Meta-autoencoder-based zero-shot learning for insulation defect diagnosis in gas-insulated switchgear

Data-driven methods are the primary methods of training models for the diagnosis of insulation defects in gas-insulated switchgear (GIS). Due to complicated operating environments, target samples are not available for training sometimes, leading to insufficient feature learning. Therefore, a meta-autoencoder-based zero-shot learning (MAZL) method is proposed for the diagnosis of GIS insulation defects. First, the visual features of insulation defects' signals are extracted by a convolutional neural network. Next, the mapping between visual and semantic spaces is learned by semantic autoencoder (SAE). For SAE, the triplet loss is adapted to align semantic information, and unknown class semantic constraints are used to alleviate the domain offset. Finally, a nearest-neighbor algorithm is used to achieve zero-sample diagnosis. In model training, episode training is introduced to make the MAZL model learn a more discriminative semantic prototype, which alleviates classification deviations. The MAZL achieved a diagnostic accuracy for single and multi-source GIS insulation defects of 96.215% and 90.41% without using test classes for training, respectively. This finding provides ideas for the diagnosis of new GIS insulation defects.