Design optimization of a centrifugal compressor impeller by multi-objective genetic algorithm

Jin Hyuk Kim, Jae Ho Choi, Afzal Husain, Kwang Yong Kim

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

This paper presents design optimization of a centrifugal compressor impeller with hybrid multi-objectives evolutionary algorithm (hybrid MOEA). Reynolds-averaged Navier-Stokes equations with shear stress transport turbulence model are discretized by finite volume approximations and solved on hexahedral grids for flow analyses. Latin hypercube sampling of design of experiments is used to generate design points within the selected design space. Two objectives, i.e., isentropic efficiency and total pressure ratio are selected with four design variables defining impeller hub and shroud contours in meridional contours to optimize the system. Non-dominated Sorting of Genetic Algorithm (NSGA-II) with ε -constraint strategy for local search coupled with surrogate model is used for multi-objective optimization. The surrogate model, Radial Basis Neural Network is trained on the numerical solutions by carrying out leave-one-out cross-validation for the data set. The trade-off between the two objectives has been found out and discussed in light of the Pareto-optimal solutions. The optimization results show that isentropic efficiencies and total pressure ratios of the cluster points at the Pareto-optimal solutions are enhanced by multi-objective optimization.

Original languageEnglish
Title of host publicationProceedings of the ASME Fluids Engineering Division Summer Conference 2009, FEDSM2009
Pages185-191
Number of pages7
Volume2
DOIs
Publication statusPublished - 2009
Event2009 ASME Fluids Engineering Division Summer Conference, FEDSM2009 - Vail, CO, United States
Duration: Aug 2 2009Aug 6 2009

Other

Other2009 ASME Fluids Engineering Division Summer Conference, FEDSM2009
CountryUnited States
CityVail, CO
Period8/2/098/6/09

Fingerprint

Centrifugal compressors
Genetic algorithms
Multiobjective optimization
Turbulence models
Sorting
Evolutionary algorithms
Design of experiments
Navier Stokes equations
Shear stress
Sampling
Neural networks
Design optimization

Keywords

  • Centrifugal compressor impeller
  • Design optimization
  • Evolutionary algorithm
  • Pareto-optimal

ASJC Scopus subject areas

  • Fluid Flow and Transfer Processes
  • Mechanical Engineering

Cite this

Kim, J. H., Choi, J. H., Husain, A., & Kim, K. Y. (2009). Design optimization of a centrifugal compressor impeller by multi-objective genetic algorithm. In Proceedings of the ASME Fluids Engineering Division Summer Conference 2009, FEDSM2009 (Vol. 2, pp. 185-191) https://doi.org/10.1115/FEDSM2009-78486

Design optimization of a centrifugal compressor impeller by multi-objective genetic algorithm. / Kim, Jin Hyuk; Choi, Jae Ho; Husain, Afzal; Kim, Kwang Yong.

Proceedings of the ASME Fluids Engineering Division Summer Conference 2009, FEDSM2009. Vol. 2 2009. p. 185-191.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kim, JH, Choi, JH, Husain, A & Kim, KY 2009, Design optimization of a centrifugal compressor impeller by multi-objective genetic algorithm. in Proceedings of the ASME Fluids Engineering Division Summer Conference 2009, FEDSM2009. vol. 2, pp. 185-191, 2009 ASME Fluids Engineering Division Summer Conference, FEDSM2009, Vail, CO, United States, 8/2/09. https://doi.org/10.1115/FEDSM2009-78486
Kim JH, Choi JH, Husain A, Kim KY. Design optimization of a centrifugal compressor impeller by multi-objective genetic algorithm. In Proceedings of the ASME Fluids Engineering Division Summer Conference 2009, FEDSM2009. Vol. 2. 2009. p. 185-191 https://doi.org/10.1115/FEDSM2009-78486
Kim, Jin Hyuk ; Choi, Jae Ho ; Husain, Afzal ; Kim, Kwang Yong. / Design optimization of a centrifugal compressor impeller by multi-objective genetic algorithm. Proceedings of the ASME Fluids Engineering Division Summer Conference 2009, FEDSM2009. Vol. 2 2009. pp. 185-191
@inproceedings{acc912beb94e46e2b247ac3dc6076e6d,
title = "Design optimization of a centrifugal compressor impeller by multi-objective genetic algorithm",
abstract = "This paper presents design optimization of a centrifugal compressor impeller with hybrid multi-objectives evolutionary algorithm (hybrid MOEA). Reynolds-averaged Navier-Stokes equations with shear stress transport turbulence model are discretized by finite volume approximations and solved on hexahedral grids for flow analyses. Latin hypercube sampling of design of experiments is used to generate design points within the selected design space. Two objectives, i.e., isentropic efficiency and total pressure ratio are selected with four design variables defining impeller hub and shroud contours in meridional contours to optimize the system. Non-dominated Sorting of Genetic Algorithm (NSGA-II) with ε -constraint strategy for local search coupled with surrogate model is used for multi-objective optimization. The surrogate model, Radial Basis Neural Network is trained on the numerical solutions by carrying out leave-one-out cross-validation for the data set. The trade-off between the two objectives has been found out and discussed in light of the Pareto-optimal solutions. The optimization results show that isentropic efficiencies and total pressure ratios of the cluster points at the Pareto-optimal solutions are enhanced by multi-objective optimization.",
keywords = "Centrifugal compressor impeller, Design optimization, Evolutionary algorithm, Pareto-optimal",
author = "Kim, {Jin Hyuk} and Choi, {Jae Ho} and Afzal Husain and Kim, {Kwang Yong}",
year = "2009",
doi = "10.1115/FEDSM2009-78486",
language = "English",
isbn = "9780791843734",
volume = "2",
pages = "185--191",
booktitle = "Proceedings of the ASME Fluids Engineering Division Summer Conference 2009, FEDSM2009",

