AN ACTIVE-SET INTERIOR-POINT TRUST-REGION ALGORITHM

In this work, an active-set method is used to transform a general nonlinear programming problem with bounds on the variables to an equality constrained optimization problem with bound on the variables. By using a Coleman-Li strategy the iterates which are generated by the proposed algorithm are strictly feasible. An interior-point Newton method is used and a trust-region globalization strategy is added to the algorithm to insure global convergence. A reduced Hessian technique is used to overcome the difficulty of having an infeasible trust-region subproblem. A global convergence analysis for this algorithm is presented under credible assumptions. Preliminary numerical results are reported.