Smoothing approximation to the k-th power nonlinear penalty function for constrained optimization problems

摘要

In this paper, a new smoothing approximation to the k-thpower nonlinear penalty function for constrained optimization problems ispresented. We prove that this type of the smoothing penalty functions has goodproperties in solving constrained optimization problems. Furthermore, based onthe smoothed penalty problem, an algorithm is presented to solve theconstrained optimization problems, with its convergence under some conditionsproved. Some numerical examples are given to illustrate the applicability ofthe present smoothing method, which show that the algorithm seems efficient.