Advertisement

Structural optimization

, Volume 9, Issue 2, pp 76–82 | Cite as

Automatic differentiation as a tool in engineering design

  • J. -F. M. Barthelemy
  • L. E. Hall
Technical Papers

Abstract

Automatic Differentiation (AD) is a tool that systematically implements the chain rule of differentiation to obtain the derivatives of functions calculated by computer programs. In this paper, it is assessed as a tool for engineering design. The paper discusses the forward and reverse modes of AD, their computing requirements, as well as approaches to implementing AD. It continues with the application of two different tools to two medium-size structural analysis problems to generate sensitivity information typically necessary in an optimization or design situation. The paper concludes with the observation that AD is to be preferred to finite differencing in most cases, as long as sufficient computer storage is available; in some instances, AD may be the alternative to consider in lieu of analytical sensitivity analysis.

Keywords

Analytical Sensitivity Civil Engineer Structural Analysis Computer Program Finite Difference 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Barthelemy, J.-F.M.; Hall, L.E. 1992: Automatic differentiation as a tool for engineering design.Proc. 4th AIAA/Air Force/NASA/OAI Symp. on Multidisciplinary Analysis and Optimization (held in Cleveland, OH)Google Scholar
  2. Barthelemy, J.-F.M.; Riley, M.F. 1988: Improved multilevel optimization approach for the design of complex engineering systems.AIAA J. 26, 353–360Google Scholar
  3. Barthelemy, J.-F.M.; Wrenn, G.A.; Dovi, A.R.; Coen, P.G. 1992: Integrating aerodynamics and structures in the minimum weight design of a supersonic transport wing.AIAA Paper 92-2372, presented at 33rd AIAA/ASME/AHS/ASC Structures, Structural Dynamics and Materials Conf. (held in Dallas, TX)Google Scholar
  4. Bathe, K.-J.; Wilson, E.L. 1976:Numerical methods in finite element analysis. New York: Prentice-HallGoogle Scholar
  5. Bischof, C.; Carle, A.; Corliss, G.; Griewank, A.; Hovland, P. 1991: ADI-FOR-Generating derivative codes for Fortran programs.Argonne Preprint, MCS-P 263-0991 Google Scholar
  6. Bischof, C.; Green, L.; Haigler, K.; Knauff, T. 1994: Parallel calculation of sensitivity dervatives for aircraft design using automatic differentiation.AIAA 94-4261. (5th AIAA/USAF/NASA/ISSMO Symp. on Multidisciplinary Analysis and Optimization, Panama City Beach, FL)Google Scholar
  7. Callahan, D.; Cooper, K.; Hood, R.T.; Kennedy, K.; Torcson, L.M. 1988: ParaScope: a parallel programming environment.Int. J. Supercomputer Applications 2, 273–285Google Scholar
  8. Carle, A.; Green, L.; Bischof, C.; Newman, P. 1994: Applications of automatic differentiation in CFD.AIAA 94-2197. (25th AIAA Fluid Dynamics Conf., Colorado Springs, CO)Google Scholar
  9. Garcia, O. 1991: A system for the differentiation of Fortran codes and an application to parameter estimation in forest growth models. In: Griewank, A.; Corliss, G.F. (eds.)Automatic differentiation of algorithms: theory, implementation, and application, pp. 273–285. Philadelphia: SIAMGoogle Scholar
  10. Giles, G.L. 1986: Equivalent plate analysis of aircraft wing box structures with general planform geometry.J. Aircraft 23, 859–864Google Scholar
  11. Giles, G.L. 1989: Further generalization of an equivalent plate representation for aircraft structural analysis.J. Aircraft 26, 67–74Google Scholar
  12. Griewank, A. 1991a: The chain rule revisited in scientific computing.SIAM News, p. 20Google Scholar
  13. Griewank, A. 1991b: The chain rule revisited in scientific computing, part II.SIAM News, p. 8Google Scholar
