Sensors, Год журнала: 2024, Номер 24(16), С. 5310 - 5310

Опубликована: Авг. 16, 2024

The Industrial Internet of Things has enabled the integration and analysis vast volumes data across various industries, with maritime sector being no exception. Advances in cloud computing deep learning (DL) are continuously reshaping industry, particularly optimizing operations such as Predictive Maintenance (PdM). In this study, we propose a novel DL-based framework focusing on fault detection task PdM marine operations, leveraging time-series from sensors installed shipboard machinery. is designed scalable cost-efficient software solution, encompassing all stages collection pre-processing at edge to deployment lifecycle management DL models. proposed architecture utilizes Graph Attention Networks (GATs) extract spatio-temporal information provides explainable predictions through feature-wise scoring mechanism. Additionally, custom evaluation metric real-world applicability employed, prioritizing both prediction accuracy timeliness identification. To demonstrate effectiveness our framework, conduct experiments three types open-source datasets relevant PdM: electrical data, bearing datasets, water circulation experiments.

Язык: Английский