Authors: Oku, Hiroshi

Abstract: This paper presents an on-line change detection method that uses a recursive subspace model identification algorithm as a residual generator. In our method, a change in the variance of the residual is monitored by a statistical change detector based on the log-likelihood ratio test. An experiment using a real-life cart system as well as a numerical example demonstrates that the proposed method can detect changes in the dynamics of a system, without being disturbed by changes in the dynamics of an input signal which are not our concern.

Keywords: Fault Detection and Diagnosis; Subspace Methods; Recursive Identification

Identifier: 10.3182/20090706-3-FR-2004.00163

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Bai, Er-Wei

Abstract: A data driven orthogonal basis function approach is proposed for non-parametric FIR nonlinear system identification. This eliminates the problem of blindly choosing the basis functions without a priori structural information. Extension to deterministic inputs are also presented. Further, based on the proposed basis functions, approaches are proposed for model order determination and regressor selection along with their theoretical justifications.

Keywords: Nonlinear System Identification; Nonparametric Methods

Identifier: 10.3182/20090706-3-FR-2004.00220

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Fey, Dirk, Bullinger, Eric

Abstract: Estimation of kinetic parameters is a key step in modelling biochemical reaction networks as, often, their direct estimation is expensive, time-consuming or even infeasible. This article proposes a parameter estimation procedure, which explicitly takes into account the model structure of the biological systems. The convergence is guaranteed using a dissipativity argument and a coordinate transformation yielding a parameter-free system description. The application to a basic enzyme kinetic model illustrates the proposed methodology.

Keywords: Biological Systems; Nonlinear System Identification; Continuous Time System Estimation

Identifier: 10.3182/20090706-3-FR-2004.00209

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Guidorzi, Roberto, Diversi, Roberto

Abstract: The extension of the Frisch scheme from the original algebraic case to the dynamic one leads to the use of errors–in–variables models where the measurements of the input and output are affected by additive white and independent noises. This problem admits a single solution when the assumptions of the scheme are exactly fulfilled but its application to real processes requires the introduction of specific model selection criteria. This paper analyzes the additional problems encountered in the extension of Frisch identification to the multivariable case and introduces a geometric approach for its solution.

Keywords: Multivariable System Identification; Errors in Variables Identification

Identifier: 10.3182/20090706-3-FR-2004.00268

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Wills, Adrian George, Mills, Adam, Ninness, Brett

Abstract: This paper describes a Matlab based software environment for the estimation of dynamic systems. It has been developed primarily as a vehicle for profiling novel approaches relative to existing methods within a common software framework in order to streamline comparisons. Key features of the toolbox include simplicity of use (particularly via automated entry of unspecified values) and the support of a wide range of scalar and multivariable model structures. The development of this software is an ongoing project, with earlier progress being reported on previously. This paper details recent advancements, including the provision of a graphical user interface environment.

Keywords: Toolboxes

Identifier: 10.3182/20090706-3-FR-2004.00123

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Ionescu, Clara, Oustaloup, Alain, Levron, Françoıs, Melchior, Pierre, Sabatier, Jocelyn, De Keyser, Robin M.C.

Abstract: The respiratory system has specific geometrical and material properties, which allow researchers to classify it as a typical fractal structure. Hitherto, only material properties in animal and human lung parenchyma have been investigated, assuming a power-law behavior of the viscoelastic properties in soft biological tissue. Consequently, lumped, fractional-order parametric models have been used to characterize such power-law behavior, in both healthy and pathologic lungs. This paper attempts to verify if the appearance of the fractional-order operator is also related to the underlying geometry and structure of the respiratory tree. Typical morphologic values are used in an electrical equivalent, based on our previous results. Simulation results show that the dichotomous and recursive, fractal-like properties of the respiratory system lead naturally to the appearance of the fractional-order operators.

Keywords: Biological Systems; Model Validation; Frequency Domain Identification

Identifier: 10.3182/20090706-3-FR-2004.00165

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Ferranti, Francesco, Rolain, Yves, Vandermot, Koen, Knockaert, Luc, Dhaene, Tom

Abstract: This paper modifies a recent robust parametric macromodeling technique called Multivariate Orthonormal Vector Fitting (MOVF), to handle noisy data in an output error estimation framework. The new method provides accurate and compact rational parametric macromodels based on measurements in the frequency domain. The performance of the multivariate method is shown on simulation as well as on real measurements.

Keywords: Multivariable System Identification; Frequency Domain Identification; Continuous Time System Estimation

Identifier: 10.3182/20090706-3-FR-2004.00271

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Pizzileo, Barbara, Li, Kang, Irwin, George W.

