Online ISSN
1751-8652
Print ISSN
1751-8644
IET Control Theory & Applications
Volume 5, Issue 7, 5 May 2011
Volumes & issues:
Volume 5, Issue 7
5 May 2011
-
- Author(s): H. Garnier ; T. Söderström ; J.I. Yuz
- Source: IET Control Theory & Applications, Volume 5, Issue 7, p. 839 –841
- DOI: 10.1049/iet-cta.2011.9043
- Type: Article
- + Show details - Hide details
-
p.
839
–841
(3)
- Author(s): J.I. Yuz ; J. Alfaro ; J.C. Agüero ; G.C. Goodwin
- Source: IET Control Theory & Applications, Volume 5, Issue 7, p. 842 –855
- DOI: 10.1049/iet-cta.2010.0246
- Type: Article
- + Show details - Hide details
-
p.
842
–855
(14)
In this study, we apply the expectation-maximisation (EM) algorithm to identify continuous-time state-space models from non-uniformly fast-sampled data. The sampling intervals are assumed to be small and uniformly bounded. The authors use a parameterisation of the sampled-data model in incremental form in order to modify the standard formulation of the EM algorithm for discrete-time models. The parameters of the incremental model converge to the parameter of the continuous-time system description as the sampling period goes to zero. The benefits of the proposed algorithm are successfully demonstrated via simulation studies. - Author(s): M. Bergamasco and M. Lovera
- Source: IET Control Theory & Applications, Volume 5, Issue 7, p. 856 –867
- DOI: 10.1049/iet-cta.2010.0228
- Type: Article
- + Show details - Hide details
-
p.
856
–867
(12)
This study deals with the problem of continuous-time model identification and presents two subspace-based algorithms capable of dealing with data generated by systems operating in closed loop. The algorithms are developed by reformulating the identification problem from the continuous-time model to equivalent ones to which discrete-time subspace identification techniques can be applied. More precisely, two approaches are considered, the former leading to the so-called all-pass domain by using a bank of Laguerre filters applied to the input–output data and the latter corresponding to the projection of the input–output data onto an orthonormal basis, again defined in terms of Laguerre filters. In both frameworks, the Predictor-Based Subspace Identification, originally developed in the case of discrete-time systems, can be reformulated for the continuous-time case. Simulation results are used to illustrate the achievable performance of the proposed approaches with respect to existing methods available in the literature. - Author(s): X. Liu ; J. Wang ; W.X. Zheng
- Source: IET Control Theory & Applications, Volume 5, Issue 7, p. 868 –877
- DOI: 10.1049/iet-cta.2010.0211
- Type: Article
- + Show details - Hide details
-
p.
868
–877
(10)
The refined instrumental variable method for continuous-time systems, abbreviated as the RIVC method, has been well accepted and used successfully for years in many disciplines. This study fills up a theoretical gap by proving the convergence property of the RIVC method: under some mild assumptions, the estimate from the RIVC method locally converges in one iteration to the true parameter in the asymptotic case. A numerical example is presented to demonstrate the convergence property of the RIVC method. - Author(s): V. Laurain ; R. Tóth ; M. Gilson ; H. Garnier
- Source: IET Control Theory & Applications, Volume 5, Issue 7, p. 878 –888
- DOI: 10.1049/iet-cta.2010.0218
- Type: Article
- + Show details - Hide details
-
p.
878
–888
(11)
Controllers in the linear parameter-varying (LPV) framework are commonly designed in continuous time (CT) requiring accurate and low-order CT models of the system. However, identification of CT-LPV models is largely unsolved, representing a gap between the available LPV identification methods and the needs of control synthesis. In order to bridge this gap, direct identification of CT-LPV systems in an input–output setting is investigated, focusing on the case when the noise part of the data generating system is an additive discrete-time (DT) coloured noise process. To provide consistent model parameter estimates in this setting, a refined instrumental variable (IV) approach is proposed and its properties are analysed based on the prediction-error framework. The benefits of the introduced direct CT-IV approach over identification in the DT case are demonstrated through a representative simulation example inspired by the Rao–Garnier benchmark. - Author(s): J.-D. Gabano ; T. Poinot ; H. Kanoun
- Source: IET Control Theory & Applications, Volume 5, Issue 7, p. 889 –899
- DOI: 10.1049/iet-cta.2010.0222
- Type: Article
- + Show details - Hide details
-
p.
889
–899
(11)
In this study, the authors present an identification procedure of a non-linear thermal system that exhibits a diffusive interface. Indeed, the system deals with heat conduction in a homogeneous material through a wall. Heating front face temperature dynamics when random heat flux is applied to the wall's front face area are considered. If the thermophysical properties of the material depend on the temperature, the heat equation is not linear anymore. Hence, in order to explicitly take into account the dependence of the system dynamics on the temperature, a fractional continuous linear parameter-varying model is used and determined thanks to a local approach composed of two steps. First, the authors show that for small heating temperature variations, the heat diffusion can be modelled with the help of a local fractional model based on a fractional integrator operator of order (1/2), which acts only over a limited spectral band. The transfer function parameters of this model are estimated thanks to an output-error technique at different initial temperature-operating points. Secondly, the transfer function parameters dependence with respect to the initial temperature of the material is obtained by direct polynomial interpolation. Simulations using ARMCO iron are used to demonstrate the performances of the proposed local approach. - Author(s): A. Narang ; S.L. Shah ; T. Chen
- Source: IET Control Theory & Applications, Volume 5, Issue 7, p. 900 –912
- DOI: 10.1049/iet-cta.2010.0718
- Type: Article
- + Show details - Hide details
-
p.
