Fine tuning straightening process using genetic algorithms and finite element methods

Abstract The process of straightening steel sections is used not only to actually straighten the product but also to reduce its internal residual stresses. Fine tuning this process within an industrial plant is complicated because of the time needed for conducting the tests and the difficulties in measuring the final residual stresses. This paper presents a methodology based on genetic algorithms and finite element analysis that seeks the best position of the rollers to produce a straightened product with the minimum amount of residual stresses. The process consists of simulating multiple roller positions using a previously validated finite element model and analysing the resulting residual stresses. Genetic programming is used to choose the best solutions that will give rise to the next generation of individuals. For several generations, the system combines a series of optimum solutions in which residual solutions are minimised. The best solutions obtained enable the rollers to be positioned in a way that guarantees a good end quality for the product.

[1]  H. Maier,et al.  High temperature fatigue behavior and residual stress stability of laser-shock peened and deep rolled austenitic steel AISI 304 , 2004 .

[2]  Dorothea Heiss-Czedik,et al.  An Introduction to Genetic Algorithms. , 1997, Artificial Life.

[3]  Ac Pankaj,et al.  Analysis of end straightness of rail during manufacturing , 2005 .

[4]  J. J. Kalker,et al.  Rail quality and maintenance for modern railway operation : International Conference on Rail Quality and Maintenance for Modern Railway Operation, Delft, June 1992 , 1993 .

[5]  J. T. Maximov,et al.  A finite element simulation of the spherical mandrelling process of holes with cracks , 2006 .

[6]  F. J. Martinez-de-Pison,et al.  Improvement and optimisation of hot dip galvanising line using neural networks and genetic algorithms , 2006 .

[7]  G. G. Stokes.  "J." , 1890, The New Yale Book of Quotations.

[8]  Franz Dieter Fischer,et al.  Residual stress formation during the roller straightening of railway rails , 2001 .

[9]  Janusz Orkisz,et al.  Residual Stress in Rails , 1992 .

[10]  Jonas W. Ringsberg,et al.  Rolling contact fatigue analysis of rails inculding numerical simulations of the rail manufacturing process and repeated wheel-rail contact loads , 2003 .

[11]  Liguang Zhu,et al.  Thermomechanical analysis and optimisation for beam blank continuous casting , 2008 .

[12]  O. Orringer,et al.  Residual stress and its consequences on both sides of the wheel-rail interface , 1996 .

[13]  Franz Dieter Fischer,et al.  Residual stresses in rails due to roll straightening , 1998 .

[14]  Alpha Verónica Pernía Espinoza.  Use of the advanced numerical simulation techniques for the study and improvement of steel-profiles manufacturing processes: Uso de técnicas avanzadas de simulación numérica para el estudio y mejora de procesos de fabricación de perfiles de acero , 2007 .

[15]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[16]  S.M.H. Mirbagheri,et al.  Study on microstructural events during hot rod rolling of steels using mathematical modelling , 2008 .