Adaptive Experimental Design for Construction of Response Surface Approximations

Sequential Approximate Optimization (SAO) is a class of methods available for the multidisciplinary design optimization (MDO) of complex systems that are composed of several disciplines coupled together. One of the approaches used for SAO, is based on a quadratic response surface approximation, where zero and first order information are required. In these methods, designers must generate and query a database of order O(n2) in order to compute the second order terms of the quadratic response surface approximation. As the number of design variables grows, the computational cost of generating the required database becomes a concern. In this paper, we present an new approach in which we require just O(ri) parameters for constructing a second order approximation. This is accomplished by transforming the matrix of second order terms into the canonical form. The method periodically requires an order O(n2) update of the second order approximation to maintain accuracy. Results show that the proposed approach is practical and convenient for engineering design problems by dramatically reducing the total number of calls to the simulation tools.

[1]  Esther Seiden,et al.  On Orthogonal Arrays , 1966 .

[2]  R. C. St. John,et al.  D-Optimality for Regression Designs: A Review , 1975 .

[3]  J. Sobieszczanski-Sobieski,et al.  A linear decomposition method for large optimization problems. Blueprint for development , 1982 .

[4]  George E. P. Box,et al.  Empirical Model‐Building and Response Surfaces , 1988 .

[5]  Jaroslaw Sobieszczanski-Sobieski,et al.  Optimization by decomposition: A step from hierarchic to non-hierarchic systems , 1989 .

[6]  J. Sacks,et al.  A system for quality improvement via computer experiments , 1991 .

[7]  J. Sobieszczanski-Sobieski,et al.  Sensitivity of control-augmented structure obtained by a system decomposition method , 1991 .

[8]  G. A. Gabriele,et al.  Improved coordination in nonhierarchic system optimization , 1993 .

[9]  Boxin Tang.  Orthogonal Array-Based Latin Hypercubes , 1993 .

[10]  J. Renaud,et al.  Approximation in nonhierarchic system optimization , 1994 .

[11]  Bernard Grossman,et al.  Noisy Aerodynamic Response and Smooth Approximations in HSCT Design , 1994 .

[12]  Douglas C. Montgomery,et al.  Response Surface Methodology: Process and Product Optimization Using Designed Experiments , 1995 .

[13]  Layne T. Watson,et al.  Dependence of optimal structural weight on aerodynamic shape for a High Speed Civil Transport , 1996 .

[14]  R. H. Myers,et al.  Response Surface Methodology: Process and Product Optimization Using Designed Experiments , 1995 .

[15]  Raphael T. Haftka,et al.  Construction of response surfaces for design optimization applications , 1996 .

[16]  Robert Lewis,et al.  A trust region framework for managing approximation models in engineering optimization , 1996 .

[17]  Farrokh Mistree,et al.  Statistical experimentation for affordable concurrent design , 1996 .

[18]  F. Jose,et al.  Convergence of Trust Region Augmented Lagrangian Methods Using Variable Fidelity Approximation Data , 1997 .

[19]  Timothy W. Simpson,et al.  On the Use of Statistics in Design and the Implications for Deterministic Computer Experiments , 1997 .

[20]  N. M. Alexandrov,et al.  A trust-region framework for managing the use of approximation models in optimization , 1997 .

[21]  B. Wujek,et al.  A new adaptive move-limit management strategy for approximate optimization. II , 1998 .

[22]  L. Watson,et al.  Trust Region Augmented Lagrangian Methods for Sequential Response Surface Approximation and Optimization , 1998 .

[23]  J. Renaud,et al.  New Adaptive Move-Limit Management Strategy for Approximate Optimization, Part 2 , 1998 .

[24]  M. Natalia.  On Managing the Use of Surrogates in General Nonlinear Optimization and MDO , 1998 .

[25]  T. W. Layne,et al.  A Comparison of Approximation Modeling Techniques: Polynomial Versus Interpolating Models , 1998 .

[26]  John E. Renaud,et al.  CONSTRUCTING VARIABLE FIDELITY RESPONSE SURFACE APPROXIMATIONS IN THE USABLE FEASIBLE REGION , 2000 .

[27]  William Crossley,et al.  Investigating response surface approaches for drag optimization of a subsonic turbofan nacelle , 2000 .

[28]  T. Simpson,et al.  Comparative studies of metamodeling techniques under multiple modeling criteria , 2000 .

[29]  J. Renaud,et al.  Decoupling the Design Sampling Region From the Trust Region in Approximate Optimization , 2000, Aerospace.

[30]  J. F. Rodríguez,et al.  Sequential approximate optimization using variable fidelity response surface approximations , 2000 .

[31]  Thomas J. Santner,et al.  Design and analysis of computer experiments , 1998 .