Robust Design of UOE Forming Process Based on Support Vector Machine

• Published in: Materials science forum Vol. 817; pp. 523 - 530
• Main Authors: Li, Da Yong, Ren, Qiang, Peng, Ying Hong, Wu, Guang Han, Zou, Tian Xia
• Format: Journal Article
• Language: English
• Published: Pfaffikon Trans Tech Publications Ltd 01.04.2015
• Subjects: Fluctuation; Forming; Optimization; Quality; Sheet metal; Statistics; Steels; Support vector machines; UOE Forming; Support Vector Machine; Robust Design; Ovality; Material Fluctuation
• ISSN: 0255-5476; 1662-9752; 1662-9752
• DOI: 10.4028/www.scientific.net/MSF.817.523

Fluctuations in material properties of the incoming steel for UOE forming process are widespread. According to the statistics, the fluctuation range of the yield strength of the same grade pipeline steel is around 80MPa. Robust optimization methods have been widely applied in sheet metal forming area. In this paper, experiments were conducted to investigate how a stochastic material behavior of noise factors affected UOE forming quality. Robust design models integrated with response surface method for UOE forming process were established to minimize impact of the variations and improve the qualified rate of UOE pipe ovality. Support vector machine in both classification and regression was adopted to map the relation between input process parameters and forming qualities. The deterministic and robust optimization results are presented and compared, demonstrating increased process robustness and decreased number of product rejects by application of the robust optimization approach.