Authors: Lennart Ljung

Abstract: Identification of nonlinear systems is a very extensive problem, with roots and branches in several diverse fields. It is not possible to survey the area in a short text. The current presentation gives a subjective view on some essential features in the area. These concern a classification of methods, the use of different shades of grey in models, and some overall issues like bias-variance trade-offs, data sparseness and the peril of local minima.

Keywords: nonlinear system identification,neural networks,nonlinear models

Identifier: 10.3182/20060329-3-AU-2901.00009

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Lennart Ljung

Abstract: Identification of nonlinear systems is a very extensive problem, with roots and branches in several diverse fields. It is not possible to survey the area in a short text. The current presentation gives a subjective view on some essential features in the area. These concern a classification of methods, the use of different shades of grey in models, and some overall issues like bias-variance trade-offs, data sparseness and the peril of local minima.

Keywords: nonlinear system identification,neural networks,nonlinear models

Identifier: 10.3182/20060329-3-AU-2901.00085

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Graham C. Goodwin; Juan Carlos Aguero; James S. Welsh; Juan I. Yuz

Abstract: This paper outlines some experiences gained with practical nonlinear system identification.

Keywords:

Identifier: 10.3182/20060329-3-AU-2901.00011

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Jan C. Willems; Ivan Markovsky; Bart De Moor

Abstract: In this presentation, we consider the problem of obtaining the state trajectory directly from an observed vector time-series. We show how the Hankel structure of the data matrix can be exploited in this construction. Both the cases of infinite as well as finite time-series are considered, but only deterministic systems are discussed.

Keywords: most powerful unfalsified model,subspace identification,state construction

Identifier: 10.3182/20060329-3-AU-2901.00043

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Huseyin Akcay; Semiha Turkay; Ata Mugan; Ahmet Kanbolat

Abstract: In this paper, we use a recently developed subspace-based identification algorithm in the modeling of road profile spectra for a variety of roads. We also study the shape filter approach to the correlation between the tracks and re-formulate this as a spectral factorization problem from corrupted spectrum samples, which can effectively be solved again by the subspace-based identification algorithm.

Keywords: estimation,road profile,track modeling,spectra,subspace-based identification,shape filter

Identifier: 10.3182/20060329-3-AU-2901.00222

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Hong Zhao; Michael Harmse; John Guiver; William M. Canney

Abstract: The subspace identification has been available in industrial advanced process control (APC) applications in a commercial APC package since 2000. After five years of industrial use, several issues were identified and addressed; the subspace identification technology has been widely accepted and is now used in numerous industrial APC projects. In this paper, a review of the 5-year application history with industrial experience and recent progress in the plant test and identification is given. Questions like why the parametric model identification (ID) methods, which have been theoretically proved optimal, such as prediction error approach, are not routinely used by industrial APC practitioners and why the APC practitioners are now in favour of subspace ID are answered from a point of view of industrial practice. Several important practical ID issues that are more concerned by industrial practitioners are discussed. More specifically, a new view on the issue of open-loop vs. closed-loop ID is provided with the recent progress in multi-variable constrained plant testing technology. It has been found that the subspace identification works very well with the innovative plant testing approach in a synergistic way, and able to substantially reduce plant test duration. As a result, improved identification efficiency from shorter but richer data sets results in better model accuracy, and consequently project costs have been reduced significantly over the past 4 years. With more applications of the subspace ID in industrial MPC projects, some known issues and the future needs are also provided to invite academic researchers to help address.

Keywords: system identification,subspace methods,model predictive control,industrial control,closed-loop identification,process control

Identifier: 10.3182/20060329-3-AU-2901.00172

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Riccardo Muradore; Enrico Fedrigo

Abstract: The goal of this paper is to describe one application of the subspace identification methods to an adaptive optics system. Adaptive optics systems are extensively used in astronomy to obtain high resolution pictures of stars and galaxies with ground telescopes. Subspace identification is used here as a method to build the MIMO model of the plant.

