CABLEGNOSIS at IEEE Conference in UK – Proactive Prognostics for Insulating Materials

As part of the project’s dissemination activities, CABLEGNOSIS researchers (Mohsen Abdolahi, Wenjuan Song, Rob Ross, Aart-Jan Graaf and Mohammad Yazdani-Asrami), received the official acceptance for an abstract in the 15th International Electrical Insulation Conference (INSUCON2026). Authors will participate in-person in INSUCON2026, which will be held in Birmingham, United Kingdom, from 21–23 April 2026. INSUCON continues its long-standing mission of advancing electrical insulation technology across power systems, aerospace, transportation, renewable energy, and emerging insulation materials. The abstract is titled as “Arc Fault Prediction of Insulation Materials in More Electric Aircraft Using Artificial Intelligence Techniques” and it contributes to the conference theme on “Aerospace Insulation Challenges” by introducing a deep learning framework for forecasting arc-fault voltage and current waveforms on Polytetrafluoroethylene (PTFE) insulation materials.

This study demonstrates that a deep learning framework can reliably predict parallel arc faults within the insulating materials by learning their complex behaviour from experimental data, overcoming limitations of traditional analytical models. It introduces a data-driven approach for modelling nonlinear insulation degradation and ageing, while it supports proactive prognostics by enabling early detection of faults and monitoring of insulation health, a capability that can be extended to power cables for improving reliability and preventing catastrophic failures. The advances offered by the research in this paper is fully aligned with the objectives of the CABLEGNOSIS project.