• DocumentCode
    1256813
  • Title

    Decoupled Stein iterations to the discrete-time generalized Riccati equations

  • Author

    Ivanov, Ivan ; Dragan, Vasile

  • Author_Institution
    Fac. of Econ. & Bus. Adm., Sofia Univ. `St. Kl.Ohridski`, Sofia, Bulgaria
  • Volume
    6
  • Issue
    10
  • fYear
    2012
  • Firstpage
    1400
  • Lastpage
    1409
  • Abstract
    The authors investigate the numerical solution of a set of discrete-time generalised Riccati equations. The class of discrete-time non-linear equations involves in various control problems for discrete-time stochastic systems and it can be considered as an important tool for solving optimisation control for such type systems. A new procedure for computing the maximal solution and the stabilising solution is proposed by Dragan et al. (`Iterative algorithm to compute the maximal and stabilising solutions of a general class of discrete-time Riccati-type equations`, Int. J. Control, 2010, 83, (4), pp. 837-847). In this study, the authors introduce a new iterative procedure based on the solution of a Stein matrix equation for computing the maximal and the stabilising solution. The convergence properties of the new iteration are proved. Sufficient conditions for computing the maximal solution and the stabilising solution are derived. Finally, some numerical examples are presented to illustrate the feasibility of the proposed algorithm.
  • Keywords
    Riccati equations; convergence of numerical methods; discrete time systems; iterative methods; nonlinear equations; numerical stability; optimisation; stochastic systems; Stein matrix equation; decoupled Stein iterations; discrete-time generalized Riccati equations; discrete-time nonlinear equations; discrete-time stochastic systems; iterative procedure; maximal solution; optimisation control; stabilising solution;
  • fLanguage
    English
  • Journal_Title
    Control Theory & Applications, IET
  • Publisher
    iet
  • ISSN
    1751-8644
  • Type

    jour

  • DOI
    10.1049/iet-cta.2011.0463
  • Filename
    6257076