Numerical experience with a reduced Hessian method for large scale constrained optimization

The reduced Hessian SQP algorithm presented in Biegler et al. [SIAM J. Opti mization, Vol. 5, no. 2, pp. 314-347, 1995.] is developed in this paper int o a practical method for large-scale optimization. The novelty of the algor ithm lies in the incorporation of a correction vector that approximates the cross term Z(T)WYp(Y). This improves the stability and robustness of the a lgorithm without increasing its computational cost. The paper studies how t o implement the algorithm efficiently, and presents a set of tests illustra ting its numerical performance. An analytic example, showing the benefits o f the correction term, is also presented.