Abstract

Within residential electricity consumption there exists significant variability from home-to-home due to the differences in building thermal properties, appliances, and inhabitants. Electricity analyses at sub-city scales using predefined geographies, such census tracts, might artificially split areas with homogenous characteristics leading to analyses that don't effectively contrast the drivers of energy use. The objective of this study is to use the spatial relationships between demographics, building types, and electricity consumption to form new geographies with less variability for use in residential energy assessment. Using Los Angeles and New York City as case studies, differences in energy use variability within predefined geographies (e.g., census tract) are compared to geographies defined by clustering on socio-technical characteristics. Socio-technical clustering, regardless of the chosen subset of variables, reduces the energy consumption variability over pre-defined geopolitical boundaries with high statistical significance (p<<0.0001). By clustering, intra-regional variability is reduced by 13% in Los Angeles and 29% in New York, therefore improving opportunities for prediction and forecasting. This is the first study to examine the role of spatial boundaries in urban energy assessment. The creation of socio-technical geographies for electricity assessment creates opportunities for improving predictions and forecasts for future sub- and cross-city energy studies.

Original languageEnglish (US)
Pages (from-to)742-754
Number of pages13
JournalEnergy
Volume112
DOIs
StatePublished - Oct 1 2016

Keywords

  • Buildings
  • Clustering
  • Electricity
  • Geospatial
  • Residential
  • Urban

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Building and Construction
  • Pollution
  • Mechanical Engineering
  • Industrial and Manufacturing Engineering
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Defining geographical boundaries with social and technical variables to improve urban energy assessments'. Together they form a unique fingerprint.

Cite this