Methodology of knowledge acquisition for machinery diagnostics

Wojciech Moczulski

Wojciech Moczulski Silesian Technical University

Abstract

The paper deals with a comprehensive methodology concerning knowledge acquisition on machinery for the purpose of expert systems suitable for aiding of diagnostic inference. ·The methodology includes selected methods of diagnostic knowledge representation, methods of knowledge acquisition from domain experts and from preclassified examples, methods of assessment of previously acquired knowledge and a scenario of knowledge acquisition process. All the methods have been implemented in a Knowledge Acquisition System. Moreover, some examples of applications of the elaborated methodology have been given.

Keywords

machinery diagnostics, knowledge acquisition, domain experts, machine learning, assessment of knowledge,

References

[1] W. Cholewa. Real-time expert systems for technical diagnostics. In: 15th IMACS World Congress on Scientific Computation, Modelling and Applied Mathematics, Proceedings, Vol. 6: 715- 720, Berlin, 1997.
[2] W. Cholewa, J . Kiciński, eds. Machinery Diagnostics. Inverted Diagnostic Models (in Polish). Monographs - Machine Building and Exploitation. Silesian Technical University, Gliwice, 1997.
[3] J . W. Grzymała-Busse. Managing uncertainty in machine learning from examples. In: M. Dąbrowski, M. Michalewicz, Z. Raś, eds., Intelligent Information Systems. Proceedings of the Conference "Practical Aspects of Artificial Intelligence III", 70-84, Wigry. Institute of Computer Science PAS, Warszawa, 1994.
[4] K. A. Kaufman. INLEN: A Methodology and integrated system for knowledge discovery in databases. Dissertation. George Mason University, Fairfax, VA, USA, 1997.
[5] P. Kostka. Expert System for Application of Multi-Dimensional Signal Analysis for Investigations of Rotor Vibrations (in Polish) . M. Sc. Thesis. Dept. of Fundamentals of Machine Design, Silesian Tech. University, Gliwice, 1997.