Advertisement

The Solution of Systems of Nonlinear Algebraic Equations

  • Siegfried Selberherr

Abstract

The main result obtained in the preceding chapter is that discretization of the basic semiconductor equations yields a large system of nonlinear algebraic equations with the values of the dependent variables of the differential equations at discrete points as unknowns. For the considerations in this chapter we adopt the following nomenclature for the system of discretized equations:
$$\vec F(\vec w) = 0$$
(7-1)
$$\vec F = \left( \begin{gathered} \vec f_\psi (\vec w) \hfill \\ \vec f_n (\vec w) \hfill \\ \vec f_n (\vec w) \hfill \\ \end{gathered} \right)$$
(7-2)
$$\vec w = \left( \begin{gathered} {\vec \psi } \hfill \\ {\vec n} \hfill \\ {\vec p} \hfill \\ \end{gathered} \right)$$
(7-3)
\({\vec F}\) is a vector funetion of rank three which itself consists of the vector funetions \(\vec f_\psi \vec f_n \,and\,\vec f_p\). These vector funetions correspond to the discrete approximations for the Poisson equation and the continuity equations, respectively. The vector of unknowns \({\vec w}\) is also comprised by three vectors which are formed by the values of the electrostatic potential \({\vec \psi }\), electron concentration \({\vec n}\) and hole concentration \({\vec p}\) at discrete points of the Simulation geometry. We shall assume that the rank of all three vector funetions \(\vec f_\psi \vec f_n ,\vec f_p\) and the three vectors \(\vec \psi ,\vec n,\vec p\) equals n. This is not a necessary assumption but it will simplify the notation. It may well happen for practical applications that the rank of \({\vec \psi }\) differs from the rank of \({\vec n}\) and \({\vec p}\) (e.g., when the Laplaee equation is solved in an insulator). For our purpose, the scalar rank of \({\vec F}\) and \({\vec w}\) is 3·n.

