skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Big data analytics in the building industry

Abstract

Catalyzed by recent market, technology, and policy trends, energy data collection in the building industry is becoming more widespread. This wealth of information allows more data-driven decision-making by designers, commissioning agents, facilities staff, and energy service providers during the course of building design, operation and retrofit. The U.S. Department of Energy’s Building Performance Database (BPD) has taken advantage of this wealth of building asset- and energy-related data by collecting, cleansing, and standardizing data from across the U.S. on over 870,00 buildings, and is designed to support building benchmarking, energy efficiency project design, and buildings-related policy development with real-world data. Here, this article explores the promises and perils energy professionals are faced with when leveraging such tools, presenting example analyses for commercial and residential buildings, highlighting potential issues, and discussing solutions and best practices that will enable designers, operators and commissioning agents to make the most of ‘big data’ resources such as the BPD.

Authors:
 [1];  [1];  [1]
  1. Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
Publication Date:
Research Org.:
Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States). Energy Analysis & Environmental Impacts Division
Sponsoring Org.:
USDOE Office of Energy Efficiency and Renewable Energy (EERE), Building Technologies Office (EE-5B)
OSTI Identifier:
1344787
Report Number(s):
LBNL-1005983
Journal ID: ISSN 0001-2491; ir:1005983
Resource Type:
Journal Article: Accepted Manuscript
Journal Name:
ASHRAE Journal
Additional Journal Information:
Journal Volume: 58; Journal Issue: 7; Journal ID: ISSN 0001-2491
Publisher:
American Society of Heating, Refrigerating and Air-Conditioning Engineers
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING; 32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATION; big data; building energy use

Citation Formats

Berger, Michael A., Mathew, Paul A., and Walter, Travis. Big data analytics in the building industry. United States: N. p., 2016. Web.
Berger, Michael A., Mathew, Paul A., & Walter, Travis. Big data analytics in the building industry. United States.
Berger, Michael A., Mathew, Paul A., and Walter, Travis. Fri . "Big data analytics in the building industry". United States. doi:. https://www.osti.gov/servlets/purl/1344787.
@article{osti_1344787,
title = {Big data analytics in the building industry},
author = {Berger, Michael A. and Mathew, Paul A. and Walter, Travis},
abstractNote = {Catalyzed by recent market, technology, and policy trends, energy data collection in the building industry is becoming more widespread. This wealth of information allows more data-driven decision-making by designers, commissioning agents, facilities staff, and energy service providers during the course of building design, operation and retrofit. The U.S. Department of Energy’s Building Performance Database (BPD) has taken advantage of this wealth of building asset- and energy-related data by collecting, cleansing, and standardizing data from across the U.S. on over 870,00 buildings, and is designed to support building benchmarking, energy efficiency project design, and buildings-related policy development with real-world data. Here, this article explores the promises and perils energy professionals are faced with when leveraging such tools, presenting example analyses for commercial and residential buildings, highlighting potential issues, and discussing solutions and best practices that will enable designers, operators and commissioning agents to make the most of ‘big data’ resources such as the BPD.},
doi = {},
journal = {ASHRAE Journal},
number = 7,
volume = 58,
place = {United States},
year = {Fri Jul 01 00:00:00 EDT 2016},
month = {Fri Jul 01 00:00:00 EDT 2016}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record
The DOI is not currently available

Save / Share:
  • Rapid increases in high performance computing are feeding the development of larger and more complex data sets in climate research, which sets the stage for so-called big data analysis challenges. However, conventional climate analysis techniques are inadequate in dealing with the complexities of today s data. In this paper, we describe and demonstrate a visual analytics system, called the Exploratory Data analysis ENvironment (EDEN), with specific application to the analysis of complex earth system simulation data sets. EDEN represents the type of interactive visual analysis tools that are necessary to transform data into insight, thereby improving critical comprehension of earthmore » system processes. In addition to providing an overview of EDEN, we describe real-world studies using both point ensembles and global Community Land Model Version 4 (CLM4) simulations.« less
  • In this work, we develop a full information capture approach for Magnetic Force Microscopy (MFM), referred to as generalized mode (G-Mode) MFM. G-Mode MFM acquires and stores the full data stream from the photodetector, captured at sampling rates approaching the intrinsic photodiode limit. The data can be subsequently compressed, denoised, and analyzed, without information loss. Here, G-Mode MFM is implemented and compared to the traditional heterodyne-based MFM on model systems, including domain structures in ferromagnetic Yttrium Iron Garnet and the electronically and magnetically inhomogeneous high entropy alloy, CoFeMnNiSn. We investigate the use of information theory to mine the G-Mode MFMmore » data and demonstrate its usefulness for extracting information which may be hidden in traditional MFM modes, including signatures of nonlinearities and mode-coupling phenomena. Finally, we demonstrate detection and separation of magnetic and electrostatic tip-sample interactions from a single G-Mode image, by analyzing the entire frequency response of the cantilever. G-Mode MFM is immediately implementable on any atomic force microscopy platform and as such is expected to be a useful technique for probing spatiotemporal cantilever dynamics and mapping material properties, as well as their mutual interactions.« less
  • Cited by 2
  • We develop a full information capture approach for Magnetic Force Microscopy (MFM), referred to as generalized mode (G-Mode) MFM. G-Mode MFM acquires and stores the full data stream from the photodetector at sampling rates approaching the intrinsic photodiode limit. The data can be subsequently compressed, denoised, and analyzed, without information loss. Also, 3 G-Mode MFM is implemented and compared to traditional heterodyne based MFM on model systems including domain structures in ferromagnetic Yttrium Iron Garnet (YIG) and electronically and magnetically inhomogeneous high entropy alloy, CoFeMnNiSn. We investigate the use of information theory to mine the G-Mode MFM data and demonstratemore » its usefulness for extracting information which may be hidden in traditional MFM modes, including signatures of nonlinearities and mode coupling phenomena. Finally we demonstrate detection and separation of magnetic and electrostatic tip-sample interactions from a single G-Mode image, by analyzing the entire frequency response of the cantilever. G-Mode MFM is immediately implementable on any AFM platform and as such is expected to be a useful technique for probing spatiotemporal cantilever dynamics and mapping material properties as well as their mutual interactions.« less
  • Electron microscopy is undergoing a transition; from the model of producing only a few micrographs, through the current state where many images and spectra can be digitally recorded, to a new mode where very large volumes of data (movies, ptychographic and multi-dimensional series) can be rapidly obtained. In this paper, we discuss the application of so-called “big-data” methods to high dimensional microscopy data, using unsupervised multivariate statistical techniques, in order to explore salient image features in a specific example of BiFeO 3 domains. Remarkably, k-means clustering reveals domain differentiation despite the fact that the algorithm is purely statistical in naturemore » and does not require any prior information regarding the material, any coexisting phases, or any differentiating structures. While this is a somewhat trivial case, this example signifies the extraction of useful physical and structural information without any prior bias regarding the sample or the instrumental modality. Further interpretation of these types of results may still require human intervention. Finally, however, the open nature of this algorithm and its wide availability, enable broad collaborations and exploratory work necessary to enable efficient data analysis in electron microscopy.« less