Acta Mathematicae Applicatae Sinica

, Volume 12, Issue 1, pp 40–49 | Cite as

A generalized gradient projection method for optimization problems with equality and inequality constraints about arbitrary initial point

  • Gao Ziyou 
  • Lai Yanlian 
  • Hu Zongying 


In this paper, we give a new generalized gradient projection algorithm for nonlinear optimization problems with arbitrary initial point. This new algorithm has some important advantages as follows: (1) The algorithm does not require initial feasible point; (2) It can deal with nonlinear equality and inequality constraints problems; (3) The structure of our algorithm is very simple; (4) Under some mild assumptions, it has global convergence.

Key words

Generalized projection matrix arbitrary initial point directional derivative global convergence 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    Hesrkovits, J. A Two-stage Feasible Directions Algorithm for Nonlinear Constrained Optimization.Mathematical Programming, 1986, 36: 19–38.MathSciNetGoogle Scholar
  2. [2]
    Mayne, D.Q. and Polak, E. Feasible Directions Algorithm for Optimization Problems with Equality and Inequality Constraints.Mathematical Programming, 1976, 11: 67–80.CrossRefMATHMathSciNetGoogle Scholar
  3. [3]
    Bazaraa, M.S. and Shetty, C.M. Nonlinear Programming: Theory and Algorithm. John Wiley & Sons, 1979.Google Scholar
  4. [4]
    Polak, E., Trahan, R. and Mayne, D.Q. Combined Phase I-Phase-II Methods of Feasible Directions.Mathematical Programming, 1979, 17: 32–61.CrossRefMathSciNetGoogle Scholar
  5. [5]
    Goldfarb D. Extension of Davidon's Variable Metric Method to Maximization under Linear Inequality and Equality Constraints.SIAM J. Appl. Math., 1969, 17: 739–764.CrossRefMATHMathSciNetGoogle Scholar
  6. [6]
    Wu F. A Superlinear Convergent Gradient Projection Type Algorithm for Linearly Constrained Problems.European J. of OR, 1981, 16: 334–344.CrossRefGoogle Scholar
  7. [7]
    Du, D.Z. and Sun, J. A New Gradient Projection Method.Mathematica Numerica Sinica, 1983, 5(4)4, 378–386 (in Chinese).MathSciNetGoogle Scholar
  8. [8]
    Du, D.Z. A Gradient Projection Algorithm for Nonlinear Constraints.Acta Mathematicae Applicatae Sinica, 1985, 8(1): 7–16 (in Chinese).MATHMathSciNetGoogle Scholar
  9. [9]
    Gao, Z.Y. and He G.P. A Generalized Gradient Projection Algorithm for Constrained Optimization Problem.Chinese Science Bulletin, 1991, 36(19): 1444–1447 (in Chinese).Google Scholar
  10. [10]
    Lai Y.L., Gao, Z.Y. and He, G.P. A Generalized Gradient Projection Algorithm for Nonlinearly Optimization Problem.Scientia Sinica, 1993, 36(2): 170–180.MATHMathSciNetGoogle Scholar
  11. [11]
    Gao Z.Y., and Yu W. A Generalized Gradient Projection Algorithm with Arbitrary Initial Point for Constrained Optimization Problem.Chinese Science Bulletin, 1992, 37(20): 1832–1836 (in Chinese).Google Scholar
  12. [12]
    Lai Y.L. and Wei Z.X. A Globally Convergent Gradient Projection Algorithm with Arbitrary Initial Point.Chinese Science Bulletin, 1990, 35(20): 1536–1539 (in Chinese).Google Scholar
  13. [13]
    Lai Y.L. and Wei Z.X., A Unified Approach to the Method of Gradient Projection with Arbitrary Initial Point.Systems Science and Mathematics Science, 1991, 3: 215–224.MathSciNetGoogle Scholar
  14. [14]
    Gao Z.Y. Robust and Efficient Feasible Direction Algorithm for Constrained Optimization Problem. Proceedings of the Symposium on Applied Mathematics, the Chinese Academy of Sciences, Beijing, 1992, 441–449.Google Scholar
  15. [15]
    S.P. Han. A Globally Convergent Method for Nonlinear Programming.JOTA, 1977, 22(3): 297–309.CrossRefMATHGoogle Scholar

Copyright information

© Science Press, Beijing, China and Allerton Press, Inc. New York, U.S.A. 1996

Authors and Affiliations

  • Gao Ziyou 
    • 1
  • Lai Yanlian 
    • 2
  • Hu Zongying 
    • 3
  1. 1.Northern Jiaotong UniversityBeijingChina
  2. 2.Institute of Applied Mathematicsthe Chinese Academy of SciencesBeijingChina
  3. 3.Gezhouba Institute of Hydro Electric EngineeringYichangChina

Personalised recommendations