The development of a rule based expert system to automate the digital analysis of condition monitoring parameters captured on rolling element bearings subjected to simulated failure

Master Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title

University of Cape Town

This synopsis provides a brief summary of the development of a rule based expert system to diagnose bearing failure. Firstly it covers the proposal of a generic, expert system based industrial condition monitoring system. It then discusses in more detail the development of a specific aspect ofthe system, viz. the analysis of rolling element bearing condition. The bearing test rig and data capture system are described, followed by primary research to define the bearing analysis solution space. This includes the use of vibration parameters, measured and derived operating conditions and the bearing running condition. It then explains the development of rulebases for the three analysis tasks of detection, diagnosis and prognosis. Included is a discussion on techniques used to normalise and adjust the vibration parameters to allow analysis under any operating conditions. Finally the synopsis is concluded with a discussion on the performance of the system and contributions made to the developing field of condition monitoring using expert systems.