Journal of Jilin University Science Edition

Previous Articles     Next Articles

Combined Homotopy Interior Point Algorithm for a Class ofUnbounded Nonconvex Optimization Problems

CAI Zhidan1, ZHAO Liqin2, SU Menglong3,4   

  1. 1. School of Science, Changchun University of Science and Technology, Changchun 130022, China;2. Editorial Department of Journal of Jilin University, Changchun 130012, China; 3. College of Mathematics, Luoyang Normal University, Luoyang 471022, Henan Province, China; 4. Key Laboratory of Symbol Computation and Knowledge Engineer of Ministry of Education, Jilin University, Changchun 130012, China
  • Received:2013-03-26 Online:2013-11-26 Published:2013-11-21
  • Contact: SU Menglong E-mail:mlsulynu@163.com

Abstract:

The authors proposed a combined homotopy interior point algorithm to solve a class of unbounded nonconvex optimization problems. Under suitable conditions, we obtained the existence of the homotopy path. By tracking this homotopy path, we can get a K-K-T point of the nonconvex optimization problem. The results provide an efficient globally convergent algorithm for the nonconvex unbounded optimization problems.

Key words: non-convex unbounded optimization problem, homotopy path, globally convergent algorithm

CLC Number: 

  • O221.2