CABLEGNOSIS Publication in Elsevier – Advanced Condition Monitoring of Power Cables

As part of the project’s dissemination activities, CABLEGNOSIS researchers (Mohsen Abdolahi, Wenjuan Song and Mohammad Yazdani-Asrami), from the “Propulsion, Electrification and Superconductivity” group, Jamse Watt school of engineering, University of Glasgow, United Kingdom, published a journal paper titled as “Intelligent Condition Monitoring of Power Cables Using Advanced Machine Learning Models” in journal of “Results in Engineering” on 11 November 2025. Results in Engineering journal is Q1 journal which is published by Elsevier with an IF of 7.9.

This study presented an intelligent health-monitoring framework that applies advanced machine learning techniques to assess the condition of 15 kV and 20 kV XLPE cables using features such as partial discharge, age, visual inspection, and neutral corrosion. 18 ML models with Bayesian-optimised hyperparameters were benchmarked, and boosting-based methods achieved the best performance with accuracies above 98%. The framework was further validated on a separate dataset of 138 kV EPR cables, where it maintained F1-scores above 98%. The framework supports condition-based online diagnostics, enabling reliable health-index prediction, improved planning and network reconfiguration, and a pathway toward future digital-twin development for power cable systems. The advances offered by the research in this paper is fully aligned with the objectives of the CABLEGNOSIS project.

CABLEGNOSIS Elsevier Results in Engineering journal paper on Intelligent Condition Monitoring of Power Cables Using Advanced Machine Learning Models DOI: https://doi.org/10.1016/j.rineng.2025.108371