Dual-stage recognition framework for open-set fault diagnosis in rotating machinery considering varying inter-class similarity

Key Points

• The dual-stage recognition framework improves fault diagnosis accuracy in rotating machinery, addressing challenges of varying inter-class similarity.
• Findings demonstrate a notable increase in diagnostic performance under open-set conditions, achieving precision rates above 85 percent.
• Observational analysis of performance metrics highlights significant strides in managing complexities of machinery faults effectively.
• The findings suggest potential application in industrial settings, but further validation in diverse machine types is warranted.