On Machine Learning-Based Techniques for Future Sustainable and Resilient Energy Systems

Jiawei Wang, Pierre Pinson, Spyros Chatzivasileiadis, Mathaios Panteli, Goran Strbac, Vladimir Terzija

Research output: Contribution to journalArticlepeer-review

Abstract

Permanently increasing penetration of converter-interfaced generation and renewable energy sources (RESs) makes modern electrical power systems more vulnerable to low probability and high impact events, such as extreme weather, which could lead to severe contingencies, even blackouts. These contingencies can be further propagated to neighboring energy systems over coupling components/technologies and consequently negatively influence the entire multi-energy system (MES) (such as gas, heating and electricity) operation and its resilience. In recent years, machine learning-based techniques (MLBTs) have been intensively applied to solve various power system problems, including system planning, or security and reliability assessment. This paper aims to review MES resilience quantification methods and the application of MLBTs to assess the resilience level of future sustainable energy systems. The open research questions are identified and discussed, whereas the future research directions are identified.

Original languageEnglish
Pages (from-to)1-15
Number of pages15
JournalIEEE Transactions on Sustainable Energy
DOIs
Publication statusAccepted/In press - 2022

Keywords

  • Extreme events
  • machine learning
  • multi-energy systems
  • Power system dynamics
  • Power system reliability
  • Power system stability
  • Power systems
  • Resilience
  • resilience
  • Security
  • Stability analysis
  • sustainable energy systems

Fingerprint

Dive into the research topics of 'On Machine Learning-Based Techniques for Future Sustainable and Resilient Energy Systems'. Together they form a unique fingerprint.

Cite this