Keywords: adaptive optics,subspace identification methods,large scale system

Identifier: 10.3182/20060329-3-AU-2901.00150

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Hajime Ase; Kiyotsugu Takaba; Tohru Katayama

Abstract: This paper deals with the subspace identification of a high temperature gasifying and direct melting (GDM) furnace with a boiler-turbine plant for power generation. Based on preliminary statistical analyses using data observed under closed-loop operation, we model the system as a multivariable linear system with four exogenous inputs, two control inputs and three outputs. The goal of this paper is to develop a discrete-time state-space model for the gasifying and direct melting plant by using the two-stage orthogonal decomposition (ORT) method (Katayama et al., 2005). Following a brief description of the two-stage ORT method as applied to the present furnace system, we show models obtained by using closed loop data. Some simulation results of closed-loop performance with a newly designed anti-windup compensator are included, together with a result of real closed-loop operation of GDM plant.

Keywords: gasifying and direct melting (GDM) plant,closed-loop identification,two-stage ORT method,anti-windup compensator,LMI design

Identifier: 10.3182/20060329-3-AU-2901.00170

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Harish J. Palanthandalam-Madapusi; Dennis S. Bernstein; Aaron J. Ridley

Abstract: Two existing Hammerstein-Wiener identification algorithms and a third novel Hammerstein-Wiener identification algorithm are considered for application to the magnetospheric system. A modified subspace algorithm that allows missing data points is described and used for identifying periodically switching Hammerstein-Wiener models, to capture the periodically time-varying nature of the system. These models are used to predict ground-based magnetometer response using the ACE satellite measurements.

Keywords: system identification,identification algorithms,modelling,models prediction,space

Identifier: 10.3182/20060329-3-AU-2901.00082

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Kentaro Kameyama; Akira Ohsumi

Abstract: This paper presents solutions to the IFAC SYSID2006 Benchmark identification problem by applying the subspace identification via distribution-based approach proposed in Ohsumi, Kameyama and Yamaguchi (2002). The application of the algorithm seems to have succeeded in establishing continuous-time models for the benchmark data.

Keywords: benchmark examples,subspace methods,system identification,continuous time systems,distribution-based approach

Identifier: 10.3182/20060329-3-AU-2901.00032

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Marion Gilson; Guillaume Mercere

Abstract: This paper deals with an optimal instrumental variable method dedicated to subspace-based closed-loop system identification. The presented solution is based on the MOESP technique but requires to modify the original scheme by proposing a new PO MOESP method which uses reconstructed past input and past output data as instrumental variables. The developed approach is then illustrated via a simulation example and a comparison with other subspace-based methods.

Keywords: closed-loop identification,MIMO systems,subspace methods,instrumental variable,optimal estimation

Identifier: 10.3182/20060329-3-AU-2901.00171

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Edmund J. Crampin

Abstract: Systems biology is the understanding through computational modelling of the function of biological systems. New high-throughput experimental technologies can measure simultaneously the levels of expression of thousands of genes. The challenge is to extract knowledge from these data sets in order to understand the regulatory machinery of the cell. This article describes recent approaches to gene network modelling, focusing on the issues arising in the attempt to identify regulatory networks directly from high-throughput gene expression data.

Keywords:

Identifier: 10.3182/20060329-3-AU-2901.00007

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Yucai Zhu

Abstract: This work reports the development of an identification technology and its application to advanced process control (APC) in the refining/petrochemical industry. We will introduce model predictive control (MPC) technology which, in the last two decades, has become the standard tool in industrial APC. Model identification plays a crucial role in MPC technology and it is also the most time consuming and difficult task in MPC projects and maintenance. Key issues of identification for MPC will be discussed. The so called ASYM method of identification is outlined that provides systematic solutions to problems from plant test to model validation. Based on the method, both off-line and on-line identification packages have been developed. A large-scale industrial application will be shown. Considerable benefits are obtained using the new identification technology: 1) reduction of identification test time and model building time by over 70%; 2) higher model quality for control; and 3) easier in use. Then, several MPC relevant identification problems will be introduced. Based on his industrial experience, the author will provide an optimistic outlook of future APC/MPC and he will point out that identification technology can play a key role in next generation control systems.

