Advertisement

Adaptive Simulation, the Adjoint State Method, and Optimization

  • Mark S. Cockenbach
  • William W. Symes
Part of the Lecture Notes in Computational Science and Engineering book series (LNCSE, volume 30)

Abstract

Adaptive grids in inverse and control problems can lead to computed objective functions that are nonsmooth, even though the underlying problem is well-behaved. This leads to the question of how to compute the linearization of the scheme—how should a nonsmooth function be differentiated? The C++ class afdtd uses automatic differentiation techniques to implement an abstract marching scheme in an object-oriented fashion, making it possible to use the resulting simulator to solve inverse or control problems. The class takes a complete specification of a single step of the scheme, and assembles from it a complete simulator, along with the linearized and adjoint simulations. The result is a (nonlinear) operator in the sense of the Hilbert Class Library, a C++ package for optimization. Moreover, afdtd supports locally “frozen” grids, allowing the implementation of an operator that is piecewise smooth in spite of the use of adaptivity.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    J. Borggaard and J. Burns. A PDE sensitivity equation method for optimal aerodynamic design. J. of Comp. Phys., pages 366–384, 1997.Google Scholar
  2. 2.
    George E. Forsythe, Michael A. Malcolm, and Cleve B. Moler. Computer Methods for Mathematical Computations. Prentice Hall, Englewood Cliffs, 1977.zbMATHGoogle Scholar
  3. 3.
    M. S. Gockenbach, M. J. Petro, and W. W. Symes. C++ classes for linking optimization with complex simulations. ACM Transactions on Mathematical Software, 25(2):191–212, 1999.CrossRefzbMATHGoogle Scholar
  4. 4.
    A. Griewank. Achieving logarithmic growth of temporal and spatial complexity in reverse automatic differentiation. Optimization Methods and Software, 1:35–54, 1992.CrossRefGoogle Scholar
  5. 5.
    Mark S. Gockenbach and William W. Symes. The Hilbert Class Library: a library of abstract C++ classes for optimization and inversion. Computers and Mathematics with Applications, 32:1–13, 1996.CrossRefzbMATHGoogle Scholar
  6. 6.
    S. Li, L. Petzold, and J. Hyman. Solution adapted mesh refinement and sensitivity analsysi for parabolic partial differential equation systems. Technical report, University of California — Santa Barbara, www.engineering.ucsb.eduj cse, 2001.Google Scholar
  7. 7.
    Lawrence F. Shampine and Marilyn Kay Gordon. Computer Solution of Ordinary Differential Equations. W. H. Freeman, San Francisco, 1975.zbMATHGoogle Scholar
  8. 8.
    R. Serban, S. Li, and L. Petzold. Adaptive algorithms for optimal control of time-dependent partial differential-algebraic equation systems. Technical report, University of California — Santa Barbara, www.engineering.ucsb.edu/~cse, 2001.Google Scholar
  9. 9.
    W.W. Symes and J. Meza. Conjugate residual methods for almost symmetric linear systems. J. Optimization Theory and Applications, 72(3):415–440, 1992.MathSciNetCrossRefzbMATHGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Mark S. Cockenbach
    • 1
  • William W. Symes
    • 2
  1. 1.Michigan Technological UniversityHoughton
  2. 2.Rice UniversityHouston

Personalised recommendations