×

A semifeasible trust-region model algorithm for minimization with inequality constraint. (English) Zbl 1006.90075

In [J. M. Martinez and L. T. Santos, J. Optim. Theory Appl. 97, 435-454 (1998; Zbl 0907.90220)] the author introduced a new algorithm of inexact-restoration type for solving minimization problems with equality constraints and bounds on the variables. The procedure to force global convergence of this algorithm makes use of an augmented Lagrangian merit function. In this paper the same ideas are applied to the problems where inequality constraints are explicitly given. The global convergence of the new method is proved, but the merit function does not allow to use arbitrary estimates of Lagrange multipliers.

MSC:

90C30 Nonlinear programming
65K05 Numerical mathematical programming methods

Citations:

Zbl 0907.90220
PDFBibTeX XMLCite