}

TY - GEN

T1 - Design optimization of a centrifugal compressor impeller by multi-objective genetic algorithm

AU - Kim, Jin Hyuk

AU - Choi, Jae Ho

AU - Husain, Afzal

AU - Kim, Kwang Yong

PY - 2009

Y1 - 2009

N2 - This paper presents design optimization of a centrifugal compressor impeller with hybrid multi-objectives evolutionary algorithm (hybrid MOEA). Reynolds-averaged Navier-Stokes equations with shear stress transport turbulence model are discretized by finite volume approximations and solved on hexahedral grids for flow analyses. Latin hypercube sampling of design of experiments is used to generate design points within the selected design space. Two objectives, i.e., isentropic efficiency and total pressure ratio are selected with four design variables defining impeller hub and shroud contours in meridional contours to optimize the system. Non-dominated Sorting of Genetic Algorithm (NSGA-II) with ε -constraint strategy for local search coupled with surrogate model is used for multi-objective optimization. The surrogate model, Radial Basis Neural Network is trained on the numerical solutions by carrying out leave-one-out cross-validation for the data set. The trade-off between the two objectives has been found out and discussed in light of the Pareto-optimal solutions. The optimization results show that isentropic efficiencies and total pressure ratios of the cluster points at the Pareto-optimal solutions are enhanced by multi-objective optimization.

AB - This paper presents design optimization of a centrifugal compressor impeller with hybrid multi-objectives evolutionary algorithm (hybrid MOEA). Reynolds-averaged Navier-Stokes equations with shear stress transport turbulence model are discretized by finite volume approximations and solved on hexahedral grids for flow analyses. Latin hypercube sampling of design of experiments is used to generate design points within the selected design space. Two objectives, i.e., isentropic efficiency and total pressure ratio are selected with four design variables defining impeller hub and shroud contours in meridional contours to optimize the system. Non-dominated Sorting of Genetic Algorithm (NSGA-II) with ε -constraint strategy for local search coupled with surrogate model is used for multi-objective optimization. The surrogate model, Radial Basis Neural Network is trained on the numerical solutions by carrying out leave-one-out cross-validation for the data set. The trade-off between the two objectives has been found out and discussed in light of the Pareto-optimal solutions. The optimization results show that isentropic efficiencies and total pressure ratios of the cluster points at the Pareto-optimal solutions are enhanced by multi-objective optimization.

KW - Centrifugal compressor impeller

KW - Design optimization

KW - Evolutionary algorithm

KW - Pareto-optimal

UR - http://www.scopus.com/inward/record.url?scp=77953821776&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77953821776&partnerID=8YFLogxK

U2 - 10.1115/FEDSM2009-78486

DO - 10.1115/FEDSM2009-78486

M3 - Conference contribution

SN - 9780791843734

VL - 2

SP - 185

EP - 191

BT - Proceedings of the ASME Fluids Engineering Division Summer Conference 2009, FEDSM2009

ER -