsburgh PA USA, 01/2013 - 07/2014

    Occupant Behavior Data Collection Architecture
    Occupant Schedule Data Mining Result

LEED Energy Performance Online Submission Tool (LEPOST)

Leadership in Energy and Environmental Design (LEED) rating system has been well recognized and widely used in the green building industry. In order to reduce the LEED submission cost and facilitate green building design process, a web-based software program - LEED Energy Performance Online Submission Tool (LEPOST) - is developed. LEPOST can automatically map EnergyPlus and eQUEST energy simulation results to LEED energy performance requirement templates. Using Energy Star Target Finder and current Energy Information Administration utility data, LEED Energy and Atmosphere Prerequisite 2 and Credit 1 submission template can be completed and points can be calculated instantly. The tool can reduce the amount of time and effort for LEED documentation and submission process from hours (if not days) to less than 5 minutes.

The research is funded by National Science Foundation (NSF) Emerging Frontiers in Research and Innovation (EFRI) in Science in Energy and Environmental Design (SEED) (Award#: 1038139).

Pittsburgh PA USA, 09/2010 - 07/2014

    LEPOST Result Display Page

LEED Building Market Influential Factor Analysis

LEED buildings were investigated in numeric studies but only used as instances in static matrices. These studies were not able to answer the question why a particular city favors LEED. However, in this sutdy, three commonly used machine learning algorithms - Linear Regression, Locally Weighted Regression and Support Vector Regression (SVR) - are compared and SVR is used to investigate, discover and evaluate the variables that could influence LEED building market in U.S. East Coast cities. Machine learned models are created and optimized with the features of city geography, demography, economy, higher education and policy. SVR model identifies the key factors by dynamically self-training and tuning the model over the dataset. Via optimization, correlation coefficient between the model's prediction and actual value is 0.79. The result suggests that population and policy can be important factors for LEED buildings. It is also interesting that higher educational institutes, especially accredited architecture schools could also be driving forces for LEED commercial building markets in East Coast cities. Click here for more information.

Pittsburgh PA USA, 09/2011 - 02/2012

    Top 10 LEED Building Cities in East Coast

Pro Environment Living - Net Zero Energy Multi-family House Design

Together with Freddie Croce, we designed a three-unit net zero energy multifamily House for DOW Design to Zero Competition. On the basis of environmental and context analysis, we proposed a complete solution for building systems and modeled the house with REM/Rate software for its energy performance. The house achieved EnergyStar V2.5 and IECC2012 Code, and achieved 2 points (0 point is net zero) for the Home Energy Rating Score (HERS). Solar and wind impact was analyzed with Autodesk Revit and Project Vasari. Click here for more information.

Pittsburgh PA USA, 10/2011 - 12/2011

    Pro-Environment Living Schematic Design

Bamboo House - Solar Decathlon Europe 2010

As the group leader for the electrical and lighting systems, I worked on the Bamboo House design-build project at Tongji University for the Solar Decathlon Europe Competition. The Solar Decathlon Competition aims to design and build a net zero energy house by using solar energy. It was organized by Spain government and supported by US Department of Energy. I was in charge of designing and installing the overall electricity distribution system and control system for lighting and electrical systems. The competition was held in Madrid, Spain in June 2010. Click here for more information.

Shanghai China, 09/2009 – 04/2010

    Tongji University Solar Decathlon Europe Bamboo House

Shanghai ECO-House for World Expo 2010

I worked on the Shanghai ECO-House exterior lighting design project as a part-time lighting design consultant for Shanghai Research Center of Engineering and Technology for Solid-State Lighting. Shanghai Eco-House is a 4-story office building in the Urban Best Practice Area of World EXPO 2010 Shanghai China. Advanced LED lighting with Schneider Electric lighting control system is installed in the building. Click here for more information.

Shanghai China, 09/2008 – 10/2009

    World Expo 2010 Shanghai Eco-House