<![CDATA[Transformer oil functions as an insulating and cooling medium in high-voltage power systems, whose dielectric condition degrades over service life due to thermal aging, moisture ingress, and metallic contamination, leading to reduced Breakdown Voltage (BDV) and increased insulation failure risk that may necessitate oil regeneration, replacement, or indicate transformer end-of-life. Unlike Dissolved Gas Analysis (DGA), which evaluates transformer faults based on gas decomposition products, BDV directly reflects the dielectric strength of insulating oil and is more sensitive to particulate contamination such as Cu and Fe, making it more suitable for material-level insulation degradation assessment. This study investigates the influence of copper (Cu) and iron (Fe) particle contamination on BDV and compares three Artificial Neural Network (ANN) training strategies for BDV prediction: gradient-based training (DFFNN-Pure), Genetic Algorithm optimization (DFFNN-GA), and Grey Wolf Optimizer-based training (DFFNN-GWO), using experimental data from 36 transformer oil samples obtained in accordance with IEC 60156:2018. The comparison represents a before–after modeling perspective in terms of training strategy rather than repeated physical testing. The results show that DFFNN-Pure achieved the highest prediction accuracy (R² = 0.996, RMSE = 0.296 kV, MAE = 0.238 kV), while DFFNN-GWO demonstrated stable convergence with competitive accuracy (R² = 0.971, RMSE = 0.886 kV), whereas DFFNN-GA exhibited unstable convergence and poor generalization. Unlike previous studies that primarily focus on transformer remaining useful life estimation at the system level, this work emphasizes material-level BDV prediction of transformer oil under metallic contamination and provides a systematic comparison between gradient-based and metaheuristic training within the same DFFNN framework, supporting non-destructive condition monitoring and predictive maintenance.]]>
Copyrights © 2026
