New Multifeature Information Health Index (MIHI) Based on a Quasi-Orthogonal Sparse Algorithm for Bearing Degradation...

Area of Science:

• Mechanical Engineering
• Data Science
• Signal Processing

Background:

• Prognostic health management (PHM) systems are crucial for monitoring and diagnosing defective bearing signals.
• Current PHM systems struggle to simultaneously identify bearing degradation trends and specific fault types.
• Accurate fault identification is essential for selecting cost-effective maintenance strategies.

Purpose of the Study:

• To develop a multifeature information health index (MIHI) capable of tracing diverse bearing degradation trends alongside various fault types.
• To enable simultaneous identification of degradation status and fault classification for improved equipment maintenance.

Main Methods:

• A novel quasi-orthogonal sparse projection algorithm was developed to process feature vector sets (e.g., spectrum).
• The algorithm converts degraded feature sets into orthogonal approximate spatial straight lines.
• A MIHI is constructed using spectral data from current state measurements and transformed via the algorithm.

Main Results:

• The proposed method effectively traces various bearing degradation trends across different fault types.
• Simultaneous identification of degradation trends and fault types was achieved.
• Case studies using bearing degradation data validated the approach's effectiveness.

Conclusions:

• The developed MIHI and quasi-orthogonal sparse projection algorithm offer a robust solution for bearing PHM.
• This approach enhances condition monitoring and diagnosis by providing simultaneous trend tracing and fault identification.
• The findings support reduced equipment operational costs through optimized, future-fault-aware maintenance planning.