Membership-set estimation using random scanning and principal componet analysis

A set-theoretic approach to parameter estimation based on the bounded-error concept is an appropriate choice when incomplete knowledge of observation error statistics and unavoidable structural model error invalidate the presuppositions of stochastic methods. Within this class the estimation of non-linear-in-the-parameters models is examined. This situation frequently occurs in modelling natural systems. The output error method proposed is based on overall random scanning with iterative reduction of the size of the scanned region. In order to overcome the problem of computational inefficiency, which is particularly serious when there is interaction between the parameter estimates, two modifications to the basic method are introduced. The first involves the use of principal component transformations to provide a rotated parameter space in the random scanning because large areas of the initial parameter space are thus excluded from further examination. The second improvement involves the standardization of the parameters so as to obtain an initial space with equal size extension in all directions. This proves to largely increase the computational robustness of the method. The modified algorithm is demonstrated by application to a simple three-parameter model of diurnal dissolved oxygen patterns in a lake.

[1]  M. B. Beck,et al.  Uncertainty and arbitrariness in ecosystems modelling: A lake modelling example , 1981 .

[2]  J. P. Norton,et al.  Fast and robust algorithm to compute exact polytope parameter bounds , 1990 .

[3]  M. Milanese,et al.  Estimation theory and uncertainty intervals evaluation in presence of unknown but bounded errors: Linear families of models and estimators , 1982 .

[4]  G. Hornberger,et al.  Modelling algal behaviour in the river thames , 1984 .

[5]  Fred C. Schweppe,et al.  Uncertain dynamic systems , 1973 .

[6]  M. Milanese,et al.  Uncertainty intervals evaluation in presence of unknown but bounded errors: non linear families of models. , 1981 .

[7]  Sylviane Gentil,et al.  Recursive membership estimation for output-error models , 1990 .

[8]  G. Belforte,et al.  An Improved Parameter Identification Algorithm for Signals with Unknown-But-Bounded Errors , 1985 .

[9]  Sylviane Gentil,et al.  Reformulation of parameter identification with unknown-but-bounded errors , 1988 .

[10]  J. P. Norton,et al.  Identification and application of bounded-parameter models , 1985, Autom..

[11]  V. Broman,et al.  A compact algorithm for the intersection and approximation of N -dimensional polytopes , 1990 .

[12]  G. V. Straten.  Analytical methods for parameter-space delimination and application to shallow-lake phytoplankton-dynamics modeling , 1985 .

[13]  Eric Walter,et al.  Estimation of non-uniquely identifiable parameters via exhaustive modeling and membership set theory , 1986 .

[14]  D. Dubois,et al.  Additions of interactive fuzzy numbers , 1981 .

[15]  D. Bertsekas,et al.  Recursive state estimation for a set-membership description of uncertainty , 1971 .

[16]  J. Norton.  Identification of parameter bounds for ARMAX models from records with bounded noise , 1987 .

[17]  J. P. Norton,et al.  Identification and Application of Bounded-Parameter Models , 1985 .

[18]  P. Eykhoff.  System Identification Parameter and State Estimation , 1974 .

[19]  Y. F. Huang,et al.  On the value of information in system identification - Bounded noise case , 1982, Autom..

[20]  Eric Walter,et al.  Exact recursive characterization of feasible parameter sets in the linear case , 1990 .