DOI QR코드

DOI QR Code

Optimization of a Centrifugal Compressor Impeller(II): Artificial Neural Network and Genetic Algorithm

원심압축기 최적화를 위한 연구(II): 인공지능망과 유전자 알고리즘

  • 최형준 (경상대학교 항공공학과 대학원) ;
  • 박영하 (경상대학교 항공공학과 대학원) ;
  • 김재실 (창원대학교 기계공학과) ;
  • 조수용 (경상대학교 항공기부품기술연구센터)
  • Received : 2010.12.13
  • Accepted : 2011.04.20
  • Published : 2011.05.01

Abstract

The optimization of a centrifugal compressor was conducted. The ANN (Artificial Neural Network) was adopted as an optimization algorithm, and it was learned and trained with the DOE (Design of Experiment). In the DOE, it was predicted the main effect and the interaction effect of design variables to the objective function. The ANN was improved in the optimization process using the GA (Genetic Algorithm). When any output at each generation was reached a standard level, it was re-calculated by the CFD (Computational Fluid Dynamics) and it was applied to develop a new ANN. After 6th generation, the prediction difference between ANN and CFD was less than 1%. A pareto of the efficiency versus the pressure ratio was obtained through the 21th generation. Using this method, the computational time for the optimization was equivalent to the time consumed by the gradient method, and the optimized results of multi-objective function were obtained.

원심압축기 임펠러의 최적화연구를 수행하였다. 최적화를 위한 알고리즘은 ANN를 기본으로 하였으며, 초기의 ANN 학습은 DOE를 사용하여 ANN을 효과적으로 형성하였다. DOE에서는 설계변수가 목적함수에 미치는 주효과와 상호 교호작용에 대한 예측을 할 수 있었다. 최적화과정에서 ANN의 향상을 위하여 GA를 사용하여 각 세대에서의 설계변수에 따른 목적함수가 일정값 이하가 되는 경우에는 수치해석을 통하여 ANN을 세대별로 향상시켰다. 6세대 이 후에는 ANN에 의한 예측값과 CFD의 예측값과의 차이가 1% 미만에 도달하였다. 총 21세대를 거쳐서 압축비와 효율과의 pareto를 형성할 수 있었다. 본 연구에서는 최적화를 위한 계산시간을 기울기 기반의 최적화시간 정도로 단축하면서도 다목적함수의 최적화의 결과를 얻을 수 있었다.

