MétaCan
Menu
Back to cohort
Record W6996958485

Supervision et prédiction des défauts des transformateurs électriques en utilisant les techniques de machine learning

2024· other· fr· W6996958485 on OpenAlexaff

Bibliographic record

VenueDepositum (Université du Québec en Abitibi-Témiscamingue) · 2024
Typeother
Languagefr
Field
Topic
Canadian institutionsCégep de l'Abitibi TémiscamingueUniversité du Québec en Abitibi-Témiscamingue
Fundersnot available
KeywordsWork (physics)Filter (signal processing)LimitingContext (archaeology)
DOInot available

Abstract

fetched live from OpenAlex

RÉSUMÉ Ce rapport présente en détail le mémoire de recherche intitulé "Supervision et prédiction des défauts des transformateurs électriques via les techniques de Machine Learning". L'ambition de ce travail est de déployer une stratégie innovante fondée sur l'apprentissage automatique pour anticiper les pannes des transformateurs électriques, garantissant ainsi leur optimal fonctionnement. Dans le cadre de cette recherche, notre approche s'est articulée autour de deux axes principaux. D'un côté, nous avons exploité des séries de données issues de simulations obtenues grâce à un modèle électrique que nous avons conçu, afin de modéliser les fluctuations de courant et de tension des transformateurs. Ces données de simulation ont été essentiel non seulement pour anticiper les pannes potentielles, mais également pour déterminer l'emplacement des irrégularités au sein des transformateurs. De l'autre côté, nous avons bénéficié de données réelles issues d'indicateurs de suivi mis à notre disposition, ce qui a ajouté une dimension pratique et concrète à notre recherche. Ces données, nous ont permis d'aborder la prédiction des défaillances des transformateurs sous un angle novateur. L'utilisation de techniques avancées d'apprentissage automatique et d'apprentissage en profondeur pour analyser ces ensembles de données s'est révélée cruciale pour prédire avec précision les défauts et les anomalies dans les transformateurs électriques. Pour cette recherche, nous avons recours à Matlab/Simulink pour élaborer un modèle électrique fidèle des transformateurs. Par ailleurs, l’implémentation des modèles d'apprentissage automatique s'est appuyée sur l'utilisation de Python, un langage largement privilégié dans le domaine de la science des données. ABSTRACT This document elaborates on the research thesis titled "Machine Learning Approaches for Monitoring and Predicting Faults in Electrical Transformers." The core objective of this study is to leverage Machine Learning for early detection and prediction of faults within electrical transformers, ensuring their efficient functioning. Our research methodology is bifurcated into two distinct segments. Initially, we utilized simulated datasets derived from an electrical model we developed, aiming to replicate the current and voltage variations observed in transformers. These simulations were pivotal for forecasting failures and pinpointing anomalies' locations. Furthermore, we incorporated actual operational data, gathered from monitoring systems, into our analysis, thereby grounding our investigation in real-world scenarios. This integration of simulated and empirical data facilitated a novel perspective on fault prediction in transformers. The employment of sophisticated machine learning and deep learning methodologies to dissect these datasets was instrumental in accurately identifying electrical system faults. To construct a detailed electrical model of the transformers, we employed Matlab/Simulink. Additionally, the development of our machine learning models was conducted using Python, the preferred programming language in data science, enhancing our data analysis capabilities and enabling the selection of appropriate learning algorithms tailored to our research requirements.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.003
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.008
Threshold uncertainty score0.015

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0010.003
Meta-epidemiology (narrow)0.0010.000
Meta-epidemiology (broad)0.0010.001
Bibliometrics0.0010.001
Science and technology studies0.0000.000
Scholarly communication0.0010.001
Open science0.0000.000
Research integrity0.0010.001
Insufficient payload (model declined to judge)0.0030.001

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.006
GPT teacher head0.212
Teacher spread0.205 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2024
Admission routes1
Has abstractyes

Explore more

Same venueDepositum (Université du Québec en Abitibi-Témiscamingue)French-language works237,207