PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
May 28, 2026Electric Power Systems Research0 citationsOpen Access

Causal inference-inspired analysis applied to non-technical loss characterization in distribution power systems

View Full Paper
RBRafael M.R. BarrosaECEdson G. CostaUniversidade Federal de Campina GrandeJAJ. F. AraújoUniversidade Federal de Campina Grande

Key Points

  • This research aims to identify and analyze engineered variables that contribute to non-technical losses in power distribution systems.
  • Causal-inference approach applied to data from 261,489 users
  • Analysis of 174 engineered variables to find correlations with non-technical loss events
  • Focus on interpretability rather than prediction accuracy
  • 76 engineered variables were linked to non-technical loss events
  • Technical, meter-reading, and consumption variables provided the most insight
  • Framework enhances decision-making rather than just improving prediction accuracy

Abstract

• Causal-inference approach characterized non-technical losses from 261,489 users. • From 174 engineered variables, 76 showed links to non-technical loss events. • Technical, meter-reading, and consumption were the most informative variables. • Framework improves interpretability and decisions rather than prediction accuracy.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Barrosa et al. (2026) studied this question.

synapsesocial.com/papers/6a17db293fad632b0f9d800fhttps://doi.org/10.1016/j.epsr.2026.113360
Ask AI
Helpful
Bookmark
Share
View Full Paper