Keywords

Iterative Method Jacobian Matrix Initial Guess Newton Method Convergence Property 
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. [7.1]
    Bank, R. E., Rose, D. J.: Parameter Selection for Newton-like Methods Applicable to Nonlinear Partial Differential Equations. SIAM J. Numer. Anal. 17, No. 6, 806–822 (1980).CrossRefGoogle Scholar
  2. [7.2]
    Bank, R. E., Rose, D. J.: Global Approximate Newton Methods. Numer. Math. 37, 279–295 (1981).CrossRefGoogle Scholar
  3. [7.3]
    Brown, G. W., Lindsay, B. W.: The Numerical Solution of Poisson’s Equation for Two-Dimensional Semiconductor Devices. Solid-State Electron. 19, 991–992 (1976).CrossRefGoogle Scholar
  4. [7.4]
    Curtis, A. R., Reid, J. K.: The Choice of Step Length when Using Differences to Approximate Jacobian Matrices. J. Inst. Math. Appl. 13, 121–126 (1974).Google Scholar
  5. [7.5]
    Den Heijer, C., Polak, S. J., Schilders, W. H. A.: A Continuation Method for the Calculation of Potentials and Currents in Semiconductors. Proc. NASECODE II Conf, pp. 182–187. Dublin: Boole Press 1981.Google Scholar
  6. [7.6]
    Deuflhard, P.: A Modified Newton Method for the Solution of Ill-Conditioned Systems of Nonlinear Equations with Application to Multiple Shooting. Numer. Math. 22, 289–315 (1974).CrossRefGoogle Scholar
  7. [7.7]
    Fichtner, W., Rose, D. J.: On the Numerical Solution of Nonlinear Elliptic PDEs Arising from Semiconductor Device Modeling. Report 80–2111-12, Bell Laboratories, 1980.Google Scholar
  8. [7.8]
    Fichtner, W., Rose, D. J.: On the Numerical Solution of Nonlinear Elliptic PDEs Arising from Semiconductor Device Modeling. In: Elliptic Problem Solvers, pp. 277–284. New York: Academic Press 1981.Google Scholar
  9. [7.9]
    Franz, A. F., Franz, G. A., Selberherr, S., Markowich, P: The Influence of Various Mobility Models on the Iteration Process and Solution of the Basic Semiconductor Equations. Proc. NASECODE III Conf., pp. 117–121. Dublin: Boole Press 1983.Google Scholar
  10. [7.10]
    Franz, A. F., Franz, G. A., Selberherr, S., Ringhofer, C., Markowich, P.: Finite Boxes — A Generalization of the Finite Difference Method Suitable for Semiconductor Device Simulation. IEEE Trans. Electron Devices ED-30, No. 9, 1070–1082 (1983).CrossRefGoogle Scholar
  11. [7.11]
    Greenfield, J. A., Hansen, S. E., Dutton, R. W.: Two-Dimensional Analysis for Device Modeling. Report G-201–7, Stanford University, 1980.Google Scholar
  12. [7.12]
    Gummel, H. K.: A Self-Consistent Iterative Scheme for One-Dimensional Steady State Transistor Calculations. IEEE Trans. Electron Devices ED-11, 455–465 (1964).CrossRefGoogle Scholar
  13. [7.13]
    Markowich, P. A.: A Singular Perturbation Analysis of the Fundamental Semiconductor Device Equations. Habilitation, Technische Universität Wien, 1983.Google Scholar
  14. [7.14]
    Meyer, G. H.: On Solving Nonlinear Equations with a One-Parameter Operator Imbedding. SIAM J. Numer. Anal. 5, 739–752 (1968).CrossRefGoogle Scholar
  15. [7.15]
    Mock, M. S.: Asymptotic Behaviour of Solutions of Transport Equations for Semiconductor Devices. J. Math. Anal. Appl. 49, 215–255 (1975).CrossRefGoogle Scholar
  16. [7.16]
    Mock, M. S.: Analysis of Mathematieal Models of Semiconductor Devices. Dublin: Boole Press 1983.Google Scholar
  17. [7.17]
    Ortega, J. M., Rheinboldt, W. C.: Iterative Solution of Nonlinear Equations in Several Variables. New York: Academic Press 1970.Google Scholar
  18. [7.18]
    Richter, S. L., DeCarlo, R. A.: Continuation Methods: Theory and Applications. IEEE Trans. Circuits and Systems CAS-30, No. 6, 347–352 (1983).Google Scholar
  19. [7.19]
    Schütz, A., Selberherr, S., Pötzl, H. W.: A Two-Dimensional Model of the Avalanche Effect in MOS Transistors. Solid-State Electron. 25, 177–183 (1982).CrossRefGoogle Scholar
  20. [7.20]
    Sherman, A. H.: On Newton-Iterative Methods for the Solution of Systems of Nonlinear Equations. SIAM J. Numer. Anal. 15, 755–771 (1978).CrossRefGoogle Scholar
  21. [7.21]
    Sutherland, A. D.: On the Use of Overrelaxation in Conjunction with Gummers Algorithm to Speed the Convergence in a Two-Dimensional Computer Model for MOSFET’s. IEEE Trans. Electron Devices ED-27, 1297–1298 (1980).CrossRefGoogle Scholar
  22. [7.22]
    Wang, C. T.: A Re-Extrapolation Technique in Newton-Sor Computer Simulation of Semiconductor Devices. Solid-State Electron. 25, No. 11, 1083–1087 (1982).CrossRefGoogle Scholar
  23. [7.23]
    Wolfe, P.: Checking the Calculation of Gradients. ACM Trans. Mathematical Software 8, No. 4, 337–343 (1982).CrossRefGoogle Scholar
  24. [7.24]
    Zaluska, E. J., Dubock, P. A., Kemhadhan, H. A.: Practical 2-Dimensional Bipolar-Transistor-Analysis Algorithm. Electron. Lett. 9, 599–600 (1973).CrossRefGoogle Scholar
  25. [7.25]
    Zarantonello, E. H.: Solving Functional Equations by Contractive Averaging. Report 160, MRC, University of Wisconsin, 1960.Google Scholar
  26. [7.26]
    Zienkiewicz, O. C.: The Finite Element Method. London: McGraw-Hill 1977.Google Scholar

Copyright information

© Springer-Verlag/Wien 1984

Authors and Affiliations

  • Siegfried Selberherr
    • 1
  1. 1.Institut für Allgemeine Elektrotechnik und ElektronikTechnische Universität WienAustria

Personalised recommendations