<ul><li>Power transformers' reliable operation is crucial for grid stability, with Dissolved Gas Analysis (DGA) being a common method for fault diagnosis.</li><li>Traditional heuristic-based approaches for fault diagnosis can be inconsistent, leading to the adoption of machine learning (ML) techniques for improved accuracy.</li><li>This work introduces a feature-weighted domain adaptation method, MCW, combining Maximum Mean Discrepancy (MMD) and Correlation Alignment (CORAL) with feature-specific weighting.</li><li>Experimental results on power transformer datasets show that MCW outperforms Fine-Tuning and MMD-CORAL (MC), achieving a 7.9% improvement and demonstrating robustness across various training sample sizes.</li></ul>

Feature-Weighted MMD-CORAL for Domain Adaptation in Power Transformer Fault Diagnosis

Discover more