decrease of the penalty parameter in differentiable penalty function methods减少的惩罚参数可微的罚函数方法.pdfVIP

Theoretical Economics Letters, 2011, 1, 8-14 doi:10.4236/tel.2011.11003 Published Online May 2011 (http://www.SciRP.org/journal/tel) Decrease of the Penalty Parameter in Differentiable Penalty Function Methods 1 2 Roohollah Aliakbari Shandiz , Emran Tohidi 1 Faculty of Mathematical Sciences, Sharif University of Technology, Tehran, Iran 2Department of Applied Mathematics, Ferdowsi University of Mashhad, Mashhad, Iran E-mail : aliakbari_r@mehr.sharif.ir, emran.tohidi@stu-mail.um.ac.ir Received March 30, 2011; revised April 28, 2011; accepted May 4, 2011 Abstract We propose a simple modification to the differentiable penalty methods for solving nonlinear programming problems. This modification decreases the penalty parameter and the ill-conditioning of the penalty method and leads to a faster convergence to the optimal solution. We extend the modification to the augmented La- grangian method and report some numerical results on several nonlinear programming test problems, show- ing the effectiveness of the proposed approach. Keywords: Nonlinear Programming, Penalty Method, Penalty Parameter, Differentiable Penalty Methods 1. Introduction Let  0 , a common penalty function for (NLP) is: H 1 H  x = f x P x , Solving nonlinear programming (NLP) problems via a   1       penalty method was first introduced by Courant [1] in m  2