TY - JOUR T1 - Data fusion in predicting internal heat gains for office buildings through a deep learning approach JF - Applied Energy Y1 - 2019 A1 - Zhe Wang A1 - Tianzhen Hong A1 - Mary Ann Piette KW - data fusion KW - deep learning KW - Internal heat gains KW - Miscellaneous electric loads KW - Occupant count KW - Predictive control AB -

Heating, Ventilation, and Air Conditioning (HVAC) is a major energy consumer in buildings. The predictive control has demonstrated a potential to reduce HVAC energy use. To facilitate predictive HVAC control, internal heat gains prediction is required. In this study, we applied Long Short-Term Memory Networks, a special form of deep neural network, to predict miscellaneous electric loads, lighting loads, occupant counts and internal heat gains in two United States office buildings. Compared with the predetermined schedules used in American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE) standard 90.1, the Long Short-Term Memory Networks method could reduce the prediction errors of internal heat gains from 12% to 8% in Building A, and from 26% to 16% in Building B. It was also found that for internal heat gains prediction, miscellaneous electric loads is a more important feature than occupant counts for two reasons. First, miscellaneous electric loads is the best proxy variable for internal heat gains, as it is the major component of and has the highest correlation coefficient with the internal heat gains. Second, miscellaneous electric loads contain valuable information to predict occupant count, while occupant count could not help improve miscellaneous electric loads prediction. These findings could help researchers and practitioners select the most relevant features to more accurately predict internal heat gains for the implementation of predictive HVAC control in buildings.

VL - 240 UR - https://linkinghub.elsevier.com/retrieve/pii/S0306261919303630 JO - Applied Energy ER - TY - JOUR T1 - DEEP: A Database of Energy Efficiency Performance to Accelerate Energy Retrofitting of Commercial Buildings Y1 - 2015 A1 - Sang Hoon Lee A1 - Tianzhen Hong A1 - Geof Sawaya A1 - Yixing Chen A1 - Mary Ann Piette AB -

The paper presents a method and process to establish a database of energy efficiency performance (DEEP) to enable quick and accurate assessment of energy retrofit of commercial buildings. DEEP was compiled from results of about 35 million EnergyPlus simulations. DEEP provides energy savings for screening and evaluation of retrofit measures targeting the small and medium-sized office and retail buildings in California. The prototype building models are developed for a comprehensive assessment of building energy performance based on DOE commercial reference buildings and the California DEER prototype buildings. The prototype buildings represent seven building types across six vintages of constructions and 16 California climate zones. DEEP uses these prototypes to evaluate energy performance of about 100 energy conservation measures covering envelope, lighting, heating, ventilation, air-conditioning, plug-loads, and domestic hot water. DEEP consists the energy simulation results for individual retrofit measures as well as packages of measures to consider interactive effects between multiple measures. The large scale EnergyPlus simulations are being conducted on the super computers at the National Energy Research Scientific Computing Center of Lawrence Berkeley National Laboratory. The pre-simulation database is a part of an on-going project to develop a web-based retrofit toolkit for small and medium-sized commercial buildings in California, which provides real-time energy retrofit feedback by querying DEEP with recommended measures, estimated energy savings and financial payback period based on users’ decision criteria of maximizing energy savings, energy cost savings, carbon reduction, or payback of investment. The pre-simulated database and associated comprehensive measure analysis enhances the ability to performance assessments of retrofits to reduce energy use for small and medium buildings and business owners who typically do not have resources to conduct costly building energy audit. DEEP will be migrated into the DEnCity - DOE’s Energy City, which integrates large-scale energy data for multi-purpose, open, and dynamic database leveraging diverse source of existing simulation data.

U2 - LBNL-180309 ER - TY - Generic T1 - Dynamic Controls for Energy Efficiency and Demand Response: Framework Concepts and a New Construction Case Study in New York T2 - 2006 ACEEE Summer Study on Energy Efficiency in Buildings Y1 - 2006 A1 - Sila Kiliccote A1 - Mary Ann Piette A1 - David S. Watson A1 - Glenn D. Hughes JF - 2006 ACEEE Summer Study on Energy Efficiency in Buildings CY - Pacific Grove, CA, USA ER - TY - Generic T1 - Demand Relief and Weather Sensitivity in Large California Commercial Buildings T2 - International Conference for Enhancing Building Operations Y1 - 2001 A1 - Satkartar Kinney A1 - Mary Ann Piette A1 - Lixing Gu A1 - Philip Haves AB -

A great deal of research has examined the weather sensitivity of energy consumption in commercial buildings; however, the recent power crisis in California has given greater importance to peak demand. Several new loadshedding programs have been implemented or are under consideration. Historically, the target customers have been large industrial users who can reduce the equivalent load of several large office buildings. While the individual load reduction from an individual office building may be less significant, there is ample opportunity for load reduction in this area. The load reduction programs and incentives for industrial customers may not be suitable for commercial building owners. In particular, industrial customers are likely to have little variation in load from day to day. Thus a robust baseline accounting for weather variability is required to provide building owners with realistic targets that will encourage them to participate in load shedding programs.

JF - International Conference for Enhancing Building Operations CY - Austin, TX ER -