Projected affine-scaling interior-point Newton’s method with line search filter for box constrained optimization

摘要

We present a new projected affine-scaling interior-point Newton’s method with a line search filter algorithm for nonlinear optimization problems with box constraints. The global convergence of the proposed algorithm is established under some reasonable conditions. Furthermore, the method is shown to be locally quadratically convergent under the strong second order sufficiency condition without assuming strict complementarity of the solution. Numerical experiments are presented that confirm the robustness and efficiency of the approach.