A novel machine learning based identification of potential adopter of rooftop solar photovoltaics

Bhavsar, S.; Pitchumani, R.

doi:10.1016/j.apenergy.2021.116503

A novel machine learning based identification of potential adopter of rooftop solar photovoltaics

Journal Article · Fri Jan 22 23:00:00 EST 2021 · Applied Energy

DOI:https://doi.org/10.1016/j.apenergy.2021.116503· OSTI ID:1848545

Bhavsar, S. ^[1]; Pitchumani, R. ^[2]

Virginia Polytechnic Inst. and State Univ. (Virginia Tech), Blacksburg, VA (United States). Advanced Materials and Technologies Lab.; Virginia Polytechnic Inst. and State Univ. (Virginia Tech), Blacksburg, VA (United States)
Virginia Polytechnic Inst. and State Univ. (Virginia Tech), Blacksburg, VA (United States). Advanced Materials and Technologies Lab.

With the proliferation of rooftop solar photovoltaic installations, there is a need to proactively predict consumer potential for solar photovoltaic adoption, for improved electric utility planning and operation. Traditional analytical modeling approaches are limited to a few survey features and a larger part of the survey would remain untouched by the decision model. This article presents a novel, data-driven modeling approach that strategically prunes a large set of consumer profile features using a machine learning framework to train a model for predicting potential solar adoption. The approach utilizes the Gradient Boosting Decision Tree model through a Light Gradient Boosting framework that improves significantly over the poor prediction accuracy of the existing approaches. Model training using focal-loss based supervision is used to overcome the difficulty in identifying the potential adopters that is inherent in conventional data-driven models. In addition, to overcome possible data sparsity in a limited survey sample, a Generative Adversarial Network is presented to create synthetic user samples and its effectiveness on model performance is assessed. A Bayesian optimization approach is used to systematically arrive at the hyperparameters of the proposed model. Validation of the presented approach on a survey data collected by the National Rural Electric Cooperative Association in Virginia in 2018 demonstrates the excellent predictive capability of the machine learning based approach to modeling solar adoption reliably.

Research Organization:: University of Virginia, Charlottesville, VA (United States)

Sponsoring Organization:: USDOE Office of Energy Efficiency and Renewable Energy (EERE)

Grant/Contract Number:: EE0007660

OSTI ID:: 1848545

Alternate ID(s):: OSTI ID: 23187433
OSTI ID: 1922533

Journal Information:: Applied Energy, Journal Name: Applied Energy Journal Issue: C Vol. 286; ISSN 0306-2619

Publisher:: ElsevierCopyright Statement

Country of Publication:: United States

Language:: English

References (10)

Economics of household technology adoption in developing countries: Evidence from solar technology adoption in rural India Aklin, M.; Bayer, P.; Harish, S. P. Energy Economics, Vol. 72 https://doi.org/10.1016/j.eneco.2018.02.011	journal	May 2018
Consumer attitudes towards domestic solar power systems Faiers, Adam; Neame, Charles Energy Policy, Vol. 34, Issue 14 https://doi.org/10.1016/j.enpol.2005.01.001	journal	September 2006
Agent-based modeling of energy technology adoption: Empirical integration of social, behavioral, economic, and environmental factors Rai, Varun; Robinson, Scott A. Environmental Modelling & Software, Vol. 70 https://doi.org/10.1016/j.envsoft.2015.04.014	journal	August 2015
An experimental comparison of classification algorithms for imbalanced credit scoring data sets Brown, Iain; Mues, Christophe Expert Systems with Applications, Vol. 39, Issue 3 https://doi.org/10.1016/j.eswa.2011.09.033	journal	February 2012
Motivations and barriers associated with adopting microgeneration energy technologies in the UK Balcombe, Paul; Rigby, Dan; Azapagic, Adisa Renewable and Sustainable Energy Reviews, Vol. 22 https://doi.org/10.1016/j.rser.2013.02.012	journal	June 2013
Spatial patterns of solar photovoltaic system adoption: The influence of neighbors and the built environment Graziano, M.; Gillingham, K. Journal of Economic Geography, Vol. 15, Issue 4 https://doi.org/10.1093/jeg/lbu036	journal	October 2014
Erratum: Estimating mutual information [Phys. Rev. E 69 , 066138 (2004)] Kraskov, Alexander; Stögbauer, Harald; Grassberger, Peter Physical Review E, Vol. 83, Issue 1 https://doi.org/10.1103/PhysRevE.83.019903	journal	January 2011
Learning from Imbalanced Data No authors listed IEEE Transactions on Knowledge and Data Engineering, Vol. 21, Issue 9 https://doi.org/10.1109/TKDE.2008.239	journal	September 2009
machine. Friedman, Jerome H. The Annals of Statistics, Vol. 29, Issue 5 https://doi.org/10.1214/aos/1013203451	journal	October 2001
A Non-Modeling Exploration of Residential Solar Photovoltaic (PV) Adoption and Non-Adoption Moezzi, Mithra; Ingle, Aaron; Lutzenhiser, Loren https://doi.org/10.2172/1379469	report	September 2017

Similar Records

Data-driven agent-based modeling, with application to rooftop solar adoption

Journal Article · Sun Jan 24 23:00:00 EST 2016 · Autonomous Agents and Multi-Agent Systems · OSTI ID:1248850

Coupled social and infrastructure approaches for enhancing solar energy adoption. Final Report

Technical Report · Thu Mar 18 00:00:00 EDT 2021 · OSTI ID:1771998

AN AUTOMATED MACHINE LEARNING-GENETIC ALGORITHM FRAMEWORK WITH ACTIVE LEARNING FOR DESIGN OPTIMIZATION

Conference · Sun Aug 01 00:00:00 EDT 2021 · OSTI ID:1855957

Related Subjects

14 SOLAR ENERGY
42 ENGINEERING
Bayesian optimization
Binary classification
Energy & Fuels
Engineering
Generative Adversarial Network
Machine learning
Rooftop solar adoption

A novel machine learning based identification of potential adopter of rooftop solar photovoltaics

Citation Formats

References (10)

Similar Records

Related Subjects