Calibration of a meanline centrifugal pump model using evolutionary algorithms

作者: E Benini , M Cenzon

DOI: 10.1243/09576509JPE742

关键词:

摘要: AbstractA one-dimensional meanline prediction code suitable for performance of small-to-medium size turbopumps (3–40 kW) is developed, which able to predict both on and off-design a turbopump given the loss diffusion system at design point. This model then calibrated against experimental data using an optimization technique. Due rapidity computations, single-objective evolutionary algorithm used calibration in order obtain wide exploration search space. Accuracy finally demonstrated by means comparison between predicted actual pump performance.

参考文章(3)
Raymond B. Furst, William D. Marscher, David Japikse, Centrifugal pump design and performance ,(1997)
Joseph P. Veres, Centrifugal and Axial Pump Design and Off-Design Performance Prediction In JHU. ,vol. 2, pp. 109- 122 ,(1995)
Andrea Toffolo, Ernesto Benini, Genetic diversity as an objective in multi-objective evolutionary algorithms Evolutionary Computation. ,vol. 11, pp. 151- 167 ,(2003) , 10.1162/106365603766646816