OPN: Open-Set Semi-Supervised Learning for Intelligent Fault Diagnosis of Rotating Machinery

Semi-supervised learning (SSL) is effective in addressing the scarcity of label information in the fault diagnosis of rotating machinery. However, existing SSL methods generally assume that the labeled and unlabeled fault samples share a consistent label space, which is difficult to guarantee in practical applications and limits the applicability of SSL. To address this issue, this study presents a method denoted as open-set semi-supervised learning (O3SL) for intelligent fault diagnosis (IFD). First, an orthogonal projection network (OPN) is proposed to classify fault samples based on feature projection. Moreover, an open-set recognition (OSR) module based on feature projection residual (FPR) is further presented to discover potential unknown fault types in unlabeled fault samples. Finally, a pseudo-label correction updating (PLCU) module is developed to further dynamically generate corrected pseudo-labels and thus to improve the performance of OPN. Extensive experiments on two gearbox fault datasets have validated the effectiveness of the proposed OPN-based O3SL method.