Sage Journals: Discover world-class research

Abstract

In the modern design system of centrifugal impellers, numerical simulations by computational fluid dynamics (CFD) have played a vital role. Despite the rapid growth in computer capability and the maturation of CFD software, the optimization design of centrifugal impellers, which rely fully on numerical simulations, is currently impractical due to the large computational costs required. An approximation model can be applied to overcome this obstacle. This article investigates the application of the Kriging model in the optimization of a centrifugal impeller. First, the feasibility of the Kriging model is verified by a test function and a measure to improve the accuracy of solution is proposed. Following this, on the basis of the impeller's parametric modelling, the Kriging model method is applied to the optimization of a typical industrial centrifugal impeller. It is found that the isentropic efficiency at the design point is increased by 2.49 per cent, and a distinct improvement is made in the flow inside the impeller.

Keywords

Kriging model optimization design centrifugal impeller

Get full access to this article

View all access options for this article.

References

Myers

R. H.

Montgomery

D. C.

Response surface methodology: process and product optimization using designed experiments, 1995 (John Wiley, New York).

Montgomery

D. C.

Design and analysis of experiments, 5th edition, 2000 (John Wiley, New York).

Madsen

J. I.

Response surface techniques for diffuser shape optimization. AIAA J., 2000, 38 (9), 1512–1518.

Papila

Shyy

Griffin

L. W.

Dorney

D. J.

Shape optimization of supersonic turbines using response surface and neural network methods. AIAA 2001-1065, 2001.

Ahn

C.-S.

Kim

K.-Y.

Aerodynamic design optimization of a compressor rotor with Navier-Stokes analysis. Proc. Instn Mech. Engrs, Part A: J. Power and Energy, 2003, 217, 179–183.

Bonaiuti

Arnine

Ermini

Baldassarre

Analysis and optimization of transonic centrifugal compressor impellers using the design of experiments technique. ASME GT-2002-30619, 2002.

Sacks

Welch

W. J.

Mitchell

T. J.

Wynn

H. P.

Design and analysis of computer experiments. Stat. Sci., 1989, 4 (4), 409–435.

Sacks

Schiller

S. B.

Welch

W. J.

Design for computer experiments. Technometrics, 1989, 31 (1), 41–47.

Jin

R. C.

X. P.

Chen

The use of metamo-deling techniques for optimization under uncertainty., ASME DETC2001/DAC-21039, 2001.

10.

Simpson

T. W.

Mauery

T. M.

Korte

J. J.

Mistree

Comparison of response surface and Kriging models for multidisciplinary design optimization. AIAA 98-4755, 1998.

11.

Alvarezl

Design optimization based on genetic programming. PhD Thesis, University of Bradford, 2000.

12.

Madsen

J. I.

Design optimization of internal flow devices., PhD Thesis, Institute of Energy Technology and Institute of Mechanical Engineering, Aalborg University, 1998.

13.

Giunta

A. A.

Aircraft multidisciplinary design optimization using design of experiments theory and response surface modeling methods., PhD Thesis, Virginia Polytechnic Institute and State University, 1997.

14.

Fang

K. T.

Uniform design and uniform design table (in Chinese), Beijing Science Press, Beijing).

15.

Sasena

M. J.

Optimization of computer simulations via smoothing splines and Kriging metamodels., PhD Thesis, University of Michigan, 1998.

16.

Hathaway

M. D.

Chriss

R. M.

Strazisar

A. J.

Wood

J. R.

Laser anemometer measurements of the three-dimensional rotor flow field in the NASA-low speed centrifugal compressor., NASA technical paper 3527, 1995.

17.

Krain

Hoffman

Verification of an impeller design by laser measurements and 3D-viscous flow calculations. ASME 89-GT-159, 1989.

18.

Eisenlohr

Krain

Richter

F.-A.

Tiede

Investigation of the flow through a high pressure ratio centrifugal impeller. ASME GT-2002-30394, 2002.

Aerodynamic optimization design of centrifugal compressor's impeller with Kriging model

Abstract

Abstract

Keywords

Get full access to this article

References