A Fuzzy-Neuro Scheme for Fault Diagnosis and Life Consumption of Rotordynamic Systems
Abstract
An intelligent health monitor for rotating machinery is presented that integrates proven neural network and fuzzy logic technologies with rotordynamic, finite element modeling. A rotor demonstration rig is used as a proof of concept tool. The approach integrates rotor shaft vibration measurements with detailed, rotordynamic, finite element models through a fuzzy-neuro scheme which is specifically developed to respond to the rotor system being monitored. The advantage of this approach over current methods lies in the use of a neural network classifier and fuzzy logic reasoning algorithms. The real time neural network is trained to contain the knowledge of a detailed finite element model whose results are integrated with system measurements to produce accurate machine fault diagnostics and component stress predictions. The availability of these real time stresses allows for critical component life estimates to be calculated during machine operation. Fuzzy logic is implemented to overcome system measurements uncertainties, provide machine fault severity information, and make informed decisions about maintenance actions that should be performed based on operator experience.
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 01, 1996
- Accession Number
- ADP010214
Entities
People
- Michael J. Roemer