900
–912
(13)
Modelling of real physical systems having long memory transients and infinite dimensional structures using fractional-order dynamic models has significantly attracted interest over the last few years. For this reason, many identification techniques both in the frequency domain and time domain have been developed to model these fractional-order systems. However, in many processes time delays are also present and estimation of time delays along with continuous-time fractional-order model parameters have not been addressed anywhere. This study deals with the continuous-time model identification of fractional-order system models with time delays. In this study, a new linear filter is introduced for simultaneous estimation of all model parameters for commensurate fractional-order system models with time delays. The proposed method simultaneously estimates time delays along with other model parameters in an iterative manner by solving simple linear regression equations. For the case when the fractional order is unknown, we also propose a nested loop optimisation method where the time delay along with other model parameters are estimated iteratively in the inner loop and the fractional order is estimated in the non-linear outer loop. The applicability of the developed procedure is demonstrated by simulations on a fractional-order system model by doing Monte Carlo simulation analysis in the presence of white noise. The proposed algorithm has also been applied to identify a process of thermal diffusion in a wall in simulation, which are characterised by fractional-order behaviour. - Author(s): A.H. Tan and C.L. Cham
- Source: IET Control Theory & Applications, Volume 5, Issue 7, p. 913 –922
- DOI: 10.1049/iet-cta.2010.0213
- Type: Article
- + Show details - Hide details
-
p.
913
–922
(10)
This study considers the identification of a continuous-time model for a cooling system. Direct and indirect approaches are compared for frequency domain identification using a periodic signal. First, the parameters of the time-invariant dynamics are estimated offline by compensating for a variable time delay using delay reconciliation, leading to more accurate estimates. Subsequently, the time delay is estimated adaptively online through a simple gridding method. - Author(s): J. Lataire and R. Pintelon
- Source: IET Control Theory & Applications, Volume 5, Issue 7, p. 923 –933
- DOI: 10.1049/iet-cta.2010.0223
- Type: Article
- + Show details - Hide details
-
p.
923
–933
(11)
A frequency-domain least-squares estimator is presented for identifying linear, continuous-time, time-varying dynamical systems. The model considered is a linear, ordinary differential equation whose coefficients vary as polynomials in time. A frequency-domain approach is used, thus allowing the user to determine easily the frequency band(s) of interest. It is shown that the bias errors because of windowing and sampling the continuous-time signals can be modelled by a polynomial function of the frequency. The regression matrices of the estimators are shown to be very efficiently computed using the fast Fourier transform algorithm and its inverse. The total least-squares, generalised total least-squares and weighted, non-linear least-squares estimators are constructed. The latter two are shown to be consistent. The estimators are illustrated on simulation and measurement data. - Author(s): C. Casenave and G. Montseny
- Source: IET Control Theory & Applications, Volume 5, Issue 7, p. 934 –942
- DOI: 10.1049/iet-cta.2010.0229
- Type: Article
- + Show details - Hide details
-
p.
934
–942
(9)
The authors introduce a new identification method for general causal convolution models of the form u↦h*u=H(∂t)u, where h is the impulse response of the system, to be identified from measurement data. This method is based on a suitable parameterisation of operator H(∂t) deduced from the so-called diffusive representation, devoted to state representations of such integral operators. Following this approach, the complex dynamic features of H(∂t) can be summarised by a few numerical parameters on which the identification method will focus. The class of concerned convolution operators includes rational as well as non-rational ones, even of complex nature. For illustration, we implement this method on a numerical example.
Editorial: Continuous-time model identification
Identification of continuous-time state-space models from non-uniform fast-sampled data
Continuous-time predictor-based subspace identification using Laguerre filters
Convergence analysis of refined instrumental variable method for continuous-time system identification
Direct identification of continuous-time linear parameter-varying input/output models
Identification of a thermal system using continuous linear parameter-varying fractional modelling
Continuous-time model identification of fractional-order models with time delays
Continuous-time model identification of a cooling system with variable delay
Frequency-domain weighted non-linear least-squares estimation of continuous-time, time-varying systems
Identification and state realisation of non-rational convolution models by means of diffusive representation
Most viewed content for this Journal
Article
content/journals/iet-cta
Journal
5
Most cited content for this Journal
-
Finite-time stability of interconnected impulsive switched systems
- Author(s): Guangdeng Zong ; Hangli Ren ; Linlin Hou
- Type: Article
-
Event-based security control for discrete-time stochastic systems
- Author(s): Derui Ding ; Zidong Wang ; Guoliang Wei ; Fuad E. Alsaadi
- Type: Article
-
Survey on semi-tensor product method with its applications in logical networks and other finite-valued systems
- Author(s): Jianquan Lu ; Haitao Li ; Yang Liu ; Fangfei Li
- Type: Article
-
Optimal control for networked control systems with disturbances: a delta operator approach
- Author(s): Yuan Yuan ; Huanhuan Yuan ; Zidong Wang ; Lei Guo ; Hongjiu Yang
- Type: Article
-
Filtering-based iterative identification for multivariable systems
- Author(s): Yanjiao Wang and Feng Ding
- Type: Article