Abstract

Presented in this paper is a primal-dual infeasible-interior-point quadratic programming (QP) algorithm and its extension for nonlinear programming that is suited for engineering design and structural optimization, where the number of variables are very large and function evaluations are computationally expensive. The computational experience in solving both test problems and optimal structural design problems using the algorithm demonstrated that the algorithm finds an approximate optimal solution in fewer iterations and function evaluations, the obtained solution usually is an interior feasible solution, and so the resulting method is very efficient and effective.

This content is only available via PDF.
You do not currently have access to this content.