Symbolic Regressor: An Interpretability Tool for Non-intrusive Load Monitoring

  • Danel Rey-Arnal*
  • , Pablo G. Bringas
  • , Ibai Laña
  • *Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Multiple sources of worries such as economic constraints and the dangers of climate change have moved society towards the process of optimizing the use of their electricity. However this approach towards energy consumption has become a source of uncertainty and worry as load monitoring becomes the norm. In order to overcome the privacy concerns techniques on Non-Intrusive Load Monitoring have been in development since the 1980s. In the field of load disaggregation applications of NILM there is constant reference to three topics to be improved on, results, interpretability and responsiveness. This paper investigates the role symbolic regression tools in the field of NILM, both as a singular tool of disaggregation and as a support instrument of deep learning models more common in the literature, such as LSTM, to improve on their prediction capabilities and adding a layer of interpretability to the results. The experimentation of this document offer two different solutions with various degrees of success depending on the proposed scenario although with quantifiable improvement over the established baseline.

Original languageEnglish
Title of host publicationHybrid Artificial Intelligent Systems - 20th International Conference, HAIS 2025, Proceedings
EditorsEmilio Corchado, Héctor Quintián, Alicia Troncoso Lora, Hilde Pérez García, Esteban Jove Pérez, José Luis Calvo Rolle, Francisco Javier Martínez de Pisón, Pablo García Bringas, Francisco Martínez Álvarez, Álvaro Herrero, Paolo Fosci, Ramos Sérgio Filipe
PublisherSpringer Science and Business Media Deutschland GmbH
Pages260-271
Number of pages12
ISBN (Print)9783032084644
DOIs
Publication statusPublished - 2026
Event20th International Conference on Hybrid Artificial Intelligence Systems, HAIS 2025 - Salamanca, Spain
Duration: 16 Oct 202517 Oct 2025

Publication series

NameLecture Notes in Computer Science
Volume16202 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference20th International Conference on Hybrid Artificial Intelligence Systems, HAIS 2025
Country/TerritorySpain
CitySalamanca
Period16/10/2517/10/25

Keywords

  • load disaggregation
  • LSTM
  • NILM
  • Symbolic Regression

Fingerprint

Dive into the research topics of 'Symbolic Regressor: An Interpretability Tool for Non-intrusive Load Monitoring'. Together they form a unique fingerprint.

Cite this