A Comparison of Advanced Techniques for Monitoring the Condition of Machinery
conference paper
Within the project described in this paper, the objective of condition monitoring is to detect (upcoming) failures as early as possible, to minimise damage to the machinery. The involvement of humans within the condition monitoring process is being reduced by incorporation of advanced and intelligent monitoring systems. We describe and compare three different state-of-the-art condition monitoring techniques: first principles, advanced feature extraction and pattern recognition, and neural networks. Each description provides an enumeration of the advantages and disadvantages. The techniques were explored and tested by applying them to the condition monitoring of a diesel engine using the torsional vibration of the outgoing shaft, using actually measured data.
TNO Identifier
95098
Publisher
World Scientific
Source title
Fuzzy Logic and Intelligent Technologies for Nuclear Science and Industry - Proceedings of the 3rd International FLINS Workshop, Antwerp, Belgium, September 14-16, 1998
Editor(s)
Ruan, D.
Abderrahim, H.A.
D'hondt, P.
Kerre, E.E.
Abderrahim, H.A.
D'hondt, P.
Kerre, E.E.
Place of publication
Singapore
Pages
238-247
Files
To receive the publication files, please send an e-mail request to TNO Repository.