A Hybrid Machine Learning Approach for Predicting Power Transformer Failures Using Internet of Things-Based Monitoring and Explainable Artificial Intelligence

Power transformers are critical components in ensuring the continuous and stable operation of power systems. Failures in these units can lead to significant power outages and costly downtime. Existing maintenance strategies often fail to accurately predict such failures, highlighting the need for no...

Full description

Saved in:
Bibliographic Details
Main Authors: Emrah Aslan, Yildirim Ozupak, Feyyaz Alpsalaz, Zakaria M. S. Elbarbary
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11053795/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849744505943621632
author Emrah Aslan
Yildirim Ozupak
Feyyaz Alpsalaz
Zakaria M. S. Elbarbary
author_facet Emrah Aslan
Yildirim Ozupak
Feyyaz Alpsalaz
Zakaria M. S. Elbarbary
author_sort Emrah Aslan
collection DOAJ
description Power transformers are critical components in ensuring the continuous and stable operation of power systems. Failures in these units can lead to significant power outages and costly downtime. Existing maintenance strategies often fail to accurately predict such failures, highlighting the need for novel predictive approaches. This study aims to improve the reliability of power systems by predicting transformer failures through the integration of IoT technologies and advanced machine learning techniques. The proposed hybrid model combines the LightGBM algorithm with GridSearch optimization to achieve both high predictive accuracy and computational efficiency. In addition, the model enhances interpretability by incorporating SHapley Additive exPlanations (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME) for transparent decision making. The study presents a detailed comparison of different classification algorithms and evaluates their performance using metrics such as accuracy, recall, and F1 score. The results show that the hybrid model outperforms other methods, achieving an accuracy of 99.91%. The SHAP and LIME analyses provide engineers and researchers with valuable insights by highlighting the most influential features in failure prediction. In addition, the model’s ability to efficiently handle large data sets enhances its practicality in real-world power systems. By proposing an innovative approach to failure prediction, this research contributes to both the theoretical foundation and practical advancement of sustainable and reliable energy infrastructures.
format Article
id doaj-art-d4281d4bc81e46d28435c9a48bbe6d3f
institution DOAJ
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-d4281d4bc81e46d28435c9a48bbe6d3f2025-08-20T03:15:35ZengIEEEIEEE Access2169-35362025-01-011311361811363310.1109/ACCESS.2025.358377311053795A Hybrid Machine Learning Approach for Predicting Power Transformer Failures Using Internet of Things-Based Monitoring and Explainable Artificial IntelligenceEmrah Aslan0https://orcid.org/0000-0002-0181-3658Yildirim Ozupak1https://orcid.org/0000-0001-8461-8702Feyyaz Alpsalaz2https://orcid.org/0000-0002-7695-6426Zakaria M. S. Elbarbary3https://orcid.org/0000-0003-1750-9244Department of Computer Engineering, Faculty of Engineering and Architecture, Mardin Artuklu University, Mardin, TürkiyeDepartment of Electricity and Energy, Silvan Vocational School, Dicle University, Diyarbakır, TürkiyeDepartment of Electricity and Energy, Akdagmadeni Vocational School, Yozgat Bozok University, Yozgat, TürkiyeElectrical Engineering Department, Faculty of Engineering, King Khalid University, Abha, Saudi ArabiaPower transformers are critical components in ensuring the continuous and stable operation of power systems. Failures in these units can lead to significant power outages and costly downtime. Existing maintenance strategies often fail to accurately predict such failures, highlighting the need for novel predictive approaches. This study aims to improve the reliability of power systems by predicting transformer failures through the integration of IoT technologies and advanced machine learning techniques. The proposed hybrid model combines the LightGBM algorithm with GridSearch optimization to achieve both high predictive accuracy and computational efficiency. In addition, the model enhances interpretability by incorporating SHapley Additive exPlanations (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME) for transparent decision making. The study presents a detailed comparison of different classification algorithms and evaluates their performance using metrics such as accuracy, recall, and F1 score. The results show that the hybrid model outperforms other methods, achieving an accuracy of 99.91%. The SHAP and LIME analyses provide engineers and researchers with valuable insights by highlighting the most influential features in failure prediction. In addition, the model’s ability to efficiently handle large data sets enhances its practicality in real-world power systems. By proposing an innovative approach to failure prediction, this research contributes to both the theoretical foundation and practical advancement of sustainable and reliable energy infrastructures.https://ieeexplore.ieee.org/document/11053795/Fault detectionpower transformersmachine learningSHAPLIME
spellingShingle Emrah Aslan
Yildirim Ozupak
Feyyaz Alpsalaz
Zakaria M. S. Elbarbary
A Hybrid Machine Learning Approach for Predicting Power Transformer Failures Using Internet of Things-Based Monitoring and Explainable Artificial Intelligence
IEEE Access
Fault detection
power transformers
machine learning
SHAP
LIME
title A Hybrid Machine Learning Approach for Predicting Power Transformer Failures Using Internet of Things-Based Monitoring and Explainable Artificial Intelligence
title_full A Hybrid Machine Learning Approach for Predicting Power Transformer Failures Using Internet of Things-Based Monitoring and Explainable Artificial Intelligence
title_fullStr A Hybrid Machine Learning Approach for Predicting Power Transformer Failures Using Internet of Things-Based Monitoring and Explainable Artificial Intelligence
title_full_unstemmed A Hybrid Machine Learning Approach for Predicting Power Transformer Failures Using Internet of Things-Based Monitoring and Explainable Artificial Intelligence
title_short A Hybrid Machine Learning Approach for Predicting Power Transformer Failures Using Internet of Things-Based Monitoring and Explainable Artificial Intelligence
title_sort hybrid machine learning approach for predicting power transformer failures using internet of things based monitoring and explainable artificial intelligence
topic Fault detection
power transformers
machine learning
SHAP
LIME
url https://ieeexplore.ieee.org/document/11053795/
work_keys_str_mv AT emrahaslan ahybridmachinelearningapproachforpredictingpowertransformerfailuresusinginternetofthingsbasedmonitoringandexplainableartificialintelligence
AT yildirimozupak ahybridmachinelearningapproachforpredictingpowertransformerfailuresusinginternetofthingsbasedmonitoringandexplainableartificialintelligence
AT feyyazalpsalaz ahybridmachinelearningapproachforpredictingpowertransformerfailuresusinginternetofthingsbasedmonitoringandexplainableartificialintelligence
AT zakariamselbarbary ahybridmachinelearningapproachforpredictingpowertransformerfailuresusinginternetofthingsbasedmonitoringandexplainableartificialintelligence
AT emrahaslan hybridmachinelearningapproachforpredictingpowertransformerfailuresusinginternetofthingsbasedmonitoringandexplainableartificialintelligence
AT yildirimozupak hybridmachinelearningapproachforpredictingpowertransformerfailuresusinginternetofthingsbasedmonitoringandexplainableartificialintelligence
AT feyyazalpsalaz hybridmachinelearningapproachforpredictingpowertransformerfailuresusinginternetofthingsbasedmonitoringandexplainableartificialintelligence
AT zakariamselbarbary hybridmachinelearningapproachforpredictingpowertransformerfailuresusinginternetofthingsbasedmonitoringandexplainableartificialintelligence