The Design of Open Engineering Systems Lab

University at Buffalo - The State University of New York

  • Increase font size
  • Default font size
  • Decrease font size

Development of a Kriging Based Surrogate Approximation Method for Large Scale Systems

Research Area: Research Publication Year: 1999
Type of Publication: Technical Report Keywords: Response Surface Methodology, Approximation Methods, Kriging, Design of Experiments
Authors: Srivastava, Amit; Hacker, Kurt; Lewis, Kemper; Simpson, Timothy
In the present world of advancing technology, designers are faced with designing increasingly complex engineering systems. Despite a steady increase in computing power, the complexity of these engineering analyses seems to advance at the same rate. In addition, the design of complex engineering systems involves the integration of multiple disciplines and the resolution of conflicting objectives. Traditional parametric design analysis is inadequate to analyze these large-scale systems because of its computational inefficiency; therefore, a departure from the traditional parametric design approach is required. Approximation techniques may be applied to build computationally inexpensive surrogate models for large-scale systems to replace expensive-to-run computer analysis codes. Response surface models are frequently utilized to construct surrogate approximations; however, they may be inefficient for systems having with a large number of design variables. Kriging, an alternative method for creating surrogate models, is applied in this work to construct approximations of computationally expensive computer analyses for a large-scale system. Comparisons between response surfaces and kriging are made based on the results of a case study, the High Speed Civil Transport (HSCT) approximation challenge. In this problem, 2490 analysis points are already generated, and the challenge is to choose 500 or fewer and construct an accurate approximation. Since the analysis points are already chosen, a modified design of experiments technique is needed to select the appropriate sample points. In this paper, a number of approaches to handling this problem are presented, and the results are compared against previous work
International Journal for Product and Process Improvement
Full text: askh_2000.pdf