Artificial Intelligence-Based Predictive Maintenance of Microgrids

Arafat, Md Yeasir

Artificial Intelligence-Based Predictive Maintenance of Microgrids

Arafat, Md Yeasir

Permalink

Publication Type:: Thesis
Issue Date:: 2026

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

Adobe PDF

Download thesisAdobe PDF (22.9 MB)

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Arafat, Md Yeasir
dc.date.accessioned	2026-10-27T03:52:05Z
dc.date.available	2026-10-27T03:52:05Z
dc.date.issued	2026
dc.identifier.uri	http://hdl.handle.net/10453/197265
dc.description	University of Technology Sydney. Faculty of Engineering and Information Technology.	en_US.UTF-8
dc.description.abstract	The increasing complexity of microgrids demands sophisticated strategies for improved maintenance and reliable operation. The integration of artificial intelligence (AI) in microgrids allows for the analysis of system performance, anomaly detection, and malfunction identification, enhancing predictive maintenance capabilities and sustainable decision-making. To improve system reliability and performance, accurate fault identification in microgrids is becoming crucial for enhanced predictive maintenance of microgrids. The first study presents advanced frameworks for microgrid predictive maintenance by performing a comprehensive correlative analysis of advanced recurrent neural network (RNN) architectures, i.e., RNNs, Long Short-Term Memory (LSTM), and Gated Recurrent Units (GRUs) for photovoltaic (PV) based dc microgrids. An algorithm is developed to detect faulty operations of inverters which is indispensable for ensuring the reliability and sustainability of distributed energy resources (DERs) in modern microgrid systems. The performances of the developed advanced models have been compared with the conventional RNN-based techniques. Results demonstrate strong predictive capabilities of all, with the proposed advanced LSTM method outperforming RNN, GRU, and the traditional techniques. The second study introduces an intelligent data-driven hybrid fault detection framework for microgrid predictive maintenance utilizing an advanced hybrid deep learning (HDL) architecture, combined with advanced data analysis and a residual-based dynamic threshold technique. An adaptive gradient technique is employed to optimize the loss function of the proposed hybrid algorithm for microgrid predictive maintenance. The results show significant accuracy in fault identification for predicting the maintenance needs within microgrids. This research offers valuable insights for designing and enhancing hybrid algorithms for fault detection in microgrid systems, contributing to advancements in predictive maintenance technology and promoting a resilient and cost-effective microgrid operation. The third study proposes a smart predictive maintenance framework for addressing the critical research problems of accurately classifying DER operational states with maintenance alerts and assessing the annual capacity degradation for enhanced predictive maintenance capabilities of complex industrial microgrid scenario by coupling a novel power-aware adaptive kernel function with a hybrid Wiener-GRU degradation algorithm. The overall accuracy of 99% demonstrates the excellent performance of the proposed framework which can allow early identification of performance degradation, enabling proactive scheduling for the maintenance actions which can ensure enhanced reliability, reduced maintenance costs, and improved operational efficiency in complex microgrid networks.	en_US.UTF-8
dc.format	Thesis (PhD)
dc.language.iso	en_US	en_US.UTF-8
dc.rights	info:eu-repo/semantics/openAccess
dc.rights	The author owns the copyright in this thesis including all reproduction and reuse rights for the work. The work may not be altered without the permission of the copyright owner. Attribution is essential when quoting or paraphrasing from this thesis.
dc.rights	© 2026 Md Yeasir Arafat
dc.rights	au.edu.uts.lib/nph
dc.title	Artificial Intelligence-Based Predictive Maintenance of Microgrids	en_US.UTF-8
dc.type	Thesis
utslib.copyright.status	open_access	*

Abstract:

The increasing complexity of microgrids demands sophisticated strategies for improved maintenance and reliable operation. The integration of artificial intelligence (AI) in microgrids allows for the analysis of system performance, anomaly detection, and malfunction identification, enhancing predictive maintenance capabilities and sustainable decision-making. To improve system reliability and performance, accurate fault identification in microgrids is becoming crucial for enhanced predictive maintenance of microgrids. The first study presents advanced frameworks for microgrid predictive maintenance by performing a comprehensive correlative analysis of advanced recurrent neural network (RNN) architectures, i.e., RNNs, Long Short-Term Memory (LSTM), and Gated Recurrent Units (GRUs) for photovoltaic (PV) based dc microgrids. An algorithm is developed to detect faulty operations of inverters which is indispensable for ensuring the reliability and sustainability of distributed energy resources (DERs) in modern microgrid systems. The performances of the developed advanced models have been compared with the conventional RNN-based techniques. Results demonstrate strong predictive capabilities of all, with the proposed advanced LSTM method outperforming RNN, GRU, and the traditional techniques. The second study introduces an intelligent data-driven hybrid fault detection framework for microgrid predictive maintenance utilizing an advanced hybrid deep learning (HDL) architecture, combined with advanced data analysis and a residual-based dynamic threshold technique. An adaptive gradient technique is employed to optimize the loss function of the proposed hybrid algorithm for microgrid predictive maintenance. The results show significant accuracy in fault identification for predicting the maintenance needs within microgrids. This research offers valuable insights for designing and enhancing hybrid algorithms for fault detection in microgrid systems, contributing to advancements in predictive maintenance technology and promoting a resilient and cost-effective microgrid operation. The third study proposes a smart predictive maintenance framework for addressing the critical research problems of accurately classifying DER operational states with maintenance alerts and assessing the annual capacity degradation for enhanced predictive maintenance capabilities of complex industrial microgrid scenario by coupling a novel power-aware adaptive kernel function with a hybrid Wiener-GRU degradation algorithm. The overall accuracy of 99% demonstrates the excellent performance of the proposed framework which can allow early identification of performance degradation, enabling proactive scheduling for the maintenance actions which can ensure enhanced reliability, reduced maintenance costs, and improved operational efficiency in complex microgrid networks.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/197265