Keywords: system identification,advanced process control,model predictive control

Identifier: 10.3182/20060329-3-AU-2901.00003

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Tomas McKelvey; Ingemar Andersson

Abstract: Future emission regulations for internal combustion engines has lead to an increasing interest in control of the combustion process in order to maximize efficiency while keeping the emission levels at a minimum. One possibility to indirectly measure the combustion process quality is by using a crankshaft mounted torque sensor and use model based signal processing techniques to calculate individual cylinder measures of combustion quality. Since the crankshaft is a flexible device the model must be dynamic. In this contribution we present methodology and results of applying system identification techniques for modeling the crankshaft dynamics. Data is collected from an experimental 5 cylinder spark ignited combustion engine and sampled every 1 degree of the crankshaft revolution. For a fixed engine speed the sampling will be almost identical to time based sampling. Data sets from 7 different engine speeds with static operating conditions are used which results in a data set with 7 different sampling frequencies. A frequency domain approach is adopted to estimate a single parametric continuous time model for all engine speeds. The estimated model is evaluated in two soft sensing applications.

Keywords: system identification,automotive application,modeling,mechanical system,engine management systems,inverse filtering

Identifier: 10.3182/20060329-3-AU-2901.00066

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Rong Deng; Patricia Davies; Anil K. Bajaj

Abstract: A five-parameter fractional derivative model and a hereditary model are being studied to predict polyurethane foam's uniaxial responses under harmonic excitation. A system identification procedure is developed to estimate the model parameters. The prediction results from both models are presented and compared. Both models provide reasonably good prediction of the observed responses from different input levels at a given compression level.

Keywords: system identification,fractional derivative model,hereditary model,polyurethane foam,harmonic response

Identifier: 10.3182/20060329-3-AU-2901.00103

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Le Yi Wang; G. George Yin; Ji-Feng Zhang

Abstract: System identification using set-valued output observations is investigated under a stochastic framework. The problem is especially relevant to system identification problems in which data-flow rates are limited due to computer networking, communications, wireless channels. Properties of empirical measures yield a general result on asymptotic analysis of space and time complexities. The result is employed to guide robust, optimal, and adaptive selections of sensor thresholds. It provides a feasible approach for optimal utility of communication bandwidth resources in enhancing identification accuracy.

Keywords: system identification,estimation,quantized observation,space and time complexity,communication resource allocation

Identifier: 10.3182/20060329-3-AU-2901.00035

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Peter Young

Abstract: The paper outlines the main algorithms and functions available in the CAPTAIN identification and time series analysis Toolbox for Matlab.

Keywords: discrete-time models,continuous-time models,nonstationary time series analysis,nonlinear time series analysis,forecasting,signal extraction

Identifier: 10.3182/20060329-3-AU-2901.00144

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: S. Gillijns; D. S. Bernstein; B. De Moor

Abstract: During the last decade, several suboptimal filtering schemes for data assimilation have been proposed. One of these algorithms, which has succesfully been used in several applications, is the Reduced Rank Square Root filter. In this paper, a numerically more efficient variation, the Reduced Rank Transform Square Root filter, is introduced. A theoretical comparison of both filters is given and their performance is analyzed by comparing assimilation results on a magnetohydrodynamic example which emulates a space storm interacting with the Earth's magnetosphere.

Keywords: Kalman filters,state estimation,Riccati-equations,large-scale systems,partial differential equations,numerical simulation,nonlinear models,systems theory

Identifier: 10.3182/20060329-3-AU-2901.00202

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Freke R. Wink; Malgorzata E. Wilinska; Ludovic J. Chassin; Martin Haluzik; Jan Blaha; Stepan Svacina; Roman Hovorka

Abstract: Critically ill subjects display high temporal variations in insulin sensitivity, which complicate efficient glucose control by glucose controllers. In the present study, we combined a simple, discrete-time model of glucoregulation with the regularisation method to obtain a time-variant marker of insulin sensitivity from hourly plasma glucose measurements and information about intravenous insulin infusion and parenteral intake in six critically ill subjects studied over 24 to 48h. The focus of the study was to determine "basal insulin requirements" representing insulin appearance, which maintains normoglycaemia. We observed up to five-fold variations in the basal insulin requirements over a period of 8-10h in a half of the studied subjects. The remaining subjects had virtually unchanged basal insulin requirements. We conclude that considerable heterogeneity exist in the temporal variation of insulin resistance in critically ill subjects.

Keywords: insulin sensitivity,physiological models,regularisation,parameter estimation,discrete time system

Identifier: 10.3182/20060329-3-AU-2901.00069

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Rachid Malti; Mohamed Aoun; Jocelyn Sabatier; Alain Oustaloup

Abstract: This paper presents a tutorial on system identification using fractional difierentiation models. The tutorial starts with some general aspects on time and frequency-domain representations, time-domain simulation, and stability of fractional models. Then, an overview on system identification methods using fractional models is presented. Both equation-error and output-error-based models are detailed.

