Abstract
This paper deals with numerical methods for solving linear variational inequalities on an arbitrary closed convex subsetC of ℝn. Although there were numerous iterations studied for the caseC=ℝ n+ , few were proposed for the case whenC is a closed convex subset. The essential difficulty in this case is the nonlinearities ofC's boundaries. In this paper iteration processes are designed for solving linear variational inequalities on an arbitrary closed convex subsetC. In our algorithms the computation of a linear variational inequality is decomposed into a sequence of problems of projecting a vector to the closed convex subsetC, which are computable as long as the equations describing the boundaries are given. In particular, using our iterations one can easily compute a solution whenC is one of the common closed convex subsets such as cube, ball, ellipsoid, etc. The non-accurate iteration, the estimate of the solutions on unbounded domains and the theory of approximating the boundaries are also established. Moreover, a necessary and sufficient condition is given for a vector to be an approximate solution. Finally, some numerical examples are presented, which show that the designed algorithms are effective and efficient. The exposition of this paper is self-contained.
Zusammenfassung
Diese Arbeit behandelt numerische Methoden zur Lösung linearer Variations-Ungleichungen auf einer abgeschlossenen konvexen Teilmenge des ℝn. Es gibt zwar zahlreiche Iterationsverfahren für den FallC=ℝ n+ für den Fall einer beliebigen abgeschlossenen konvexen TeilmengeC wurde aber nur wenig vorgeschlagen. Die wesentliche Schwierigkeit in diesem Fall liegt in den Nichtlinearitäten des Randes vonC. In dieser Arbeit werden Iterationsverfahren zur Lösung linearer Variations-Ungleichungen auf einer beliebigen abgeschlossenen konvexen TeilmengeC entwickelt. In unseren Algorithmen wird die Berechnung einer linearen Variations-Ungleichung zerlegt in eine Folge von Projektionen eines Vektors auf die abgeschlossene konvexe TeilmengeC, die berechnet werden können, solange die Bestimmungsgleichungen des Randes gegeben sind. Insbesondere kann mit unseren Iterationsverfahren leicht eine Lösung berechnet werden fürC als Würfel, Kugel, Ellipsoid etc. Außerdem werden Näherungs-Iterationen, Abschätzung der Lösungen für unbeschränkte Bereiche und die Theorie der Randapproximation untersucht. Weiters wird eine notwendige und hinreichende Bedingung dafür angegeben, daß ein Vektor eine Näherungslösung ist. Schließlich werden einige numerische Beispiele präsentiert, die zeigen, daß die vorgestellten Algorithmen effektiv und effizient sind.
Similar content being viewed by others
References
Allen, G.: Variational inequalities, complementarity problems, and duality theorems. J. Math. Anal. Appl.58, 1–10 (1977).
Ahn, B. H.: Solution of non-symmetric linear complementarity problems by iterative methods. JOTA33, 175–185 (1981).
Cottle, R. W., Golub, G., Sacher, R. S.: On the solution of large structured linear complementarity problems: the block partitioned case. Appl. Math. Optim.4, 347–364 (1978).
Cryer, C. W.: The solution of quadratic programming problems using systematic over relaxation, SIAM J. Control9, 385–392 (1971).
Dafermos, S.: Traffic equilibrium and variational inequalities. Transport. Sci.14, 42–54 (1980).
Dafermos, S.: An iterative scheme for variational inequalities. Math. Prog.26, 40–47 (1983).
Fridman, V. M., Chernina, V. S.: An iteration process for the solution of finite-dimensional contact problem. USSR Comp. Math. Math. Phys.8, 210–214 (1967).
Glowinski, R. Lions, J. L. Tremolieres, R.: Numerical analysis of variational inequalities. Amsterdam: North-Holland 1981.
Isac, G.: Fixed-point theory, coincidence equations on convex cones, and complementarity problems. Contemp. Math.72, 139–155 (1988).
Kinderlehrer, D., Stampacchia, G.: An introduction to variational inequalities and their applications. New York: Academic Press 1980.
Kostreva, M. M.: Direct algorithms for complementarity problems. Ph.D. Dissertation, Rensselaer Polytechnic Institute, Troy, New York, 1976.
Kostreva, M. M.: Block pivot methods for solving the complementarity problems. Lin. Alg. Appl.21, 207–215 (1978).
Mangasarian, O. L.: Solution of symmetric linear complementarity problems by iterative methods. JOTA22, 465–485 (1977).
Moré, J. J.: Coercivity conditions for nonlinear complementarity problems. SIAM Rev.16, 1–16 (1974).
Ostrowski, A. M.: Solution of equations and systems of equations, 2nd edn. New York: Academic Press 1969.
Ortega, J. M.: Numerical analysis, a second course. New York: Academic Press 1972.
Pang, J. S. Chan, D.: Iterative methods for solving variational and complementarity problems. Math. Prog.24, 284–314 (1982).
Raimondi, A. A., Boyd, J.: A solution for finite journal bearing and its application to analysis and design, III. Trans. Am. Soc. Lubr. Eng.1, 194–209 (1958).
You, Z. Y., Zhu, S. Q.: A constructive approach for proving the existence of solutions of variational inequalities. Kexue Tongbao31 (1986).
You, Z. Y., Zhu, S. Q.: On global convergence and approximate iteration of the linear approximation method for solving variational inequalities J. Comput.4 (1986).
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Zhu, S.Q. Iterations for solving linear variational inequalities on domains with nonlinear boundaries. Computing 54, 251–272 (1995). https://doi.org/10.1007/BF02253616
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF02253616