Keywords

References

  1. Cumpsty, N. A., "Compressor Aerodynamics", Longman Group UK, 1989.
  2. Whitfield, A. and Baines, N. C., "Design of Radial Turbomachines", Longman Group UK, 1990.
  3. Aungier, R. H., "Centrifugal Compressor", ASME Press, 2000.
  4. Bonaiuti, D. and Pediroda, V., “Aerodynamic Optimization of an Industrial Centrifugal Compressor Impeller Using Genetic Algorithms", Proceedings of Eurogen 2001.
  5. Cosentino, R., Alsalihi, Z., and Braembussche, V.R.A., "Expert System for Radial Impeller Optimization", 2001 Proceedings of Euroturbo4 ATI-CST-039/01.
  6. Bonaiuti, D, Arnone, A., Ermini, M. Baldassarre, L., "Analysis and Optimization of Transonic Centrifugal Compressor Impellers Using the Design of Experimental Technique", 2002, GT-2002-30619.
  7. 김세미, 박준영, 안국영, 백제현, “반응면기법을 이용한 원심압축기 최적설계”, 한국전산유체공학 추계학술대회, 2007, pp. 92-99.
  8. Bonaiuti, D. and Zangeneh, M., "On the Coupling of Inverse Design and Optimization Techniques for the Multiobjective, Multipoint Design of Turbomachinery Blades", J. of Turbomachinery, 2009, Vol. 131, 021014-1. https://doi.org/10.1115/1.2950065
  9. Verstraete, T., Alsalihi, Z. and Braembussche, R.A.V., "Multidisciplinary Optimization of a Radial Compressor for Microgas Turbine Applications", J. of Turbomachinery, 2010, Vol. 132, 031004-1. https://doi.org/10.1115/1.3144162
  10. 최형준, 박영하, 안국영, 조수용, “원심압축기 최적화를 위한 연구 I-형상변수 및 설계변수에 관한연구”, 2011, 항공우주학회지, 39권 5호. https://doi.org/10.5139/JKSAS.2011.39.5.424
  11. Eckardt, D., "Instantaneous Measurements in the Jet-Wake Discharge Flow of a Centrifugal Compressor Impeller", J. of Engineering and Power, 1975, pp. 337-346.
  12. Eckardt, D., "Detailed Flow Investigations Within a High-Speed Centrifugal Compressor Impeller", J. of Fluid Engineering, 1976, pp. 390-402. https://doi.org/10.1115/1.3448334
  13. Eckardt, D., "Flowfield Analysis of Radial and Backswept Centrifugal Compressor Impellers Part 1: Flow Measurements Using a Laser Velocimeter", In Performance Prediction of Centrifugal Pumps and Compressor edited by Gopalskrishnan, S. and Cooper, P., 1980, ASME.
  14. Cho, C. H., Cho, S. Y., Ahn, K. Y. and Kim, Y. C., “Study of an Axial-Type Fan Design Technique Using an Optimization Method", Proc. IMechE Vol.223, Part E:J. Process Mechanical Engineering, 2009, pp. 101-111. https://doi.org/10.1243/09544089JPME254
  15. Wang, Z., Xi, G. and Wang, X., "Aerodynamics Design Optimization of Vaned Diffuser for Centrifugal Compressors Using Kriging Model", Proc. Asian Joint Conf. on Propul. and Power, China, Beijing, 2006, AJCPP2006-22023.
  16. Shu, X., Gu, C., Xiao, J. ad Gao C., "Centrifugal Compressor Blade Optimization Based on Uniform Design and Genetic Algorithms'" Frontier Energy Power Eng. China, Vol. 2(4), 2008, pp. 454-456. https://doi.org/10.1007/s11708-008-0083-5
  17. Bonaiuti, D. and Pediroda, V., “Aerodynamic Optimization of an Industrial Centrifugal Compressor Impeller using Genetic Algorithms”, Proceedings of Eurogen 2001.
  18. Fan, H. Y., "An Inverse Design Method of Diffuser Blades by Genetic Algorithms", Proceedings Institutes Mech. Engineering(A), 1998, Vol. 212, pp. 261-268.
  19. Pierret, S. and Braembussche, R. A., "Turbomachinery Blade Design Using a Navier-Stokes Solver and Artificial Neural Network“, J. of Turbomachinery, 1999, Vol. 121, pp. 326-332. https://doi.org/10.1115/1.2841318
  20. Fan, H. Y., "A Neural Network Approach for Centrifugal Impeller Inverse Design", Proceedings Institutes Mech. Engineering(A), 2000, Vol. 214, pp. 183-186. https://doi.org/10.1243/0957650001538272
  21. Kim, J. H.. Choi, J. H., Husain, A. and Kim, K. Y., "Multi-Objective Optimization of a Centrifugal Compressor Impeller Through Evolutionary Algorithms", Proc. IMechE Vol. 224, Paer A:J. Power and Energy, 2010, pp. 711-721. https://doi.org/10.1243/09576509JPE884
  22. Bonaiuti, D., Arone, A., Ermini, M. and Baldassarre, L., "Analysis and Optimization of Transonic Centrifugal Compressor Impellers Using the Design of Experiments Technique", J. of Turbomachinery, 2006, Vol. 128, pp. 786-797. https://doi.org/10.1115/1.1579507
  23. Hagan, M. T., and Menhaj. M. “Training Feed-Forward Networks with the Marquardt Algorithm”, IEEE Transactions on Neural Networks, 1999, Vol. 5, No. 6, pp. 989-993.
  24. Hagan, M. T., Demuth, H. B. and Beale, M. H. "Neural Network Design", Boston, MA: PWS Publishing, 1996.
  25. Demuth, H., Beale, M. and Haga, M., "Neural Network Toolbox 6, User's Guide", 2007. Matlab R2007b.

Cited by

  1. Optimal Design of Impeller Shroud for Centrifugal Compressor Using Response Surface Method vol.18, pp.4, 2015, https://doi.org/10.5293/kfma.2015.18.4.043