Development and implementation of statistical models for estimating diversified adoption of energy transition technologies

R. Bernards, J. Morren, J.G. Slootweg

Research output: Contribution to journalArticleAcademicpeer-review

8 Citations (Scopus)
181 Downloads (Pure)

Abstract

For efficient network investments, insight in the expected spatial spread of new load and generation units is of prime importance. This paper presents and applies a method to determine key factors for adoption of photovoltaics, electric vehicles and heat pumps. Using a logistic regression analysis the impact of geographical and socio-economic factors on adoption probabilities of these new energy technologies is quantified. Income level, average age and household composition are shown to be important factors. Additionally for photovoltaics, peer effects were also shown to significantly influence the likelihood of adoption. The implementation of the developed models and the achievable improvement in prediction accuracy is demonstrated by application to a scenario study based on historical data. The models can be incorporated in future energy scenarios to provide a probabilistic spatial forecast of future penetration levels of the mentioned technologies and identify key areas of interest.

Original languageEnglish
Article number8260957
Pages (from-to)1540-1554
Number of pages15
JournalIEEE Transactions on Sustainable Energy
Volume9
Issue number4
DOIs
Publication statusPublished - 1 Oct 2018

Keywords

  • Adaptation models
  • Buildings
  • Forecast uncertainty
  • Investment
  • Planning
  • Power system planning
  • Predictive models
  • Probabilistic logic
  • Resistance heating
  • Statistical learning
  • Technology adoption
  • power system planning
  • statistical learning
  • technology adoption

Fingerprint Dive into the research topics of 'Development and implementation of statistical models for estimating diversified adoption of energy transition technologies'. Together they form a unique fingerprint.

Cite this