Rational model data smoothers and identification algorithms (Q4364040)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 1086118
Language Label Description Also known as
default for all languages
No label defined
    English
    Rational model data smoothers and identification algorithms
    scientific article; zbMATH DE number 1086118

      Statements

      Rational model data smoothers and identification algorithms (English)
      0 references
      0 references
      0 references
      18 June 1998
      0 references
      structure detection
      0 references
      nonlinear rational model smoothing algorithm
      0 references
      identification
      0 references
      parameter estimation
      0 references
      A new nonlinear rational model smoothing algorithm has been presented. One considers a system with nonlinearity of the input, output, noise and parameters of the rational function models, which make the identification of such a model very complicated and computationally uneconomical. To overcome the hard noise problem in rational model estimation, a two step identification algorithm has been utilized: the proposed new smoother is estimated without any a priori information from the raw data, based on past, present and future sample values about the current time step; then data smoothing is performed and tests are introduced to determine when the noise has been acceptably suppressed. Structure detection and parameter estimation is then applied to the smoothed data to obtain unbiased estimates of the dynamic nonlinear rational model. Thus, the fit to complex nonlinear rational models with excellent approximation properties has been achieved, using polynomial algorithms. Illustrative examples are presented to demonstrate that the proposed approach provides unbiased estimates with the same degree of accuracy, compared with known estimators.
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references
      0 references