Interpreting What Typical Fault Signals Look Like via Prototype-matching

arXiv:2403.07033v1 Announce Type: new
Abstract: Neural networks, with powerful nonlinear mapping and classification capabilities, are widely applied in mechanical fault diagnosis to ensure safety. However, being typical black-box models, their application is limited in high-reliability-required scenarios. To understand the classification logic and explain what typical fault signals look like, the prototype matching network (PMN) is proposed by combining the human-inherent prototype-matching with autoencoder (AE). The PMN matches AE-extracted feature with each prototype and selects the most similar prototype as the …

abstract application arxiv box capabilities classification cs.ai cs.lg diagnosis however logic look mapping networks neural networks reliability safety type via