On implementation of selected methods of artificial intelligence in expert systems for thermal and flow diagnostics

• Autorzy: Głuch J.
• Języki publikacji: EN

Abstrakty

EN

Discussed has been a problem of aiding the operators of steam turbine cycles with the expert systems on the topic of exploitation decisions. Attention has been focused on realization of a one of the problems based on determination of the extent of exploitation degradation of geometric parameters of steam turbine blading. Discussion has been conducted on the sample of one of the components of the methods of artificial intelligence: a selected type of artificial neural network (ANN). Such ANN, which detects the extent of geometric degradation indicates a high quality and accuracy based on a good identification of the extent of degradation both in the case of complete and incomplete measurement data. In such way fulfilled are the requirements faced by the aiding methods regarding the automatic expert systems.

Słowa kluczowe

EN

thermal and flow diagnostics; expert systems; neural networks

• Wydawca: Wydawnictwo Instytutu Maszyn Przepływowych PAN
• Czasopismo: Transactions of the Institute of Fluid-Flow Machinery
• Rocznik: 2003
• Tom: nr 113
• Strony: 179--190
• Opis fizyczny: Bibliogr.15 poz., tab., wykr.

Twórcy

autor

Głuch J.

• Gdańsk Univerity of Technology, Faculty of Ocean Engineering and Ship Technology, Chair of Ship Automatic Control and Turbine Propulsion, Narutowicza 11/12, 80-952 Gdańsk, Poland,

Bibliografia

• 1. Błażko E., Lidke M.: Instruction of the TURBINA program, , Int. Reports of the IFFM, Gdańsk, Poland. (1977-1992), (in Polish),
• 2. Cholewa W., Kiciński J.: Technical Diagnostics - Inverse Diagnostic Models, Silesian Technical University, Gliwice 1997 (in Polish).
• 3. Demuth H., Beale M.: Neural Network Toolbox for use with Matlab, The MathWorks Inc., Natick 1996.
• 4. Fan J. Y., Nicolau M., White R. E.: An approach to fault diagnosis of chemical processes via neural networks, AIChE Journal, Vol. 39, No. 1.
• 5. Gardzilewicz A., Głuch J., Bogulicz M,, Uziębło W., Jankowski T: Instruction of the DIAGAR program for TURÓW Power Station,, Rep. No. 20/01, Diagnostyka Maszyn, Sp. z o.o., Gdańsk 2001 (in Polish).
• 6. Głuch J., Krzyżanowski J.: On Thermal diagnostics of power cycles on the example of regenerative heat exchanger, Proc. of KDT'76 Conf., Gdańsk, September 1996, 1/ł-17 (in Polish).
• 7. Głuch J.: Flow and heat diagnostic analysis for a group of steam turbine stages, Proc. of the 2-nd Europ. Conf. on Turbomachinery - Fluid Dynamics and Therrnodynamics, March 5-7, 1997, Antwerpen, Belgium
• 8. Głuch J., Krzyżanowski J.: Searching for inefficient components of compler power systems, Proc. ASME-TURBO-EXPO'98, Paper No. 98-GT-76, Stockholm, Sweden
• 9. Głuch J.,Krzyżanowski J.: Application of incomplete measurement data to thermal diagnostics of complex power systems components, Proc. ASMB TURBO-EXPO'2000, Paper No. 2000-GT-0056, May 5-11, Munich 2000, Germany.
• 10. Głuch J.,Krzyżanowski J,: Diagnostics of steam turbine IP part under condition of insuficient measurement data, Proc of the Diagnostics'2O0O Conf. September 18-22,2000, Warsaw (in Polish).
• 11. Gluch, J.,Krzyżanowski, J.: Applicatźon of preprocessed classifier type neural network for searching ol faulty components of power cycles in case of incomplete measumerent data. ASME-TURB O-EXP O' 2002 Conference Paper No. GT-2002-30028, Amsterdam 2002, Nederland
• 12. Kościelny J, M.: The problem of multiple defects in industrial processes diagnostic Proc. of the Ind. Processes Diagnostics Conf. DPP'2001, September 17-19,2001, Łagów, Poland (in Polish)
• 13. Krzyżanowski J.,'Krzyślak P. Głuch J.: On application of statistic to validation of turbomachinery calculation methods Proc. ASME-TURBO EXPO'92, Paper No. 92-GT-272, Cologne 1992, Germany.
• 14. Masters T.: Practical Neural Network Recipes in C++, Academic Press, San Diego, CA, 1993
• 15. Pieczyński A.: Computerised Diagnostic Systems of Indriustal Processes Tech. Univ. of Zielona Góra, Zielona Góra, Poland 1999, (in Polish)