The tribological failure diagnosis of spur gear by an expert system

Y. L. Su, J. S. Lin, S. K. Hsieh

Research output: Contribution to journalArticlepeer-review

3 Citations (Scopus)


The failure of tribo-elements at an early stage, before the designed lifetime, is attributable to the complex interaction of many factors, which can be diagnosed by various techniques. These techniques can be learned in a certain period of time, while the knowledge of failure analysis must have accumulated from a long experience of practical work. For this reason, a computerized expert system program, developed from artificial intelligence (PC Plus, an inference engine shell), was constructed for spur gear tribological failure diagnosis. The knowledge was expressed as many "if-then" rules and a "frame" structure of inheritance. Note that the certainty factor of the rules is itself a special characteristic of this system and the "man-machine" interface is very friendly, the graphical interpretation being an example. The system was finally validated by the twin roller wear test which can be recognized as the motion of a spur gear near the pitch-line region. The failure characteristics of the worn rollers were transferred to the expert system by means of a "user-friendly" interface to deduce the reason for the failure.

Original languageEnglish
Pages (from-to)187-196
Number of pages10
Issue number2
Publication statusPublished - 1993 Jul 1

All Science Journal Classification (ASJC) codes

  • Condensed Matter Physics
  • Mechanics of Materials
  • Surfaces and Interfaces
  • Surfaces, Coatings and Films
  • Materials Chemistry


Dive into the research topics of 'The tribological failure diagnosis of spur gear by an expert system'. Together they form a unique fingerprint.

Cite this