No label defined (Q3973672)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 16988
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | No label defined |
scientific article; zbMATH DE number 16988 |
Statements
26 June 1992
0 references
continuation methods
0 references
constrained nonlinear optimization
0 references
Karush-Kuhn- Tucker conditions
0 references
bifurcation
0 references
Lagrange multipliers
0 references
parametric nonlinear optimization
0 references
[For the entire collection see Zbl 0721.00029.] This paper considers parametric nonlinear optimization problems of the form \(\min f(x,t)\), \(t\in[0,1]\) or \(t\in\mathbb{R}\), subject to \(h_ i(x,t)=0\), \(i\in I\), \(g_ j(x,t)\leq 0\), \(j\in J\), where \(I\) and \(J\) are integer index sets. Of interest is to follow paths of solutions from \(t=0\) to \(t=1\). The approach is to track solutions to a slight generalization of the Karush-Kuhn-Tucker equations. Various contingencies, related to changing of the active set of constraints and encountering singularities in the reduced Hessian matrix of the Lagrangian, may prevent straightforward tracking from \(t=0\) to \(t=1\). In this paper, all possible such cases are outlined. Numerous illustrative figures appear. It is shown how to pose modified problems in a neighborhood of such a case, so that solutions of the modified problem connect to solutions of the original problem on both sides of the exceptional point. Algorithms appear which incorporate the analysis. A special technique is used in an attempt to jump from one connected component of solutions to another. Six numerical examples, including tables and well-labelled figures, appear. The paper appears to be typewritten, and the figures were produced by hand.
0 references