Advertisement

Numerische Mathematik

, Volume 55, Issue 6, pp 619–632 | Cite as

Newton's method for gradient equations based upon the fixed point map: Convergence and complexity study

  • Joseph W. Jerome
Article

Summary

An approximate Newton method, based upon the fixed point mapT, is introduced for scalar gradient equations. Although the exact Newton method coincides for such scalar equations with the standard iteration, the structure of the fixed point map provides a way of defining anR-quadratically convergent finite element iteration in the spirit of the Kantorovich theory. The loss of derivatives phenomenon, typically experienced in approximate Newton methods, is thereby avoided. It is found that two grid parameters are sssential,h and\(\bar h \approx h^2 \). The latter is used to calculate the approximate residual, and is isolated as a fractional step; it is equivalent to the approximation ofT. The former is used to calculate the Newton increment, and this is equivalent to the approximation ofT′. The complexity of the finite element computation for the Newton increment is shown to be of optimal order, via the Vituškin inequality relating metric entropy andn-widths.

Subject Classifications

AMS(MOS):65H10 CR:G1.5 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Aziz, A.K.: The mathematical foundations of the finite element method with applications to partial differential equations. London: Academic Press 1972Google Scholar
  2. 2.
    Bank, R.E., Rose, D.J.: Parameter selection for Newton-like methods applicable to nonlinear partial differential equations. SIAM J. Numer. Anal.17, 806–822 (1980)CrossRefGoogle Scholar
  3. 3.
    Gilbarg, D., Trudinger, N.S.: Elliptic partial differential equations of second order. Heidelberg Berlin New York: Springer 1977Google Scholar
  4. 4.
    Hewitt, E., Stromberg, K.: Real and abstract analysis. New York: Springer 1965Google Scholar
  5. 5.
    Jerome, J.W.: Approximation of nonlinear evolution systems. New York London: Academic Press 1983Google Scholar
  6. 6.
    Jerome, J.W.: An adaptive Newton algorithm based on numerical inversion: regularization as postconditioner. Numer. Math.47, 123–138 (1985)Google Scholar
  7. 7.
    Jerome, J.W.: Approximate Newton methods and homotopy for stationary operator equations. Constructive Approximation1, 271–285 (1985)Google Scholar
  8. 8.
    Lorentz, G.G.: Metric entropy and approximation. Bull. Amer. Math. Soc.72, 903–937 (1966)Google Scholar
  9. 9.
    Lorentz, G.G.: Approximation of functions. Holt, Rinehart and Winston 1966Google Scholar
  10. 10.
    Mostefai, A.: Évaluations de l'ε-entropy dans les espaces de Sobolev. Dissertation, University of Algier 1970Google Scholar
  11. 11.
    Nirenberg, L.: Topics in nonlinear functional analysis. Courant Institute of Mathematical Sciences, New York University 1973–1974Google Scholar
  12. 12.
    Strang, G., Fix, G.J.: An analysis of the finite element method. Englewood Cliffs, NJ: Prentice-Hall 1973Google Scholar
  13. 13.
    Taylor, A.E.: Introduction to Functional Analysis New York: Wiley 1958Google Scholar

Copyright information

© Springer-Verlag 1989

Authors and Affiliations

  • Joseph W. Jerome
    • 1
  1. 1.Department of MathematicsNorthwestern UniversityEvanstonUSA

Personalised recommendations