Abstract: In identification of complex dynamic systems using fuzzy neural networks, one of the main issues is the curse-of-dimensionality, which makes it difficult to quickly compute all the parameters associated with the network with all possible inputs and rules being included. In the literature this issue has been addressed by the selection of either the inputs or the rules. This is due to the fact that not all possible inputs or rules have to be necessarily included because of the correlations between them. Adding unnecessary inputs or rules simply increases the model complexity and worsens the network generalization performance. Selecting the best set of inputs or rules is a combinational problem and can be computationally too expensive. In this paper, the problem is solved by first proposing a refinement procedure for rule selection. The algorithm is then adapted and integrated with prior input selection to further improve the model accuracy. Simulation results confirm the efficacy of the method.

Keywords: Nonlinear System Identification; Neural Networks; Machine Learning and Data Mining

Identifier: 10.3182/20090706-3-FR-2004.00250

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Ohsumi, Akira, Kimura, Takuro, Kono, Michio

Abstract: In this paper, a novel approach to identify an unknown parameter vector of the exogenous input to a class of stochastic linear systems is proposed. The key of the approach is to introduce an additional information about the unknown parameter vector which is called the pseudomeasurement. Augmenting this pseudomeasurement with the original observation data, the identification of unknown vector as well as the state estimation is performed. The efficacy of the proposed approach is confirmed by simulation studies.

Keywords: Identification for Control; Multivariable System Identification

Identifier: 10.3182/20090706-3-FR-2004.00048

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Delgado, Ramón A., Yuz, Juan I., Aguero, Juan C, Goodwin, Graham C.

Abstract: Different maximum likelihood formulations have been proposed in the literature for dynamic system identification in the time and frequency domains. In this paper we present numerical examples to study and compare these approaches for short and long data sets. In particular, in the time domain, different likelihood functions are obtained depending on whether or not the initial state is considered as a random vector, as a deterministic parameter, or equal to zero. Similar assumptions can be made in the frequency domain regarding an extra term that contains the difference between the initial and final state.

Keywords: Maximum Likelihood Methods; Frequency Domain Identification

Identifier: 10.3182/20090706-3-FR-2004.00188

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Lauer, Fabien, Vidal, Rene, Bloch, Gerard

Abstract: We propose a general framework for identification of linear discrete-time hybrid systems in which arbitrary loss functions can be easily included. Our framework includes the algebraic (Vidal et al., 2003) and support vector regression (Lauer and Bloch, 2008a,b) methods as particular cases. Inspired by these approaches, we then propose an optimization framework that relies on the minimization of a product of loss functions. Here, the identification problem is recast as a nonlinear and non-convex, though continuous, optimization program that involves only the model parameters as variables. As a result, its complexity scales linearly with the number of data and it can easily be solved using standard global optimization methods. Moreover, we show that by choosing a saturated loss function, such as Hampel’s loss function, the algorithm can efficiently deal with noise and outliers in the data. The final result is a general framework for linear hybrid system identification that can deal efficiently with noise, outliers, and large data sets. Numerical experiments demonstrate the efficiency and robustness of the proposed approach.

Keywords: Hybrid and Distributed System Identification; Machine Learning and Data Mining

Identifier: 10.3182/20090706-3-FR-2004.00093

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Welsh, James, Rojas, Cristian

Abstract: Robust optimal experiment design is an infinite dimensional optimisation problem. Typically it is solved by discretisation of the design space resulting in a discrete semi-infinite convex programming problem which is computationally expensive. In this paper we propose a new computational approach to solve robust optimal experiment design problems based on a recently developed method for robust convex optimisation known as the `scenario approach'.

Keywords: Input and Excitation Design

Identifier: 10.3182/20090706-3-FR-2004.00031

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Vazquez, Emmanuel, Bect, Julien

Abstract: This paper deals with the problem of estimating the probability of failure of a system, in the challenging case where only an expensive-to-simulate model is available. In this context, the budget for simulations is usually severely limited and therefore classical Monte~Carlo methods ought to be avoided. We present a new strategy to address this problem, in the framework of sequential Bayesian planning. The method uses kriging to compute an approximation of the probability of failure, and selects the next simulation to be conducted so as to reduce the mean square error of estimation. By way of illustration, we estimate the probability of failure of a control strategy in the presence of uncertainty about the parameters of the plant.

