The Role of Hierarchy in Response Surface Modeling of Wind Tunnel Data

This paper is intended as a tutorial introduction to certain aspects of response surface modeling, for the experimentalist who has started to explore these methods as a means of improving productivity and quality in wind tunnel testing and other aerospace applications. A brief review of the productivity advantages of response surface modeling in aerospace research is followed by a description of the advantages of a common coding scheme that scales and centers independent variables. The benefits of model term reduction are reviewed. A constraint on model term reduction with coded factors is described in some detail, which requires such models to be “well-formulated”, or “hierarchical”. Examples illustrate the consequences of ignoring this constraint. The implication for automated regression model reduction procedures is discussed, and some opinions formed from the author’s experience are offered on coding, model reduction, and hierarchy.

[1]  Derek J. Pike,et al.  Empirical Model‐building and Response Surfaces. , 1988 .

[2]  Richard DeLoach,et al.  National Transonic Facility Wall Pressure Calibration Using Modern Design of Experiments (Invited) , 2001 .

[3]  A. E. Hoerl,et al.  Ridge regression: biased estimation for nonorthogonal problems , 2000 .

[4]  Richard DeLoach,et al.  Applications of modern experiment design to wind tunnel testing at NASA Langley Research Center , 1998 .

[5]  Richard DeLoach,et al.  Uncertainty Analysis for the Evaluation of a Passive Runway Arresting System , 2009 .

[6]  Richard DeLoach.  A Factorial Data Rate and Dwell Time Experiment in the National Transonic Facility , 2000 .

[7]  Elizabeth A. Peck,et al.  Introduction to Linear Regression Analysis , 2001 .

[8]  N. Draper,et al.  Applied Regression Analysis , 1966 .

[9]  J. Peixoto.  Hierarchical Variable Selection in Polynomial Regression Models , 1987 .

[10]  Richard DeLoach,et al.  Using Modern Design of Experiments to Create a Surface Pressure Database From a Low Speed Wind Tunnel Test , 2004 .

[11]  Philip J. Brown.  Centering and Scaling in Ridge Regression , 1977 .

[12]  Richard DeLoach.  Tactical Defenses Against Systematic Variation in Wind Tunnel Testing , 2002 .

[13]  S. S. Srivastava,et al.  Correlation in Polynomial Regression , 1979 .

[14]  J. Peixoto.  A Property of Well-Formulated Polynomial Regression Models , 1990 .

[15]  R. A. Fisher,et al.  Statistical methods and scientific inference. , 1957 .

[16]  R. Fisher.  Statistical Methods for Research Workers , 1971 .

[17]  Ronald Christensen,et al.  Plane Answers to Complex Questions: The Theory of Linear Models , 1989 .

[18]  M. Kendall.  Statistical Methods for Research Workers , 1937, Nature.

[19]  Andrew D. Cutler,et al.  APPLICATION OF MODERN DESIGN OF EXPERIMENTS TO CARS THERMOMETRY IN A SUPERSONIC COMBUSTOR , 2002 .

[20]  Richard DeLoach,et al.  An Airbag Landing Dynamics Experiment Using the Modern Design of Experiments , 2009 .

[21]  R. DeLoach.  Improved Quality in Aerospace Testing Through the Modern Design of Experiments , 2000 .

[22]  Richard DeLoach,et al.  Tailoring wind tunnel data volume requirements through the formal design of experiments , 1998 .

[23]  Eugene A. Morelli,et al.  Response Surface Modeling Using Multivariate Orthogonal Functions , 2001 .

[24]  Richard DeLoach,et al.  Assessment of Response Surface Models Using Independent Confirmation Point Analysis , 2010 .

[25]  A. E. Hoerl,et al.  Ridge Regression: Applications to Nonorthogonal Problems , 1970 .

[26]  R. Snee,et al.  Ridge Regression in Practice , 1975 .

[27]  P. A. Parker.  Response surface methods for force balance calibration modeling , 2001, ICIASF 2001 Record, 19th International Congress on Instrumentation in Aerospace Simulation Facilities (Cat. No.01CH37215).

[28]  C. Radhakrishna Rao,et al.  Statistics and probability : essays in honor of C.R. Rao , 1983 .

[29]  David A. Belsley,et al.  Regression Analysis and its Application: A Data-Oriented Approach.@@@Applied Linear Regression.@@@Regression Diagnostics: Identifying Influential Data and Sources of Collinearity , 1981 .