TRUST REGION METHODS FOR NONLINEAR EQUALITY CONSTRAINED OPTIMIZATION

This paper presents two trust region algorithms for solving general nonlinear equality constrained optimization problems. The different techniques are used in minimizing a quadrdtic approximate model of the original optimization problems in trust region. By the convergence analysis, the global convergences and local superlinear convergences of the algorithms are obtained. Finally,the implementations of the algorithms are given.

Project Supported:

Clc Number:

O221.2