Sequential metamodelling application to improve porthole die design

Abstract The conventional trial—error method and empirical approaches are time consuming for the design of complex shaped products like porthole dies. These methods are associated with higher production costs, lower efficiency, and design inaccuracies pertaining to ambiguity and uncertainty. Owing to these deficiencies, there is a need for a more reliable and better design approach. In this article, a Kriging metamodel and differential evolution-based random simulation design methodology is proposed in order to reduce the cognitive load on the designer. The proposed methodology helps in selecting the set of parameters to be used to perform a simulation such that an improved design is delivered with reduced time and effort. The combination of the input parameters and their probable effect on the final design is evaluated and provided to the designer beforehand. This information, when juxtaposed with the designer's knowledge, gives greater opportunities to produce an optimal design. The sequential sampling strategy is used to select this set of parameters. It depends on the confidence value: a function of the design variables and the desired performance parameter. A Kriging metamodel is employed for modelling a random simulation of porthole extrusion with different influencing parameters. It converts the black-box region (no information zone in the design space) into a grey region (design space with some available information). Differential evolution (an evolutionary algorithm) is used to search for the black-box region carrying the least information in the design space. Three-dimensional extrusion of aluminium is considered in this article for designing a porthole die. The effect of variation of the design parameters is described, sampling points are generated, and the effective set of parameters are evaluated. The results obtained with the proposed sequential methodology are comparable with the simulation results presented in the literature for porthole extrusion.

[1]  Boris Štok,et al.  Tool design optimization in extrusion processes , 1998 .

[2]  Chung-Gil Kang,et al.  Effects of chamber shapes of porthole die on elastic deformation and extrusion process in condenser tube extrusion , 2005 .

[3]  Cristina H. Amon,et al.  An engineering design methodology with multistage Bayesian surrogates and optimal sampling , 1996 .

[4]  Donald R. Jones,et al.  Efficient Global Optimization of Expensive Black-Box Functions , 1998, J. Glob. Optim..

[5]  M. Sasena,et al.  Exploration of Metamodeling Sampling Criteria for Constrained Global Optimization , 2002 .

[6]  Thomas J. Santner,et al.  The Design and Analysis of Computer Experiments , 2003, Springer Series in Statistics.

[7]  Tadasu Murakami,et al.  Metal flow in porthole die extrusion of aluminium , 2002 .

[8]  Mark Jolly,et al.  Finite element modelling investigations upon the influence of pocket die designs on metal flow in aluminium extrusion: Part II. Effect of pocket geometry configurations on metal flow , 2003 .

[9]  Jack P. C. Kleijnen,et al.  Kriging for interpolation in random simulation , 2003, J. Oper. Res. Soc..

[10]  J. H. Starnes,et al.  Construction of Response Surface Approximations for Design Optimization , 1998 .

[11]  Rainer Storn,et al.  Differential Evolution – A Simple and Efficient Heuristic for global Optimization over Continuous Spaces , 1997, J. Glob. Optim..

[12]  Mark Jolly,et al.  Finite element investigations upon the influence of pocket die designs on metal flow in aluminium extrusion: Part I. Effect of pocket angle and volume on metal flow , 2003 .

[13]  Dong-Yol Yang,et al.  Investigation into the improvement of welding strength in three dimensional extrusion of tubes using porthole dies , 2002 .

[14]  Rong-Shean Lee,et al.  Optimal Die Design for Three-Dimensional Porthole Extrusion Using the Taguchi Method , 2006 .

[15]  Janez Brest,et al.  Self-Adapting Control Parameters in Differential Evolution: A Comparative Study on Numerical Benchmark Problems , 2006, IEEE Transactions on Evolutionary Computation.

[16]  Jack P. C. Kleijnen,et al.  Kriging Metamodeling in Simulation: A Review , 2007, Eur. J. Oper. Res..

[17]  L. Watson,et al.  Adaptive Experimental Design for Construction of Response Surface Approximations , 2001 .

[18]  T. Ozdemir,et al.  Fast parameter optimization using Kriging metamodeling [antenna EM modeling/simulation] , 2003, IEEE Antennas and Propagation Society International Symposium. Digest. Held in conjunction with: USNC/CNC/URSI North American Radio Sci. Meeting (Cat. No.03CH37450).

[19]  Rahul Rai,et al.  Q2S2: A New Methodology for Merging Quantitative and Qualitative Information in Experimental Design , 2008 .

[20]  Jouni Lampinen,et al.  A Fuzzy Adaptive Differential Evolution Algorithm , 2005, Soft Comput..

[21]  Jack P. C. Kleijnen,et al.  A Comment on Blanning's “Metamodel for Sensitivity Analysis: The Regression Metamodel in Simulation” , 1975 .

[22]  J. Huetink,et al.  An effective split of flow and die deformation calculations of aluminium extrusion , 1999 .

[23]  Hans-Jürgen Zimmermann,et al.  Fuzzy Set Theory - and Its Applications , 1985 .