Keywords: Error Quantification; Bayesian Methods; Nonparametric Methods

Identifier: 10.3182/20090706-3-FR-2004.00090

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Bettinger, Régis, Pronzato, Luc, Duchêne, Pascal

Abstract: A design method presented in a previous paper for the sequential generation of observation sites used for the inversion of a prediction model is extended to cope with practical issues such as delayed observations and design of batches of imposed size. The final objective of the construction is to be able to associate with any target T in the output space a value xT of the input factors such that the response of the system at xT will be “close” to T (from an industrial point of view, xT corresponds to manufactory conditions that yield a product whose feature of interest is described by T). The problem is thus much different from the more standard one where one wishes to build a precise model over the whole input space: here the model only has to be precise over a set of values xT that permit to reach any target T, that is, the observation sites should not be spread over the entire admissible input space, but should rather concentrate in areas that cover the reachable output space when mapped by the system. Examples in low dimensions are presented that illustrate the behavior of the method and allow a comparison to be made with a standard sequential method for designing exploratory experiments.

Keywords: Grey Box Modelling; Maximum Likelihood Methods

Identifier: 10.3182/20090706-3-FR-2004.00216

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Novara, Carlo, Ruiz, Fredy, Milanese, Mario

Abstract: A simple approach to the design of Linear Parameter Varying (LPV) filters for nonlinear systems is proposed. The approach relies on individuating several working conditions of the system. For each of these conditions, an almost optimal Linear Time Invariant (LTI) filter is directly identified from data. The designed LTI filters are then suitably combined to give an LPV filter. The identification of a nonlinear model and the design of a nonlinear filter are thus avoided. The direct filter design approach is applied to a problem involving real data, regarding the estimation of vehicles yaw rate.

Keywords: Filtering and Smoothing; Nonlinear System Identification; Mechanical and Aerospace

Identifier: 10.3182/20090706-3-FR-2004.00057

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Young, Peter, Garnier, Hugues, Gilson, Marion

Abstract: The paper describes a simple, two-stage instrumental variable method of closed loop identification and estimation. This can be used with both continuous and discrete-time transfer function models and the enclosed system can be unstable. The paper also shows briefly how a third stage of estimation can be added that induces statistical efficiency when the enclosed system is stable.

Keywords: Continuous Time System Estimation; Closed Loop Identification; Identification for Control

Identifier: 10.3182/20090706-3-FR-2004.00191

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Straka, Ondrej, Flídr, Miroslav, Dunik, Jindrich, Simandl, Miroslav

Abstract: The goal of the article is to describe a software framework designed for nonlinear state estimation of discrete time dynamic systems. The framework was designed with the aim to facilitate implementation, testing and use of various nonlinear state estimation methods in mind. The main strength of the framework is its versatility due to the possibility of either structural or probabilistic description of the problem. Besides the well-known basic nonlinear estimation methods such as the extended Kalman filter, the divided difference filters and the unscented Kalman filter, the framework implements particle filter with advanced features as well. As the framework is designed on the object oriented basis, further extension by user-specified nonlinear estimation algorithms is extremely easy. The paper provides a brief introduction into nonlinear state estimation problem and describes the individual components of the framework, their key features and use. The strengths of the framework are presented in two examples.

Keywords: Toolboxes; Bayesian Methods; Filtering and Smoothing

Identifier: 10.3182/20090706-3-FR-2004.00084

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Said, Salem, Le Bihan, Nicolas, Sangwine, Stephen J.

Abstract: Local linearization is employed in the analysis of rotation time series in order to overcome difficulties related to non linear constraints and non commutativity. This paper obtains results on the stability and invariance properties of local linearization and their importance to applications.

Keywords: Filtering and Smoothing; Time Series

Identifier: 10.3182/20090706-3-FR-2004.00241

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Ase, Hajime, Katayama, Tohru, Tanaka, Hideyuki

Abstract: We develop a state-space method of identifying a Wiener-Hammerstein system, where a nonlinearity is sandwiched by two linear systems. By dividing it into the linear system and Hammerstein system composed of the nonlinearity and the second linear system, we propose an iterative method of identifying the Hammerstein system by the orthogonal decomposition subspace method (ORT) and the linear system by minimizing the square norm of output prediction error, for which the identified Hammerstein model plays a role of instrument of measuring the output of the linear system. The data driven local coordinate (DDLC)-based gradient method is applied to this minimization. Numerical results for the benchmark problem are included to show the applicability of the present method.

Keywords: Nonlinear System Identification

Identifier: 10.3182/20090706-3-FR-2004.00181

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Straka, Ondrej, Simandl, Miroslav

Abstract: The paper deals with the particle filter in discrete-time nonlinear non-Gaussian system state estimation. One of the key parameters affecting estimate quality of the particle filter is the sample size. In the literature, there is a number of techniques coming from various ideas that aim at adapting the sample size while keeping quality in some sense fixed. The goal of the paper is to provide a survey of sample size adaptation techniques, to classify them and to discuss various aspects concerning the techniques.

