Graph Neural Network-Based Semi-Supervised Open-Set Fault Diagnosis for Marine Machinery Systems

TL;DR

The paper tackles open-set fault diagnosis for marine propulsion systems under limited labeled data. It introduces SOFD-GCN, a semi-supervised framework that uses a Graph Convolutional Network for supervised feature learning on graph-structured sensor data, constructs a reliability subset for pseudo-labeling unknown faults, and trains a semi-supervised classifier that includes an unknown class $(K+1)$. Key contributions include a robust reliability-subset construction via multi-layer feature fusion and discriminant statistics, and a demonstrated performance boost (macro-F1 > 0.97 across speeds) on a public maritime benchmark relative to open-set baselines. This approach enables simultaneous accurate recognition of known faults and reliable detection of unknown faults, supporting practical deployment in complex marine environments.

Abstract

Recently, fault diagnosis methods for marine machinery systems based on deep learning models have attracted considerable attention in the shipping industry. Most existing studies assume fault classes are consistent and known between the training and test datasets, and these methods perform well under controlled environment. In practice, however, previously unseen or unknown fault types (i.e., out-of-distribution or open-set observations not present during training) can occur, causing such methods to fail and posing a significant challenge to their widespread industrial deployment. To address this challenge, this paper proposes a semi-supervised open-set fault diagnosis (SOFD) framework that enhances and extends the applicability of deep learning models in open-set fault diagnosis scenarios. The framework includes a reliability subset construction process, which uses a multi-layer fusion feature representation extracted by a supervised feature learning model to select an unlabeled test subset. The labeled training set and pseudo-labeled test subset are then fed into a semi-supervised diagnosis model to learn discriminative features for each class, enabling accurate classification of known faults and effective detection of unknown samples. Experimental results on a public maritime benchmark dataset demonstrate the effectiveness and superiority of the proposed SOFD framework.

Figures (7)

• Figure 1: Schematic diagram of the propulsion system tan2021multi
• Figure 2: Pipeline of the proposed fault diagnosis method.
• Figure 3: The proposed fault diagnosis framework.
• Figure 4: Confusion matrices of the compared methods and the proposed SOFD-GCN at 9 speeds (New denotes the unknown fault).
• Figure 5: T-SNE visualization of features extracted by model $M_0$ and after semi-supervised learning by $M_1$ for three known classes ($F_1$, $F_2$, $F_3$) and one unknown class ($F_4$) at 9 speeds.