  14. Griewank, A. 1991c: Automatic evaluation of first and higher derivative vectors. In: Schneider; Dupper; Troger (eds.)Analysis, algorithms, applications (Proc. Conf. on Bifurcation and Chaos, held in Wurzburg, 1990), pp. 124–137. Basel: BirkhäuserGoogle Scholar
  15. Griewank, A. 1992: Achieving logarithmic growth of temporal and spatial complexity in reverse automatic differentiation.Optimization Methods and Software 1, 35–54Google Scholar
  16. Griewank, A.; Corliss, G.F. (eds.) 1991:Automatic differentiation of algorithms: theory, implementation, and application. Philadelphia: SIAMGoogle Scholar
  17. Griewank, A; Reese, S. 1991: On the calculation of Jacobian matrices by the Markowitz rule. In: Griewank, A.; Corliss, G.F. (eds.)Automatic differentiation of algorithms: theory, implementation, and application, pp. 126–135. Philadelphia: SIAMGoogle Scholar
  18. Haftka, R.T.; Gurdal, Z. 1992:Elements of structural optimization, 3rd edition. Dordrecht: KluwerGoogle Scholar
  19. Haug, E.J.; Arora, J.S. 1979:Applied optimal design. New York: J. Wiley and SonsGoogle Scholar
  20. Horwedel, J.E. 1991a: GRESS version 2.0 user's manual.Report ORNL/TM-11951 Google Scholar
  21. Horwedel, J.E. 1991b: GRESS, a preprocessor for sensitivity analysis of Fortran programs. In: Griewank, A.; Corliss, G.F. (eds.)Automatic differentiation of algorithms: theory, implementation, and application, pp. 243–250. Philadelphia: SIAMGoogle Scholar
  22. Iri, M. 1984: Simultaneous computation of functions, partial derivatives and estimates of rounding errors — complexity and practicality.Japan J. Appl. Math. 1, 223–252Google Scholar
  23. Iri, M. 1988: Automatic computation of partial derivatives and rounding error estimates with applications to large-scale systems of nonlinear equations.J. Comp. Appl. Math. 24, 365–392Google Scholar
  24. Iri, M. 1991: History of automatic differentiation and rounding error estimation. In: Griewank, A.; Corliss, G.F. (eds.)Automatic differentiation of algorithms: theory, implementation, and application, pp. 3–16. Philadelphia: SIAMGoogle Scholar
  25. Juedes, D.W. 1991: A taxonomy of automatic differentiation tools. In: Griewank, A.; Corliss, G.F. (eds.) Automatic differentiation of algorithms: theory, implementation, and application, pp. 315–329. Philadelphia: SIAMGoogle Scholar
  26. Rall, L.B. 1981:Automatic differentiation: techniques and applications. Lecture Notes in Computer Science No. 120. Berlin, Heidelberg, New York: SpringerGoogle Scholar
  27. Talagrand, O. 1991: The use of adjoint equations in numerical modeling of the atmospheric circulation. In: Griewank, A.; Corliss, G.F. (eds.) Automatic differentiation of algorithms: theory, implementation, and application, pp. 169–180. Philadelphia: SIAMGoogle Scholar
  28. Thacker, W.M. 1991: Automatic differentiation from an oceanographer's perspective. In: Griewank, A.; Corliss, G.F. (eds.) Automatic differentiation of algorithms: theory, implementation, and application, pp. 191–201. Philadelphia: SIAMGoogle Scholar
  29. Worley, B.A. 1991: Experience with the forward and reverse mode of GRESS in contaminant transport modeling and other applications. In: Griewank, A.; Corliss, G.F. (eds.) Automatic differentiation of algorithms: theory, implementation, and application, pp. 307–314. Philadelphia: SIAMGoogle Scholar

Copyright information

© Springer-Verlag 1995

Authors and Affiliations

  • J. -F. M. Barthelemy
    • 1
  • L. E. Hall
    • 2
  1. 1.NASA Langley Research CenterHamptonUSA
  2. 2.Computer Science CorporationHamptonUSA

Personalised recommendations