Keywords: Particle Filtering/Monte Carlo Methods; Filtering and Smoothing

Identifier: 10.3182/20090706-3-FR-2004.00226

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Schwartz, Jay D., Rivera, Daniel E.

Abstract: Efficient supply chain management is a crucial imperative for modern, global enterprises. Tactical decision policies based on process control principles have been developed in the literature for managing production-inventory systems and supply chain networks. To be effective these decision policies depend on accurate nominal models. With a discrete-event simulation acting as a "truth model,'' we employ system identification techniques to parameterize a nonlinear Partial Differential Equation (PDE) model of the semiconductor manufacturing process. A case study shows that the identified PDE model can accurately predict the output of the discrete-event simulation, but without the high computational burden.

Keywords: Other; Hybrid and Distributed System Identification; Nonlinear System Identification

Identifier: 10.3182/20090706-3-FR-2004.00160

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Sjoberg, Jonas E., Hjalmarsson, Håkan

Abstract: In this contribution we describe a new signals, systems and identification toolbox for the symbolic and numerical computation system Mathematica. The toolbox provides functionality for computation of properties of systems and signals ranging from frequency responses, zeros and poles to signal spectra and spectral factorizations. It also includes a wide range of identification algorithms ranging from spectral analysis to subspace and prediction error identification of models for non-linear systems. The symbolic capabilities of Mathematica are used to allow the user to construct very general model structures, and for pre-processing, such as gradient calculations, when optimizing the parameters in such structures.

Keywords: Toolboxes; Nonlinear System Identification; Subspace Methods

Identifier: 10.3182/20090706-3-FR-2004.00124

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Chen, Han Fu

Abstract: A unified approach to recursively identifying ARMAX systems, Hammerstein systems, Wiener systems, and nonlinear ARX systems is presented, by which the problem is solved by two steps. First, the task of identification is transformed to a root-seeking problem by selecting a regression function, whose roots coincide with the estimated parameters, and by forming an available value called ``observation" at each time. However, the resulting root-seeking problem is hardly to be solved by the classical Robbins-Monro (RM) algorithm. Instead, the stochastic approximation (SA) algorithm with expanding truncations (SAAWET) and its general convergence theorem (GCT) serve as the main tool. The second step of the unified approach is to verify conditions required by GCT. Reasonable conditions are respectively given for each system mentioned above, under which the estimates given by the recursive algorithms are strongly consistent.

Keywords: Recursive Identification; Nonlinear System Identification

Identifier: 10.3182/20090706-3-FR-2004.00069

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Mikhaylova, Ljubov, Alexandrov, A. G., Orlov, Juriy

Abstract: Description of ADAPLAB-3 (MATLAB ToolBox) for finite-frequency identification and adaptation is given. As distinct from known MATLAB ToolBoxes, ADAPLAB-3 algoritms proceed from assumption that an external disturbance applied to a plant and a measurement noise are unknown but bounded functions. A test signal as a sum of a minimal number of harmonics is used for identification. An adaptive control is formed on the base of H-infinity optimization and the results of identification of the plant and a closed-loop system. ADAPLAB-3 has some substantial differences from the early developed toolbox ADAPLAB-M: the discrete-time plants are considered; the amplitudes and the frequencies of the test signal are tuned automatically during the identification and adaptation process; the time of identification and adaptation is determined automatically in dependence on the current external disturbances and the noise.

Keywords: Toolboxes; Frequency Domain Identification; Identification for Control

Identifier: 10.3182/20090706-3-FR-2004.00082

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009

Authors: Chen, Xinkai

Abstract: The piezo-actuated nano-positioner is composed of a piezo electric actuator (PEA) and a positioning mechanism (PM). Due to the existence of hysteretic nonlinearity in the PEA and the friction behavior in the PM, the accurate position control of the piezo-actuated stage is a challenging task. This paper discusses the adaptive sliding mode control for the piezo-actuated nano-positioner, where the hysteresis is described by Prandtl-Ishlinskii model. This paper tries to fuse the hysteresis model with the adaptive control techniques, where the real value of the parameters of the stage need neither to be identified nor to be measured. The proposed control law ensures the global stability of the controlled nano-positioner, and the position error can be controlled to be as small as required by choosing the design parameters. Experimental results show the effectiveness of the proposed method.

Keywords: Mechanical and Aerospace; Identification for Control

Identifier: 10.3182/20090706-3-FR-2004.00290

Conference: 15th IFAC Symposium on System Identification, 2009

Location: Saint-Malo Convention Center, Saint-Malo, France

Start Date: Mon Jul 06 2009 - End Date: Wed Jul 08 2009