Machine condition monitoring enables reliable and e conomical way of action for maintenance operations in modern industrial plants. Increasing number of m easurement points and more demanding problems requi re a tomatic fault detection. Advanced signal processing methods exposed failures earlier and then it’s possible to plan more operating time and less shutdowns. Intelligent meth ods have been increasingly used in model based faul t di gnosis and intelligent analysers. Intelligent methods provide various techniques for combining a large number of features. A test rig was used to simulate different fault types and changes in operating conditions. Linguistic equation (LE) models were developed for the normal operation and nine fault c ases including rotor unbalance, bent shaft, misalig nment and bearing faults. Classification is based on the degrees of m embership developed for each case from the fuzzines s of the LE models. The classification results of the experiment al cases are very good and logical. As even very sm all faults are detected by a slight increase of membership, the re sults are very promising for early detection of fau lts. Together with the compact implementation and the operability of t he normal model, this makes the extension to real w or d problems feasible.
[1]
Esko Juuso,et al.
Industrial utilization of linguistic equations for defect detection on printed circuit boards
,
2002,
IEEE 2002 28th Annual Conference of the Industrial Electronics Society. IECON 02.
[2]
蔡 篤儀,et al.
XIV IMEKO World Congress, New Measurements : Challenges and Visions 報告
,
1998
.
[3]
Esko Juuso,et al.
Integration of intelligent systems in development of smart adaptive systems
,
2004,
Int. J. Approx. Reason..
[4]
Mika Ruusunen,et al.
Monitoring of Automated Screw Insertion Processes-A Soft Computing Approach
,
2003
.
[5]
Esko Juuso,et al.
Fuzzy Control in Process Industry: The Linguistic Equation Approach
,
1999
.