Search

This paper describes the identification of a temperature dependent FitzHugh-Nagumo model directly from experimental observations with controlled inputs. By studying the steady states and the trajectory of the phase of the variables, the stability of the model is analyzed and a rule to generate oscillation waves is proposed. The dependence of the oscillation frequency and propagation speed on the model parameters is then investigated to seek the appropriate control variables, which then become functions of temperature in the identified model. The results show that the proposed approach can provide a good representation of the dynamics of the oscillatory behavior of a Belousov-Zhabotinskii reaction. [ABSTRACT FROM AUTHOR]

A simple scalar coupled map lattice (sCML) model for excitable media is derived in this paper. The new model, which has a simple structure, is shown to be closely related to the observed phenomena in excitable media. Properties of the sCML model are also investigated. Illustrative examples show that this kind of model is capable of reproducing the behavior of excitable media and of generating complex spatiotemporal patterns. [ABSTRACT FROM AUTHOR]

The wavefront profile and the propagation velocity of waves in an experimentally observed Belousov-Zhabotinskii reaction are analyzed and a revised FitzHumgh-Nagumo (FHN) model of these systems is identified. The ratio between the excitation period and the recovery period, for a solitary wave is studied, and included within the model. Averaged traveling velocities at different spatial positions are shown to be consistent under the same experimental conditions. The relationship between the propagation velocity and the curvature of the wavefront are also studied to deduce the diffusion coefficient in the model, which is a function of the curvature of the wavefront and not a constant. The application of the identified model is demonstrated on real experimental data and validated using multistep ahead predictions. [ABSTRACT FROM AUTHOR]

The identification problem for excitable media is investigated in this paper. A new scalar coupled map lattice (SCML) model is introduced and the orthogonal least squares algorithm is employed to determinate the structure of the SCML model and to estimate the associated parameters. A simulated pattern and a pattern observed directly from a real Belousov–Zhabotinsky reaction are identified. The identified SCML models are shown to possess almost the same local dynamics as the original systems and are able to provide good long term predictions. [ABSTRACT FROM AUTHOR]

A method of identifying the transition rule, encapsulated in a modified cellular automata (CA) model, is demonstrated using experimentally observed data of the evolution of dendritic crystal growth patterns in NH4Br crystals. The influence of the factors, such as experimental set-up and image preprocessing, color and size calibrations, on the method of identification are discussed in detail. A noise reduction parameter and the diffusion velocity of the crystal boundary are also considered. The results show that the proposed method can in principle provide a good representation of the dendritic growth anisotropy. [ABSTRACT FROM AUTHOR]

Particle swarm optimization (PSO) is introduced to implement a new constructive learning algorithm for training generalized cellular neural networks (GCNNs) for the identification of spatio-temporal evolutionary (STE) systems. The basic idea of the new PSO-based learning algorithm is to successively approximate the desired signal by progressively pursuing relevant orthogonal projections. This new algorithm will thus be referred to as the orthogonal projection pursuit (OPP) algorithm, which is in mechanism similar to the conventional projection pursuit approach. A novel two-stage hybrid training scheme is proposed for constructing a parsimonious GCNN model. In the first stage, the orthogonal projection pursuit algorithm is applied to adaptively and successively augment the network, where adjustable parameters of the associated units are optimized using a particle swarm optimizer. The resultant network model produced at the first stage may be redundant. In the second stage, a forward orthogonal regression (FOR) algorithm, aided by mutual information estimation, is applied to refine and improve the initially trained network. The effectiveness and performance of the proposed method is validated by applying the new modeling framework to a spatio-temporal evolutionary system identification problem. [ABSTRACT FROM AUTHOR]

Many complex and interesting spatiotemporal patterns have been observed in a wide range of scientific areas. In this paper, two kinds of spatiotemporal patterns including spot replication and Turing systems are investigated and new identification methods are proposed to obtain Coupled Map Lattice (CML) models for this class of systems. Initially, a new correlation analysis method is introduced to determine an appropriate temporal and spatial data sampling procedure for the identification of spatiotemporal systems. A new combined Orthogonal Forward Regression and Bayesian Learning algorithm with Laplace priors is introduced to identify sparse and robust CML models for complex spatiotemporal patterns. The final identified CML models are validated using correlation-based model validation tests for spatiotemporal systems. Numerical results illustrate the identification procedure and demonstrate the validity of the identified models. [ABSTRACT FROM AUTHOR]

The identification problem for spatiotemporal patterns which are generated by autonomous Cellular Neural Networks (CNN) is investigated in this paper. The application of traditional identification algorithms to these special spatiotemporal systems can produce poor models due to the inherent piecewise nonlinear structure of CNN. To solve this problem, a new type of Coupled Map Lattice model with output constraints and corresponding identification algorithms are proposed in the present study. Numerical examples show that the identified CML models have good prediction capabilities even over the long term, and the main dynamics of the original patterns appears to be well represented. [ABSTRACT FROM AUTHOR]

