Evaluation of Surrogate Modelling Methods for Turbo-Machinery Component design Optimization

Gianluca Badjan, Carlos Poloni, Andrew Pike, Nadir Ince

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

Surrogate models are used to approximate complex problems in order to reduce the final cost of the design process. This study has evaluated the potential for employing surrogate modelling methods in turbo-machinery component design optimization. Specifically four types of surrogate models are assessed and compared, namely: neural networks, Radial Basis Function (RBF) Networks, polynomial models and Kriging models. Guidelines and automated setting procedures are proposed to set surrogate models, which are applied to two turbo-machinery application studies.
Original languageEnglish
Title of host publicationAdvances in Evolutionary and Deterministic Methods for Design, Optimization and Control in Engineering and Sciences.
EditorsDavid Greiner, Blas Galván, Jacques Périaux, Nicolas Gauger, Kyriakos C. Giannakoglou, Gabriel Winter
PublisherSpringer
Chapter13
Pages209-223
Number of pages14
ISBN (Print)978-3-319-11540-5
Publication statusPublished - 31 Jan 2015

Publication series

NameComputational Methods in Applied Sciences
PublisherSprnger
Volume36

Keywords

  • Surrogate models
  • Neural networks
  • Turbo-machinery

Fingerprint Dive into the research topics of 'Evaluation of Surrogate Modelling Methods for Turbo-Machinery Component design Optimization'. Together they form a unique fingerprint.

  • Cite this

    Badjan, G., Poloni, C., Pike, A., & Ince, N. (2015). Evaluation of Surrogate Modelling Methods for Turbo-Machinery Component design Optimization. In D. Greiner, B. Galván, J. Périaux, N. Gauger, K. C. Giannakoglou, & G. Winter (Eds.), Advances in Evolutionary and Deterministic Methods for Design, Optimization and Control in Engineering and Sciences. (pp. 209-223). (Computational Methods in Applied Sciences; Vol. 36). Springer.