Path-Following Algorithms for Posynomial Geometric Programming Problems

Geometric programming (GP) is a class of optimization problems with a nonlinear objective function subject to a set of nonlinear constraints. The problems considered in this paper are posynomial GP problems, whose solutions can be obtained by solving a dual linear programing problem. The primary pur...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	Journal of the Chinese Institute of Industrial Engineers Ročník 17; číslo 2; s. 223 - 232
Hlavní autor:	Yang, Hsu-Hao
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	Taylor & Francis Group 01.03.2000
Témata:	geometric programming problem Interior point methods path-following algorithm path-following演算法內點法幾何規劃問題
ISSN:	1017-0669, 2151-7606
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	Geometric programming (GP) is a class of optimization problems with a nonlinear objective function subject to a set of nonlinear constraints. The problems considered in this paper are posynomial GP problems, whose solutions can be obtained by solving a dual linear programing problem. The primary purpose of this paper is to investigate the quality of solutions found by, the application of a path-following algorithm, which is one of interior point methods. Two majorr approches to solve posynomial GP problems are investigated. One is dual GP as Generalized Lmear Programming (GLP) and the other is the traditional dual GP. The test results suggest that dual GP as GLP approach is advantageous over the posynomial one in the event that high a~curacy of the solution is required or critically important.
ISSN:	1017-0669 2151-7606
DOI:	10.1080/10170669.2000.10433300