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

Title: Generating realistic building electrical load profiles through the Generative Adversarial Network (GAN)

Abstract

Building electrical load profiles can improve understanding of building energy efficiency, demand flexibility, and building-grid interactions. Current approaches to generating load profiles are time-consuming and not capable of reflecting the dynamic and stochastic behaviors of real buildings; some approaches also trigger data privacy concerns. In this study, we proposed a novel approach for generating realistic electrical load profiles of buildings through the Generative Adversarial Network (GAN), a machine learning technique that is capable of revealing an unknown probability distribution purely from data. The proposed approach has three main steps: (1) normalizing the daily 24-hour load profiles, (2) clustering the daily load profiles with the k-means algorithm, and (3) using GAN to generate daily load profiles for each cluster. The approach was tested with an open-source database – the Building Data Genome Project. We validated the proposed method by comparing the mean, standard deviation, and distribution of key parameters of the generated load profiles with those of the real ones. The KL divergence of the generated and real load profiles are within 0.3 for majority of parameters and clusters. Additionally, results showed the load profiles generated by GAN can capture not only the general trend but also the random variations ofmore » the actual electrical loads in buildings. We report the proposed GAN approach can be used to generate building electrical load profiles, verify other load profile generation models, detect changes to load profiles, and more importantly, anonymize smart meter data for sharing, to support research and applications of grid-interactive efficient buildings.« less

Authors:
 [1];  [1]
  1. Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
Publication Date:
Research Org.:
Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
Sponsoring Org.:
USDOE Office of Energy Efficiency and Renewable Energy (EERE), Energy Efficiency Office. Building Technologies Office
OSTI Identifier:
1784288
Grant/Contract Number:  
AC02-05CH11231
Resource Type:
Journal Article: Accepted Manuscript
Journal Name:
Energy and Buildings
Additional Journal Information:
Journal Volume: 224; Journal ID: ISSN 0378-7788
Publisher:
Elsevier
Country of Publication:
United States
Language:
English
Subject:
32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATION; Electrical load profile; Generative Adversarial Network (GAN); Machine learning; Smart meter data; Load shape

Citation Formats

Wang, Zhe, and Hong, Tianzhen. Generating realistic building electrical load profiles through the Generative Adversarial Network (GAN). United States: N. p., 2020. Web. doi:10.1016/j.enbuild.2020.110299.
Wang, Zhe, & Hong, Tianzhen. Generating realistic building electrical load profiles through the Generative Adversarial Network (GAN). United States. https://doi.org/10.1016/j.enbuild.2020.110299
Wang, Zhe, and Hong, Tianzhen. 2020. "Generating realistic building electrical load profiles through the Generative Adversarial Network (GAN)". United States. https://doi.org/10.1016/j.enbuild.2020.110299. https://www.osti.gov/servlets/purl/1784288.
@article{osti_1784288,
title = {Generating realistic building electrical load profiles through the Generative Adversarial Network (GAN)},
author = {Wang, Zhe and Hong, Tianzhen},
abstractNote = {Building electrical load profiles can improve understanding of building energy efficiency, demand flexibility, and building-grid interactions. Current approaches to generating load profiles are time-consuming and not capable of reflecting the dynamic and stochastic behaviors of real buildings; some approaches also trigger data privacy concerns. In this study, we proposed a novel approach for generating realistic electrical load profiles of buildings through the Generative Adversarial Network (GAN), a machine learning technique that is capable of revealing an unknown probability distribution purely from data. The proposed approach has three main steps: (1) normalizing the daily 24-hour load profiles, (2) clustering the daily load profiles with the k-means algorithm, and (3) using GAN to generate daily load profiles for each cluster. The approach was tested with an open-source database – the Building Data Genome Project. We validated the proposed method by comparing the mean, standard deviation, and distribution of key parameters of the generated load profiles with those of the real ones. The KL divergence of the generated and real load profiles are within 0.3 for majority of parameters and clusters. Additionally, results showed the load profiles generated by GAN can capture not only the general trend but also the random variations of the actual electrical loads in buildings. We report the proposed GAN approach can be used to generate building electrical load profiles, verify other load profile generation models, detect changes to load profiles, and more importantly, anonymize smart meter data for sharing, to support research and applications of grid-interactive efficient buildings.},
doi = {10.1016/j.enbuild.2020.110299},
url = {https://www.osti.gov/biblio/1784288}, journal = {Energy and Buildings},
issn = {0378-7788},
number = ,
volume = 224,
place = {United States},
year = {2020},
month = {7}
}

Works referenced in this record:

Impacts of Raw Data Temporal Resolution Using Selected Clustering Methods on Residential Electricity Load Profiles
journal, November 2015


Review of Smart Meter Data Analytics: Applications, Methodologies, and Challenges
journal, May 2019


Electric load shape benchmarking for small- and medium-sized commercial buildings
journal, October 2017


Modeling of end-use energy consumption in the residential sector: A review of modeling techniques
journal, October 2009


Real-time retail price determination in smart grid from real-time load profiles: JOSEPH AND E. A.
journal, January 2018


Silhouettes: A graphical aid to the interpretation and validation of cluster analysis
journal, November 1987


Residential Energy Load Profile Generation Using a Probabilistic Approach
conference, November 2012

  • Gruber, Jorn K.; Prodanovic, Milan
  • 2012 European Modelling Symposium (EMS), 2012 Sixth UKSim/AMSS European Symposium on Computer Modeling and Simulation
  • https://doi.org/10.1109/EMS.2012.30

Mutual Privacy Preserving $k$ -Means Clustering in Social Participatory Sensing
journal, August 2017


Sparse and Redundant Representation-Based Smart Meter Data Compression and Pattern Extraction
journal, May 2017


Price sensitivity of residential energy consumption in Norway
journal, December 1999


CITIESData: a smart city data management framework
journal, April 2017


Private memoirs of a smart meter
conference, January 2010

  • Molina-Markham, Andrés; Shenoy, Prashant; Fu, Kevin
  • Proceedings of the 2nd ACM Workshop on Embedded Sensing Systems for Energy-Efficiency in Building - BuildSys '10
  • https://doi.org/10.1145/1878431.1878446

A New Index and Classification Approach for Load Pattern Analysis of Large Electricity Customers
journal, February 2012


Multi-Layered Clustering for Power Consumption Profiling in Smart Grids
journal, January 2017


Residential Load Shape Modelling Based on Customer Behavior
journal, July 1985


Privacy-Enhanced Data Aggregation Scheme Against Internal Attackers in Smart Grid
journal, February 2014


Dynamic clustering segmentation applied to load profiles of energy consumption from Spanish customers
journal, February 2014


Buildings.Occupants : a Modelica package for modelling occupant behaviour in buildings
journal, November 2018


Determination of household energy using ‘fingerprints’ from energy billing data
journal, October 1986


Energy Big Data Analytics and Security: Challenges and Opportunities
journal, September 2016


A method of formulating energy load profile for domestic buildings in the UK
journal, June 2005


Analysis of a privacy-preserving PCA algorithm using random matrix theory
conference, December 2016