. An off-line algorithm for semi-empirical modeling of nonlinear dynamic systems is presented. The model representation is based on the interpolation of a number of simple local models, where the validity of each local model is restricted to an operating regime, but where the local models yield a complete global model when interpolated. The input to the algorithm is a sequence of empirical data and a set of candidate local model structures. The algorithm searches for an optimal decomposition into operating regimes, and local model structures. The method is illustrated using simulated and real data. The transparency of the resulting model and the flexibility with respect to incorporation of prior knowledge is discussed. 1 Introduction The problem of identifying a mathematical model of an unknown system from a sequence of empirical data is a fundamental one which arises in many branches of science and engineering. The complexity of solving such a problem depends on many factors, such as...
|
804
|
System Identification, Theory for the user
– Ljung
- 1987
|
|
578
|
Adaptive mixtures of local experts
– Jacobs, Jordan, et al.
- 1991
|
|
576
|
Hierarchical mixtures of experts and the EM algorithm
– Jordan, Jacobs
- 1994
|
|
487
|
Fuzzy identification of systems and its application to modelling and control
– Takagi, Sugeno
- 1985
|
|
458
|
Cross-validatory choice and assessment of statistical predictions
– Stone
- 1974
|
|
231
|
Smoothing noisy data with spline functions
– Craven, Wahba
- 1979
|
|
170
|
Multiple adaptive regression splines (with discussion
– Friedman
- 1991
|
|
81
|
Fitting autoregressive models for prediction
– Akaike
- 1969
|
|
78
|
Structure identification of fuzzy model
– Sugeno, Kang
- 1988
|
|
54
|
Constructing NARMAX models using ARMAX models
– JOHANSEN, FOSS
- 1992
|
|
36
|
Selection of the order of an autoregressive model by Akaike’s information criterion
– Shibata
- 1976
|
|
32
|
Exploiting Neurons with Localized Receptive Fields to Learn Chaos
– STOKBRO, UMBERGER, et al.
- 1990
|
|
29
|
Convergence analysis of parametric identification methods
– Ljung
- 1978
|
|
25
|
Construction of fuzzy models through clustering techniques
– Yoshinari, Pedrycz, et al.
- 1993
|
|
24
|
Detection and characterization of cluster substructure, I. linear structure: fuzzy c-lines
– Bezdek, Coray, et al.
- 1981
|
|
22
|
ASMOD: An algorithm for adaptive spline modeling of observation data
– Kavli
- 1993
|
|
14
|
A Comparison of Two Nonparametric Estimation Schemes
– Veaux, Psichogios, et al.
- 1993
|
|
12
|
Model-structure selection by cross-validation
– Stoica, Eykhoff, et al.
- 1986
|
|
9
|
State-space Modelling using operating Regime Decomposition
– JOHANSEN, FOSS
- 1993
|
|
8
|
Nonlinear adaptive networks: A little theory, a few applications
– Jones, Barnes, et al.
- 1990
|
|
8
|
A Generalization Error Estimate for Nonlinear Systems
– Larsen
- 1992
|
|
8
|
A constructive learning algorithm for local model networks
– Murray-Smith, Gollee
- 1994
|
|
8
|
Construction of composite models from observed data, Int
– SKEPPSTEDT, LJUNG, et al.
- 1992
|
|
7
|
A comparison of four methods for non-linear data modelling
– Carlin, Kavli, et al.
- 1994
|
|
6
|
Local Model Networks and Local Learning
– Murray-Smith
- 1994
|
|
6
|
Unified structure and parameter identification of fuzzy models
– Yager, Filev
- 1993
|
|
4
|
Kinetic analysis of gluconic acid production by pseudomonas ovalis
– Ghose, Ghosh
- 1976
|
|
3
|
Model structure identification using separate validation data - asymptotic properties. Submitted to the European Control Conference
– Johansen, Weyer
- 1995
|
|
2
|
Nonuniformly partitioned piecewise linear representation of continuous learned mappings
– Kavli
- 1990
|
|
1
|
A dynamic modeling framework based on local models and interpolation -- combining empirical and mechanistic knowledge and data. Submitted to Computers and Chemical Engineering
– Johansen, Foss
- 1994
|
|
1
|
Application of a general multimodel approach for identification of highly non-linear processes -- A case study
– Pottmann, Unbehauen, et al.
- 1993
|
|
1
|
A combined network architecture using ART2 and back propagation for adaptive estimation of dynamical processes
– S��rheim
- 1990
|
