Modeling Systems' Disruption and Social Acceptance - A Proof-of-Concept Leveraging Reinforcement Learning

Research output: Contribution to journalArticlepeer-review

Abstract

As the need for a just and equitable energy transition accelerates, disruptive clean energy technologies are becoming more visible to the public. Clean energy technologies, such as solar photovoltaics and wind power, can substantially contribute to a more sustainable world and have been around for decades. However, the fast pace at which they are projected to be deployed in the United States (US) and the world poses numerous technical and nontechnical challenges, such as in terms of their integration into the electricity grid, public opposition and competition for land use. For instance, as more land-based wind turbines are built across the US, contention risks may become more acute. This article presents a methodology based on reinforcement learning (RL) that minimizes contention risks and maximizes renewable energy production during siting decisions. As a proof-of-concept, the methodology is tested on a case study of wind turbine siting in Illinois during the 2022–2035 period. Results show that using RL halves potential delays due to contention compared to a random decision process. This approach could be further developed to study the acceptance of offshore wind projects or other clean energy technologies.

Original languageAmerican English
Article number10231
Number of pages13
JournalSustainability (Switzerland)
Volume15
Issue number13
DOIs
StatePublished - 2023

Bibliographical note

Publisher Copyright:
© 2023 by the authors.

NREL Publication Number

  • NREL/JA-6A20-86039

Keywords

  • energy transition
  • reinforcement learning
  • social acceptance
  • wind turbines

Fingerprint

Dive into the research topics of 'Modeling Systems' Disruption and Social Acceptance - A Proof-of-Concept Leveraging Reinforcement Learning'. Together they form a unique fingerprint.

Cite this