Many-to-Many Regression Analysis of Loading Characteristic in Dieless Tube Hydroforming

Article Preview

Abstract:

Based on uniform design, finite element simulation of dieless tube hydroforming and double-screening stepwise regression method, a many-to-many polynomial regression equation set was established, whose independent variables are time t1 at which axial compression displacement from 0 reaches 25mm and time t2 at which internal pressure from 0 achieves 65MPa in 60 seconds, while dependent variables are hoop plastic strain s11 and radial plastic strain s22 in the midmost position of hydroformed tube. Analysis of partial correlation coefficient indicates that the effect of t2 on s11 is a little stronger than t1 and there is no interaction between t1 and t2 on s11, while s22 is influenced by t22 and t1t22. The many-to-many regression model not only quantificationally describes the action of loading factors on plastic strains but also reflects the inherent relevance between plastic strains, which is more useful than several one-to-many regression models to solve problems such as loading characteristic analysis, loading parameters design and optimization, etc.

You might also be interested in these eBooks

Info:

Periodical:

Pages:

534-537

Citation:

Online since:

March 2014

Export:

Price:

Permissions CCC:

Permissions PLS:

Сopyright:

© 2014 Trans Tech Publications Ltd. All Rights Reserved

Share:

Citation:

* - Corresponding Author

[1] K. T. Fang and C. X. Ma, Uniform Designs, Science Press, Beijing, (2001).

Google Scholar

[2] J. X. Qiu, S. H. Zhang, G. L. Li, Z. G Li and J. L. Zhang, Application of combining uniform design, neural network and genetic algorithm to optimize process parameters of internal high pressure forming, Journal of Plasticity Engineering, 2005, 12(4): 76-79.

Google Scholar

[3] H. X. Gao, Applied Multivariable Statistical Analysis, Peking University Press, Beijing, (2005).

Google Scholar

[4] Y. T. Zhang and K. T Fang, An Introduction to Multivariate Statistical Analysis, Science Press, Beijing, (2003).

Google Scholar