The purpose of model validation is to test the goodness of fit of the identified model and to check whether the model is an appropriate representation of the underlying system. Model validation is a final and essential stage in most system identification procedures. Although model validation for nonlinear temporal systems has been extensively studied, model validation for spatiotemporal systems is still an open question. In this paper, correlation based methods, which have been successfully applied in nonlinear temporal systems are extended and enhanced to validate models of spatiotemporal systems. New nonlinear correlation functions are constructed based on the inputs, one-step-ahead predicted outputs and the residuals to test the dependency between the residuals and inputs/outputs. Examples are included to demonstrate the application of the tests. [ABSTRACT FROM AUTHOR]

The identification of binary cellular automata from spatio-temporal binary patterns is investigated in this paper. Instead of using the usual Boolean or multilinear polynomial representation, the Fourier transform representation of Boolean functions is employed in terms of a Fourier basis. In this way, the orthogonal forward regression least-squares algorithm can be applied directly to detect the significant terms and to estimate the associated parameters. Compared with conventional methods, the new approach is much more robust to noise. Examples are provided to illustrate the effectiveness of the proposed approach. [ABSTRACT FROM AUTHOR]

New methods of identifying the transition rule of a Belousov–Zhabotinskii (BZ) reaction directly from experimental data using cellular automata (CA) models are investigated. The experimental set-up and new techniques for image pre-processing to ensure the identification of representative models are discussed including noise reduction, pixel and color calibration. Two kinds of models, the Greenberg–Hasting model (GHM) and the polynomial CA model are studied in detail. It is shown that the results of identifying a real BZ reacting system are very encouraging and the predicted patterns compare well with the imaged patterns both visually and quantitatively. [ABSTRACT FROM AUTHOR]

Excitable media represent an important class of spatio-temporal systems which can be studied using cellular automata models. In the present study examples of excitable media behavior generated using simple cellular automata models are introduced. A mutual information algorithm is then derived to determine the neighborhood, the excitation threshold, and the number of excitation states. Based on this information two methods of identifying the rule which describes the excitable media pattern using a multimodel and a polynomial model are introduced. The results are illustrated using simulated examples and real data from a Belousov–Zhabotinskii experiment. [ABSTRACT FROM AUTHOR]

A comparison between polynomial and wavelet expansions for the identification of coupled map lattice (CML) models for deterministic spatiotemporal dynamical systems is presented in this paper. The pattern dynamics generated by smooth and nonsmooth nonlinear maps in a well-known two-dimensional CML structure are analyzed. By using an orthogonal feedforward regression algorithm (OFR), polynomial and wavelet models are identified for the CML's in chaotic regimes. The quantitative dynamical invariants such as the largest Lyapunov exponents and correlation dimensions are estimated and used to evaluate the performance of the identified models. [ABSTRACT FROM AUTHOR]

A new neighborhood selection method is presented for both deterministic and probabilistic cellular automata models. The detection criteria are built explicitly on the corresponding contribution which is made to the value of each updated cell from each detected cell in the evolution. Theoretical analysis and numerical simulations demonstrate the effectiveness of this new method. [ABSTRACT FROM AUTHOR]

A new NARMA based smoothing algorithm is introduced for chaotic and nonchaotic time series. The new algorithm employs a cross-validation method to determine the smoother structure, requires very little user interaction, and can be combined with wavelet thresholding to further enhance the noise reduction. Numerical examples are included to illustrate the application of the new algorithm. [ABSTRACT FROM AUTHOR]

The Response Spectrum Map (RSM) is introduced as a frequency domain equivalent to the Bifurcation Diagram. The RSM is a map of the energy distribution of a system in the frequency domain, where subharmonics, superharmonics and chaos generation can be revealed. The RSM is used in this paper to qualitatively analyze and detect various dynamical states exhibited by a nonlinear system. [ABSTRACT FROM AUTHOR]

Nonlinear system identification and analysis methods are employed to study the low-frequency oscillations present in time-series data obtained from reflectance imagery of microvasculature. Using the method of surrogate data testing the analysis reveals the deterministic nature of these oscillations believed by many to be chaotic. Further investigations by means of nonlinear system identification techniques indicate however that the underlying dynamics can described by a periodically driven nonlinear dynamical model exhibiting quasiperiodic behavior. [ABSTRACT FROM AUTHOR]

This paper develops an original approach for identifying models of chaotic systems directly from noise-corrupted data. The nonlinear functional describing the process is constructed using a new multiresolution model structure implemented with B-spline wavelet and scaling functions. Following an iterative strategy, a sequence of model sets of increasing complexity are postulated and tested until a suitable model is found. An orthogonal-forward-regression routine coupled with model validity tests is used to select parsimonious wavelet models and to measure the quality of the fit. The effectiveness of the identification procedure is demonstrated using both simulated and experimental data. It is shown that the proposed method can produce accurate models which exhibit qualitatively the same dynamical behavior as the observed system and are characterized by dynamical invariants which are very close to those of the original system. [ABSTRACT FROM AUTHOR]

In this paper a new concept, spatio-temporal generalized frequency response functions (STGFRF), is introduced for the first time to characterize nonlinear spatio-temporal dynamical systems in the frequency domain. A probing method is developed to calculate the STGFRFs recursively for both continuous and discrete spatio-temporal systems. The algorithm is computationally compact and exposes the explicit relationship between the continuous and discrete models and the elements of the generalized frequency response functions. [ABSTRACT FROM AUTHOR]