A modified penalty term for the sequential unconstrained minimization technique for convex programming problems
Leahy, Vincent J.
MetadataShow full item record
The Sequential Unconstrained Minimization Technique (SUMT) for Convex Programming Problems is modified by the introduction of an exponent in the penalty term. The exponent is introduced to increase the rate of convergence of the method for nonlinear problems with solutions on the boundary of one or more constraints. Convergence to the solution of the constrained problem is proved, and it is shown that SUMT is a special case of the general unconstrained function with the exponent equal to one. Results of a sample problem indicate that the rate of convergence is improved and that the computational time for solution is decreased for an exponent less than one.
Showing items related by title, author, creator and subject.
Connections between the covector mapping theorem and convergence of pseudospectral methods for optimal control Gong, Qi; Ross, I. Michael; Kang, Wei; Fahroo, Fariba (Springer Science + Business Media, LLC, 2008);In recent years, many practical nonlinear optimal control problems have been solved by pseudospectral (PS) methods. In particular, the Legendre PS method offers a Covector Mapping Theorem that blurs the distinction between ...
Luscombe, James H.; Luban, Marshall; Reynolds, Joseph P. (American Physical Society, 1996-06);We obtain the exact critical relaxation time TL(E), where E is the bulk correlation length, for the Glauber kinetic Ising model of spins on a one-dimensional lattice of finite length L for both periodic and free boundary ...
Royset, Johannes O.; Szechtman, Roberto (2012-12);The sample average approximation approach to solving stochastic programs induces a sampling error, caused by replacing an expectation by a sample average, as well as an optimization error due to approximating the solution ...