Keywords: fractional differentiation,fractional integration,identification,simulation,output error models,equation error models,state variable filter

Identifier: 10.3182/20060329-3-AU-2901.00093

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Torsten Soderstrom; Erik Larsson; Kaushik Mahata; Magnus Mossberg

Abstract: Continuous-time identification is applied to an errors-in-variables setting. A continuous-time model is fitted to data consisting of discrete-time noise corrupted input and output measurements. The noise-free input is modelled as a continuous-time ARMA process. It is described how the Cramér-Rao lower bound for the estimation problem can be computed. Several parameter estimation approaches for the problem are presented, and also illustrated in a short numerical study.

Keywords: system identification,continuous-time,errors-in-variables

Identifier: 10.3182/20060329-3-AU-2901.00064

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Markus Gerdin

Abstract: If a model structure is not identifiable, then it is not possible to uniquely identify its parameters from measured data. This contribution describes how solvers for differential-algebraic equations (DAE) can be used to examine if a model structure is locally identifiable. The procedure can be applied to both linear and nonlinear systems. If a model structure is not identifiable, it is also possible to examine which functions of the parameters that are locally identifiable.

Keywords: identifiability,nonlinear systems,modelling,identification,descriptor systems

Identifier: 10.3182/20060329-3-AU-2901.00127

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Graham C. Goodwin; James S. Welsh; Arie Feuer; Milan Derpich

Abstract: In this paper we propose a new approach to robust optimal experiment design. The key departure from earlier work is that we specifically account for the fact that, prior to the experiment, we possess only partial knowledge of the system. We also give a detailed analysis of the solution for a simple case and propose a concave optimization algorithm that can be applied more generally.

Keywords: experiment design,optimal input design

Identifier: 10.3182/20060329-3-AU-2901.00220

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Abd-Krim Seghouane; Shun-ichi Amari

Abstract: The Akaike information criterion, AIC, is a widely used tool for model selection. AIC is derived as an asymptotically unbiased estimator of a function used for ranking candidate models which is a variant of the Kullback-Leibler divergence between the true model and the approximating candidate model. Despite the Kullback-Leibler's computational and theoretical advantages, what can become a nuisance in model selection applications is its lack of symmetry. Simple examples can show that reversing the role of the arguments in the Kullback-Leibler divergence can yield substantially different results. In this paper, three new functions for ranking candidate models are proposed. These functions are constructed by symmetrizing the Kullback-Leibler divergence between the true model and the approximating candidate model. The operations used for symmetrizing are the average, geometric and harmonic means. It is found that the original AIC criterion is an asymptotically unbiased estimator of these three difierent functions. A simulation study based on polynomial regression is also provided to compare the difierent proposed ranking functions with the AIC asymptotic estimation.

Keywords: akaike information criterion,Kullback-Leibler divergence,model selection,geometric and harmonic means

Identifier: 10.3182/20060329-3-AU-2901.00130

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006

Authors: Piet M. T. Broersen

Abstract: The program ARMAsel automatically selects a single time series model for given stationary stochastic data. Three model types are candidates for selection. The type can be AR or autoregressive, MA or moving average and the combined ARMA type. The parameters of that selected model accurately represent the power spectral density and the autocorrelation function of the data. The reduced statistics ARMAsel-rs algorithm uses a long AR model as only input to compute models of the other types and to select the best. ARMAsel-mis is a new program that can handle missing data. It computes AR models with a numerically stable maximum likelihood algorithm and uses ARMAsel-rs to determine MA and ARMA models. The order and the type of the best candidate are automatically selected with dedicated missing data criteria, supposing that data are randomly missing. Missing less than 10 % of the data has not much influence on the accuracy, missing 50 % generally looses a factor two in the accuracy in comparison with the model estimated from the same number of contiguous observations. Low order time series models can be computed as long as the product of the remaining fraction and the remaining number of observations is greater than about 10.

Keywords: ARMA model,autoregressive model,autocovariance estimation,missing observations,order selection,parametric model,spectral estimation

Identifier: 10.3182/20060329-3-AU-2901.00109

Conference: 14th IFAC Symposium on System Identification, 2006

Location: , Australia

Start Date: Wed Mar 29 2006 - End Date: Fri Mar 31 2006