Anomaly detection in flexible mechanical couplings via symbolic time series analysis

Amol Khatkhate, Shalabh Gupta, Asok Ray, Ravi Patankar

Research output: Contribution to journalArticlepeer-review

11 Scopus citations

Abstract

Critical components of a rotating machinery such as bearings and couplings are often subjected to unbalanced axial and radial loads due to excessive machine vibrations arising from shaft misalignment(s). The paper presents Symbolic Time Series Analysis (STSA) of bearing acceleration data for detection and estimation of gradually developing parametric changes in flexible disc/diaphragm couplings. The analytical method is built upon the principles of Symbolic Dynamics, Automata Theory, and Statistical Pattern Recognition. The anomaly detection methodology is validated on a real-time simulation test bed, where the dynamic model of a flexible mechanical coupling is subjected to angular misalignment(s) leading to coupling failure. Damage patterns are identified from STSA of multiple data sets generated for different input torques. Statistical estimates are obtained for small changes in the coupling stiffness based on the information derived from the ensemble of damage patterns.

Original languageEnglish (US)
Pages (from-to)608-622
Number of pages15
JournalJournal of Sound and Vibration
Volume311
Issue number3-5
DOIs
StatePublished - Apr 8 2008

All Science Journal Classification (ASJC) codes

  • Condensed Matter Physics
  • Mechanics of Materials
  • Acoustics and Ultrasonics
  • Mechanical Engineering

Fingerprint Dive into the research topics of 'Anomaly detection in flexible mechanical couplings via symbolic time series analysis'. Together they form a unique fingerprint.

Cite this