An Effective Line Search for the Subgradient Method

  • C. Beltran
  • F. J. Heredia

Abstract

One of the main drawbacks of the subgradient method is the tuning process to determine the sequence of steplengths. In this paper, the radar subgradient method, a heuristic method designed to compute a tuning-free subgradient steplength, is geometrically motivated and algebraically deduced. The unit commitment problem, which arises in the electrical engineering field, is used to compare the performance of the subgradient method with the new radar subgradient method.

Keywords

Lagrangian relaxation multiplier updating subgradient method radar unit commitment 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Bertsekas, D.P. 1999Nonlinear Programming2Athena ScientificBelmont, MassachusettsGoogle Scholar
  2. 2.
    Luenberger, D. G. 1984Linear and Nonlinear Programming2Addison-Wesley, Reading Massachusetts Google Scholar
  3. 3.
    Hiriart-Urruty, J.B., Lemaréchal, C. 1996Convex Analysis and Minimization Algorithms.Springer VerlagBerlin, Germany Vols. I and II. Google Scholar
  4. 4.
    Shor, N.Z. 1998Nondifferentiable Optimization and Polynomial Problems, Nonconvex Optimization and Its Applications.Kluwer Academic PublishersDordrecht Holland Google Scholar
  5. 5.
    Feltenmark, S. 1997On Optimization of Power ProductionDepartment of Mathematics, Optimization, and Systems Theory. Royal Institute of TechnologyStockholm, SwedenPhD ThesisGoogle Scholar
  6. 6.
    Kelley, J. E. 1960The Cutting-Plane Method for Solving Convex ProgramsSIAM Journal on Applied Mathematics8703712Google Scholar
  7. 7.
    Feltenmark, S., Kiwiel, K., Dual, C. 2000Applications of Proximal Bundle Methods,Including Lagrangian Relaxation of Nonconvex ProblemsSIAM Journal on Optimization10697721Google Scholar
  8. 8.
    Kiwiel, K. C., A Survey of Bundle Methods for Nondifferentiable Optimization, Proceedings of the 13th International Symposium on Mathematical Programming, Tokyo, Japan, 1988.Google Scholar
  9. 9.
    Schramm, H., Zowe, J.,  et al. 1988

    A Combination of the Bundle Approach and the Trust-Region Concept

    Guddat, J.Bank, B. eds. Advances in Mathematical OptimizationAkademie VerlagBerlin, Germany
    Google Scholar
  10. 10.
    Schramm, H., Zowe, J. 1992A Version of the Bundle Idea for Minimizing a Nonsmooth Function: Conceptual Idea, Convergence Analysis, Numerical ResultsSIAM Journal on Optimization2121152Google Scholar
  11. 11.
    Du Merle, O., Goffin, J.L.,  et al. 1998On Improvements to the Analytic Center Cutting Plane MethodComputational Optimization and Applications113752Google Scholar
  12. 12.
    Goffin, J.L., Vial, J.P. 1999Convex Nondifferentiable Optimization:A Survey Focused on the Analytic Center Cutting Plane MethodGeneva UniversityHEC, LogilabTechnical Report 99.02Google Scholar
  13. 13.
    Gondzio, J., Sarkissian, R.,.,  et al. 2001Parallel Implementation of a Central Decomposition Method for Solving Large-Scale Planning ProblemsComputational Optimization and Applications19529Google Scholar
  14. 14.
    Nemirovsky, A.S., Yudin, D. B. 1983Problem Complexity and Method Efficiency in Optimization.John Wiley and Sons New York, NY Google Scholar
  15. 15.
    Baker, B. M., Sheasby, J. 1999Accelerating the Convergence of Subgradient OptimizationEuropean Journal of Operational Research117136144Google Scholar
  16. 16.
    Barahona, F., Anbil, R. 2000The Volume Algorithm: Producing Primal Solutions with a Subgradient MethodMathematical Programming87385399Google Scholar
  17. 17.
    Nedic, A., Bertsekas, D. P. 2001Incremental Subgradient Methods for Nondifferentiable OprimizationSIAM Journal on Optimization12109138Google Scholar
  18. 18.
    Kiwiel, K., Larsson, C.T.,  et al. 1999The Efficiency of Ballstep Subgradient Level Methods for Convex OptimizationMathematics of Operations Research24237254Google Scholar
  19. 19.
    Beltran, C., Heredia, F. J. 1999Short-Term Hydrothermal Coordination by Augmented Lagrangian Relaxation: A New Multiplier UpdatingInvestigación Operativa86376Google Scholar
  20. 20.
    Lai, S., Baldick, R. 1999Unit Commitment with Ramp MultipliersIEEE Transactions on Power Systems145864Google Scholar
  21. 21.
    Wang, S.J., Shahidehpour, S. M.,  et al. 1995Short-Term Generation Scheduling with Transmission and Environmental Constraints Using an Augmented Lagrangian RelaxationIEEE Transactions on Power Systems1012941301Google Scholar
  22. 22.
    Pellegrino, F., Renaud, A.,  et al. 1996Bundle and Augmented Lagrangian Methods for Short-Term Unit Commitment12th Power System Computation Conference2730739Google Scholar
  23. 23.
    Beltran, C., Heredia, F. J. 2002Unit Commitment by Augmented Lagrangian Relaxation: Testing Two Decomposition ApproachesJournal of Optimization Theory and Applications112295314Google Scholar
  24. 24.
    Beltran, C., Generalized Unit Commitment by the Radar Multiplier Method, PhD Thesis, Statistics and Operations Research, Polytechnic University of Catalonia, Barcelona, Catalonia, Spain, 2001; see http://www.tdcat.cesca.es/TDCat-0720101-132715/, 2001.Google Scholar

Copyright information

© Springer Science+Business Media, Inc. 2005

Authors and Affiliations

  • C. Beltran
    • 1
  • F. J. Heredia
    • 2
  1. 1.Researcher, Logilab, HECUniversity of GenevaGenevaSwitzerland
  2. 2.Professor, Department of Statistics and Operations ResearchPolytechnical University of CataloniaBarcelonaSpain

Personalised recommendations