Yuzhu Guo

An iterative orthogonal forward regression algorithm

A novel iterative learning algorithm is proposed to improve the classic Orthogonal Forward Regression (OFR) algorithm in an attempt to produce an optimal solution under a purely OFR framework without using any other auxiliary algorithms. The new algorithm searches for the optimal solution on a global solution space while maintaining the advantage of simplicity and computational efficiency. Both a theoretical analysis and simulations demonstrate the validity of the new algorithm.

IDENTIFICATION OF EXCITABLE MEDIA USING A SCALAR COUPLED MAP LATTICE MODEL

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.

Ultra-Orthogonal Forward Regression Algorithms for the Identification of Non-Linear Dynamic Systems

A new ultra-least squares (ULS) criterion is introduced for system identification. Unlike the standard least squares criterion which is based on the Euclidean norm of the residuals, the new ULS criterion is derived from the Sobolev space norm. The new criterion measures not only the discrepancy between the observed signals and the model prediction but also the discrepancy between the associated weak derivatives of the observed and the model signals. The new ULS criterion possesses a clear physical interpretation and is easy to implement. Based on this, a new Ultra-Orthogonal Forward Regression (UOFR) algorithm is introduced for nonlinear system identification, which includes converting a least squares regression problem into the associated ultra-least squares problem and solving the ultra-least squares problem using the orthogonal forward regression method. Numerical simulations show that the new UOFR algorithm can significantly improve the performance of the classic OFR algorithm.

Identification of nonlinear systems with non-persistent excitation using an iterative forward orthogonal least squares regression algorithm

A new iterative orthogonal least squares forward regression (iOFR) algorithm is proposed to identify nonlinear systems which may not be persistently excited. By slightly revising the classic forward orthogonal regression (OFR) algorithm, the new iterative algorithm provides search solutions on a global solution space. Examples show that the new iterative algorithm is computationally efficient and capable of producing a good model even when the input is not completely persistently excited.

Time-Varying System Identification Using an Ultra-Orthogonal Forward Regression and Multiwavelet Basis Functions With Applications to EEG

A new parametric approach is proposed for nonlinear and nonstationary system identification based on a time-varying nonlinear autoregressive with exogenous input (TV-NARX) model. The TV coefficients of the TV-NARX model are expanded using multiwavelet basis functions, and the model is thus transformed into a time-invariant regression problem. An ultra-orthogonal forward regression (UOFR) algorithm aided by mutual information (MI) is designed to identify a parsimonious model structure and estimate the associated model parameters. The UOFR-MI algorithm, which uses not only the observed data themselves but also weak derivatives of the signals, is more powerful in model structure detection. The proposed approach combining the advantages of both the basis function expansion method and the UOFR-MI algorithm is proved to be capable of tracking the change of TV parameters effectively in both numerical simulations and the real EEG data.

Identification of continuous-time models for nonlinear dynamic systems from discrete data

ABSTRACT A new iOFR-MF (iterative orthogonal forward regression--modulating function) algorithm is proposed to identify continuous-time models from noisy data by combining the MF method and the iOFR algorithm. In the new method, a set of candidate terms, which describe different dynamic relationships among the system states or between the input and output, are first constructed. These terms are then modulated using the MF method to generate the data matrix. The iOFR algorithm is next applied to build the relationships between these modulated terms, which include detecting the model structure and estimating the associated parameters. The relationships between the original variables are finally recovered from the model of the modulated terms. Both nonlinear state-space models and a class of higher order nonlinear input–output models are considered. The new direct method is compared with the traditional finite difference method and results show that the new method performs much better than the finite difference method. The new method works well even when the measurements are severely corrupted by noise. The selection of appropriate MFs is also discussed.

A Volterra series representation for a class of nonlinear infinite dimensional systems with periodic boundary conditions

Abstract This paper proves the existence of a Volterra series representation for the mild solutions of a class of nonlinear infinite dimensional systems. More specifically, given the evolutionary system/operator { U ( t , s ) : 0 ≤ s ≤ t ∞ } associated with a semilinear evolution equation ∂ u / ∂ t = ∂ 2 u / ∂ x 2 + f ( u ) , u ( 0 ) = u 0 ∈ X with periodic boundary conditions, it is proved that, under suitable conditions, the unique (mild) solution u ( t ) = U ( t , 0 ) u ( 0 ) , t ≥ 0 can be expanded by a Volterra series. A recursive algorithm is given to construct the Volterra kernels/series terms and a nonlinear heat equation is discussed to illustrate the proposed method.

SPATIO-TEMPORAL MODELING OF WAVE FORMATION IN AN EXCITABLE CHEMICAL MEDIUM BASED ON A REVISED FITZHUGH–NAGUMO MODEL

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.

A parametric frequency response method for non-linear time-varying systems

A new parametric frequency response algorithm is introduced to investigate linear and non-linear dynamic systems with time-varying parameters. In the new algorithm the time-varying parameters are regarded as additional inputs of the systems and the non-linear generalised frequency response functions for multi-input-single-output systems are then employed to obtain Zadehs system functions from a differential equation representation. The parametric frequency response method reveals how the time-varying parameters affect the behaviour of the systems through a time-varying term. The new method can be applied to both linear and non-linear time-varying systems.

IDENTIFICATION OF A TEMPERATURE DEPENDENT FITZHUGH-NAGUMO MODEL FOR THE BELOUSOV-ZHABOTINSKII REACTION

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.