New PG-AMF framework enhances bearing fault diagnosis

Researchers have developed a new framework called Parametric Generalized Adaptive Moment Features (PG-AMF) for bearing fault diagnosis and machine health monitoring. This approach learns feature characteristics directly from data, moving beyond predefined statistical descriptors that have limited adaptability. PG-AMF extracts multiple complementary representations, including absolute, signed moment, and AC-coupled moment features, while also modeling interactions between sensor channels for enhanced fault representation. Evaluations on a gearbox bearing dataset demonstrated improved classification performance and generalization capability compared to conventional methods. AI

IMPACT This new framework could improve the accuracy and adaptability of predictive maintenance systems in industrial settings.

RANK_REASON Research paper detailing a new feature extraction framework for machine health monitoring. [lever_c_demoted from research: ic=1 ai=0.7]

Read on arXiv cs.AI →

• alphaXiv
• arXiv
• CatalyzeX Code Finder for Papers
• Connected Papers
• DagsHub
• Gotit.pub
• Hugging Face
• Litmaps
• Parametric Generalized Adaptive Moment Features (PG-AMF)
• Rajeev Kumar
• ScienceCast
• scite Smart Citations

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →