Showing papers on "System identification published in 1999"
TL;DR: The paper describes a general methodology for the fitting of measured or calculated frequency domain responses with rational function approximations by replacing a set of starting poles with an improved set of poles via a scaling procedure.
Abstract: The paper describes a general methodology for the fitting of measured or calculated frequency domain responses with rational function approximations. This is achieved by replacing a set of starting poles with an improved set of poles via a scaling procedure. A previous paper (Gustavsen et al., 1997) described the application of the method to smooth functions using real starting poles. This paper extends the method to functions with a high number of resonance peaks by allowing complex starting poles. Fundamental properties of the method are discussed and details of its practical implementation are described. The method is demonstrated to be very suitable for fitting network equivalents and transformer responses. The computer code is in the public domain, available from the first author.
2,950 citations
TL;DR: In this paper, a novel approach of stochastic subspace identification is presented that incorporates the idea of the reference sensors already in the identification step: the row space of future outputs is projected into the rowspace of past reference outputs.
1,236 citations
TL;DR: A new projection approach to closed-loop identification with the advantage of allowing approximation of the open-loop dynamics in a given, and user-chosen frequency domain norm, even in the case of an unknown, nonlinear regulator.
742 citations
Book•
01 Jan 1999713 citations
01 Jan 1999
TL;DR: This thesis phrases the application of terrain navigation in the Bayesian framework, and develops a numerical approximation to the optimal but intractable recursive solution, and derives explicit expressions for the Cramer-Rao bound of general nonlinear filtering, smoothing and prediction problems.
Abstract: Recursive estimation deals with the problem of extracting information about parameters, or states, of a dynamical system in real time, given noisy measurements of the system output. Recursive estimation plays a central role in many applications of signal processing, system identification and automatic control. In this thesis we study nonlinear and non-Gaussian recursive estimation problems in discrete time. Our interest in these problems stems from the airborne applications of target tracking, and autonomous aircraft navigation using terrain information.In the Bayesian framework of recursive estimation, both the sought parameters and the observations are considered as stochastic processes. The conceptual solution to the estimation problem is found as a recursive expression for the posterior probability density function of the parameters conditioned on the observed measurements. This optimal solution to nonlinear recursive estimation is usually impossible to compute in practice, since it involves several integrals that lack analytical solutions.We phrase the application of terrain navigation in the Bayesian framework, and develop a numerical approximation to the optimal but intractable recursive solution. The designed point-mass filter computes a discretized version of the posterior filter density in a uniform mesh over the interesting region of the parameter space. Both the uniform mesh resolution and the grid point locations are automatically adjusted at each iteration of the algorithm. This Bayesian point-mass solution is shown to yield high navigation performance in a simulated realistic environment.Even though the optimal Bayesian solution is intractable to implement, the performance of the optimal solution is assessable and can be used for comparative evaluation of suboptimal implementations. We derive explicit expressions for the Cramer-Rao bound of general nonlinear filtering, smoothing and prediction problems. We consider both the cases of random and nonrandom modeling of the parameters. The bounds are recursively expressed and are connected to linear recursive estimation. The newly developed Cramer-Rao bounds are applied to the terrain navigation problem, and the point-mass filter is verified to reach the bound in exhaustive simulations.The uniform mesh of the point-mass filter limits it to estimation problems of low dimension. Monte Carlo methods offer an alternative approach to recursive estimation and promise tractable solutions to general high dimensional estimation problems. We provide a review over the active field of statistical Monte Carlo methods. In particular, we study the particle filters for recursive estimation. Three different particle filters are applied to terrain navigation, and evaluated against the Cramer-Rao bound and the point-mass filter. The particle filters utilize an adaptive grid representation of the filter density and are shown to yield a performance equal to the point-mass method.A Markov Chain Monte Carlo (MCMC) method is developed for a highly complex data association problem in target tracking. This algorithm is compared to previously proposed methods and is shown to yield competitive results in a simulation study.
577 citations
TL;DR: In this paper, the authors present various applications of neural networks in energy problems in a thematic rather than a chronological or any other order, including modeling the heat-up response of a solar steam-generating plant, estimation of a parabolic trough collector intercept factor, and the estimation of the local concentration ratio.
431 citations
01 Apr 1999
TL;DR: An alternate ANN structure called functional link ANN (FLANN) for nonlinear dynamic system identification using the popular backpropagation algorithm and performs as good as and in some cases even better than the MLP structure for the problem of nonlinear system identification.
Abstract: We have presented an alternate ANN structure called functional link ANN (FLANN) for nonlinear dynamic system identification using the popular backpropagation algorithm. In contrast to a feedforward ANN structure, i.e., a multilayer perceptron (MLP), the FLANN is basically a single layer structure in which nonlinearity is introduced by enhancing the input pattern with nonlinear functional expansion. With proper choice of functional expansion in a FLANN, this network performs as good as and in some cases even better than the MLP structure for the problem of nonlinear system identification.
382 citations
TL;DR: In this paper, the output-only modal analysis has typically been approached by applying a peak-picking technique to the auto and cross-powers of the measured responses, resulting in operational deflection shapes and approximate estimates for the resonance frequencies.
380 citations
TL;DR: In this paper, a memory-based technique for local modeling and control of unknown non-linear dynamical systems is proposed, which uses a query-based approach to select the best model configuration by assessing and comparing different alternatives.
Abstract: This paper presents local methods for modelling and control of discrete-time unknown non-linear dynamical systems, when only input-output data are available. We propose the adoption of lazy learning, a memory-based technique for local modelling. The modelling procedure uses a query-based approach to select the best model configuration by assessing and comparing different alternatives. A new recursive technique for local model identification and validation is presented, together with an enhanced statistical method for model selection. A lso, three methods to design controllers based on the local linearization provided by the lazy learning algorithm are described. In the first method the lazy technique returns the forward and inverse models of the system which are used to compute the control action to take. The second is an indirect method inspired by self-tuning regulators where recursive least squares estimation is replaced by a local approximator. The third method combines the linearization provided by t...
248 citations
TL;DR: A unified view of algorithms for adaptive transversal FIR filtering and system identification has been presented, and the LMS algorithm and its offspring have been presented and interpreted as stochastic approximations of iterative deterministic steepest descent optimization schemes.
Abstract: A unified view of algorithms for adaptive transversal FIR filtering and system identification has been presented. Wiener filtering and stochastic approximation are the origins from which all the algorithms have been derived, via a suitable choice of iterative optimization schemes and appropriate design parameters. Following this philosophy, the LMS algorithm and its offspring have been presented and interpreted as stochastic approximations of iterative deterministic steepest descent optimization schemes. On the other hand, the RLS and the quasi-RLS algorithms, like the quasi-Newton, the FNTN, and the affine projection algorithm, have been derived as stochastic approximations of iterative deterministic Newton and quasi-Newton methods. Fast implementations of these methods have been discussed. Block-adaptive, and block-exact adaptive filtering have also been considered. The performance of the adaptive algorithms has been demonstrated by computer simulations.
232 citations
01 Jan 1999
TL;DR: In this paper, an accurate, high-bandwidth, linear state-space model was derived for the hover condition of a fully-instrumented model-scale unmanned helicopter (Yamaha R-SO with loft. diameter rotor) for dynamic model identification.
Abstract: Abstmcf: Flight testing of a fully-instrumented model-scale unmanned helicopter (Yamaha R-SO with loft. diameter rotor) was conducted for the purpose of dynamic model identification. This paper describes the application of CIFER' system identification techniques, which have been developed for full size helicopters, to this aircraft. An accurate, high-bandwidth, linear state-space model was derived for the hover condition. The model structure includes the explicit representation of regressive rotor-flap dynamics, rigid-body fuselage dynamics, and the yaw damper. The R-50 codiguration and identified dynamics are compared with those of a dynamically scaled UH-1H. The identified model shows excellent predictive capability and is well suited for flight control design and simulation applications.
TL;DR: Genetic programming is introduced, which is an evolutionary computing method that provides a ‘transparent’ and structured system identification, to rainfall-runoff modelling and is applied to flow prediction for the Kirkton catchment in Scotland.
Abstract: Planning for sustainable development of water resources relies crucially on the data available. Continuous hydrologic simulation based on conceptual models has proved to be the appropriate tool for studying rainfall-runoff processes and for providing necessary data. In recent years, artificial neural networks have emerged as a novel identification technique for the modelling of hydrological processes. However, they represent their knowledge in terms of a weight matrix that is not accessible to human understanding at present. This paper introduces genetic programming, which is an evolutionary computing method that provides a ‘transparent’ and structured system identification, to rainfall-runoff modelling. The genetic-programming approach is applied to flow prediction for the Kirkton catchment in Scotland (U.K.). The results obtained are compared to those attained using two optimally calibrated conceptual models and an artificial neural network. Correlations identified using data-driven approaches (genetic programming and neural network) are surprising in their consistency considering the relative size of the models and the number of variables included. These results also compare favourably with the conceptual models.
TL;DR: In this paper, the Steiglitz-McBride algorithm, the eigensystem realization algorithm, and the Prony method were used to identify low order linear systems of power systems modeled in standard transient stability programs.
Abstract: This paper describes the results of a study to evaluate the performance of three identification methods for the study of low frequency electromechanical oscillations. The three identification methods considered are: the Steiglitz-McBride algorithm; the eigensystem realization algorithm; and the Prony method. The identification methods are used to identify low order linear systems of power systems modeled in standard transient stability programs. This is accomplished by processing the system response to a simple probing pulse. The frequency domain characteristics of several identified systems are compared using three power systems with lightly damped electromechanical modes.
TL;DR: In this paper, a method based on adaptive estimation approaches is presented for the on-line identification of hysteretic systems under arbitrary dynamic environments, where no information is available on the system parameters, even the mass distribution.
Abstract: A method based on adaptive estimation approaches is presented for the on-line identification of hysteretic systems under arbitrary dynamic environments. The availability of such an identification approach is crucial for the on-line control and monitoring of time-varying structural systems. Previous work by the writers is extended to handle the general case when no information is available on the system parameters, even the mass distribution. A robust, least-squares-based adaptive identification algorithm, incorporating a Bouc-Wen hysteresis element model with additional polynomial-type nonlinear terms, is used to investigate the effects of persistence of excitation and of under- and overparameterization: challenging problems in realistic applications. In spite of the challenges encountered in the identification of the hereditary nature of the restoring force of such nonlinear systems, it is shown through the use of simulation studies of single-degree-of-freedom and certain multi-degree-of-freedom systems ...
TL;DR: In this article, the identification of a linear parameter-varying syster whose parameter dependence can be written as a linear fractional transformation (LFT) is formulated as an output-error identification problem.
Abstract: This paper deals with the identification of a linear parameter-varying (LPV) syster whose parameter dependence can be written as a linear-fractional transformation (LFT). We formulate an output-error identification problem and present a parameter estimation scheme in which a prediction error-based cost function is minimized using nonlinear programming; its gradients and (approximate) Hessians can he computed using LPV filters and inne products, and identifiable model sets (i.e., local canonical forms) are obtained efficiently using a natural geometrical approach. Some computational issues and experiences are discussed, and a simple numerical example is provided fo r illustration.
TL;DR: In this article, the authors address the issue of system identification for linear structural systems using earthquake induced time histories of the structural response and propose a methodology based on the Eigensystem Realization Algorithm (ERA) and on the Observer/Kalman filter IDentification (OKID) approach to perform identification of structural system using general input-output data via Markov parameters.
Abstract: This paper addresses the issue of system identification for linear structural systems using earthquake induced time histories of the structural response. The proposed methodology is based on the Eigensystem Realization Algorithm (ERA) and on the Observer/Kalman filter IDentification (OKID) approach to perform identification of structural systems using general input-output data via Markov parameters. The efficiency of the proposed technique is shown by numerical examples for the case of eight-storey building finite element models subjected to earthquake excitation and by the analysis of the data from the dynamic response of the Vincent-Thomas cable suspension bridge (Long Beach, CA) recorded during the Whittier and the North-ridge earthquakes. The effects of noise in the measurements and of inadequate instrumentation are investigated. It is shown that the identified models show excellent agreement with the real systems in predicting the structural response time histories when subjected to earthquake-induced ground motion.
TL;DR: In this paper, an observer-based least square method and an observer/filtered-regressor-based method were proposed for identifying the tire-road friction coefficient, where the observer is used to estimate signals which are difficult or expensive to measure.
Abstract: This paper presents methods for identifying the tire-road friction coefficient. The proposed methods are: an observer-based least square method and an observer/filtered-regressor-based method. These methods were designed assuming that some of the states are not available since physical parameter identification methods developed assuming that the system states are available are not attractive from a practical point of view. The observer is used to estimate signals which are difficult or expensive to measure. Using the estimated states of the system and the filtered-regressor, the parameter estimates are obtained. The proposed methods are evaluated on an eight state nonlinear vehicle/transmission simulation model with a Bakker-Pacejka's formula tire model. Vehicle tests have been performed on dry and wet roads to verify the performance of the methods. It has been shown through simulations and vehicle tests how the RPM sensors can be used with observer based identification methods to estimate the tire-road f...
TL;DR: The authors generalize linear subspace Identification theory to an analog theory for the subspace identification of bilinear systems and shows that most of the properties of linear sub space identification theory can be extended to similar properties for bilinears systems.
Abstract: The authors generalize linear subspace identification theory to an analog theory for the subspace identification of bilinear systems. A major assumption they make is that the inputs of the system should be white and mutually independent. It is shown that in that case most of the properties of linear subspace identification theory can be extended to similar properties for bilinear systems. The link between the presented bilinear subspace method and Kalman filter theory is made. Finally, the practical relevance of the method is illustrated by making a direct comparison between linear and bilinear subspace identification methods when applied on data from a model of a distillation column.
01 Jan 1999
TL;DR: In this paper, the authors considered communication using chaotic systems from a control point of view and showed that parameter identification methods may be effective in building reconstruction mechanisms, even when a synchronizing system is not available.
Abstract: Communication using chaotic systems is considered from a control point of view It is shown that parameter identification methods may be effective in building reconstruction mechanisms, even when a synchronizing system is not available. Three worked examples show the potentials of the proposed method.
TL;DR: Results indicate that the proposed method can successfully be used for the estimation of the useful energy extracted from the system and the temperature rise in the stored water of solar domestic water heating (SDHW) systems with the minimum of input data.
TL;DR: In this article, some remarkable analogies between subspace identification and model predictive control are uncovered, which can be combined in a very elegant way to form a numerically robust and easily implementable control/identification algorithm.
TL;DR: Results indicate that the approach described here will allow the estimation of endpoint stiffness dynamics in an experimentally efficient manner with minimal assumptions about the specific form of these properties.
Abstract: This study presents time-domain and frequency-domain, multiple-input, multiple-output (MIMO) linear system identification techniques that can be used to estimate the dynamic endpoint stiffness of a multijoint limb. The stiffness of a joint or limb arises from a number of physiological mechanisms and is thought to play a fundamental role in the control of posture and movement. Estimates of endpoint stiffness can therefore be used to characterize its modulation during physiological tasks and may provide insight into how the nervous system normally controls motor behavior. Previous MIMO stiffness estimates have focused upon the static stiffness components only or assumed simple parametric models with elastic, viscous, and inertial components. The method presented here captures the full stiffness dynamics during a relatively short experimental trial while assuming only that the system is linear for small perturbations. Simulation studies were performed to investigate the performance of this approach under typical experimental conditions. It was found that a linear MIMO description of endpoint stiffness dynamics was sufficient to describe the displacement responses to small stochastic force perturbations. Distortion of these linear estimates by nonlinear centripetal and Coriolis forces was virtually undetectable for these perturbations. The system identification techniques were also found to be robust in the presence of significant output measurement noise and input coupling. These results indicate that the approach described here will allow the estimation of endpoint stiffness dynamics in an experimentally efficient manner with minimal assumptions about the specific form of these properties.
TL;DR: This paper addresses a number of open problems concerning the generalized likelihood ratio rules for online detection of faults and parameter changes in control systems and shows that with an appropriate choice of the threshold and window size, these GLR rules are asymptotically optimal.
Abstract: This paper addresses a number of open problems concerning the generalized likelihood ratio (GLR) rules for online detection of faults and parameter changes in control systems It is shown that with an appropriate choice of the threshold and window size, these GLR rules are asymptotically optimal The rules are also extended to non-likelihood statistics that are widely used in monitoring adaptive algorithms for system identification and control by establishing Gaussian approximations to these statistics when the window size is chosen suitably Recursive algorithms are developed for practical implementation of the procedure, and importance sampling techniques are introduced for determining the threshold of the rule to satisfy prescribed bounds on the false alarm rate
01 Jan 1999
TL;DR: This contribution discusses the aspects of model validation in the light of model error models that are explicit descriptions of the model error, which allows a better visualization of the possible deficiencies of the nominal model.
Abstract: To validate an estimated model and to have a good understanding of its reliability is a central aspect of System Identification. This contribution discusses these aspects in the light of model error models that are explicit descriptions of the model error. A model error model is implicitly present in most model validation methods, so the concept is more of a representation form than a set of new techniques. Traditional model validation is essentially a test of whether the confidence region of the model error model contains the zero model. However, the model error model allows a better visualization of the possible deficiencies of the nominal model. Based on such information, the nominal model may very well be accepted even if the model error model does not contain the zero model. Conversely, it will be illustrated that the model error model may give good reason - because of if its more precise infomation - to reject a nominal model, that has passed a conventional model validation test.
TL;DR: In this article, an artificial neural network (ANN) was used to predict the performance of a thermosiphon solar domestic water heating system, which is measured in terms of the useful energy extracted and the stored water temperature rise.
TL;DR: Drawing on the rich theory of wavelets, a system identification scheme based on orthogonal wavelets is proposed and illustrated by applying the procedure to determine a speed-controller for an electric vehicle.
Abstract: Compactly supported orthogonal wavelets have certain properties that are useful for system identification and learning control. Drawing on the rich theory of wavelets, we propose a system identification scheme based on orthogonal wavelets. Better accuracy of estimation can be obtained by adding more terms to the wavelet based identifier, and these terms do not alter the coefficients of the existing terms. These terms can be selectively added depending upon the region of interest, for example we may require more terms in regions where the identified functions vary rapidly. We illustrate the concepts by applying the procedure to determine a speed-controller for an electric vehicle.
TL;DR: Wiener model predictive control (WMPC) is evaluated experimentally, and also compared with benchmark proportional integral derivative (PID) and linear MPC strategies, considering the effects of output constraints and modeling error.
Abstract: pH control is recognized as an industrially important, yet notoriously difficult control problem. Wiener models, consisting of a linear dynamic element followed in series by a static nonlinear element, are considered to be ideal for representing this and several other nonlinear processes. Wiener models require little more effort in development than a standard linear step-response model, yet offer superior characterization of systems with highly nonlinear gains. These models may be incorporated into model predictive control (MPC) schemes in a unique way which effectively removes the nonlinearity from the control problem, preserving many of the favorable properties of linear MPC. In this paper, Wiener model predictive control (WMPC) is evaluated experimentally, and also compared with benchmark proportional integral derivative (PID) and linear MPC strategies, considering the effects of output constraints and modeling error.
TL;DR: The separable least-squares technique of Golub and Pereyra can be exploited in the identification of both linear and non-linear systems based on the prediction error formulation to solve the associated optimisation problem in the linear case using only np parameters.
TL;DR: A new set of sensitivity functions, called the generalized sensitivity functions (GSF), are proposed, based on information theoretical criteria, for the analysis of input–output identification experiments and provide a more accurate picture on the information content of measured outputs on individual model parameters at different times.
Abstract: Parameters of physiological models are commonly associated in an input–output experiment with a specific pattern of the system response. This association is often made on an intuitive basis by traditional sensitivity analysis, i.e., by inspecting the variations of model output trajectories with respect to parameter variations. However, this approach provides limited information since, for instance, it ignores correlation among parameters. The aim of this study is to propose a new set of sensitivity functions, called the generalized sensitivity functions (GSF), for the analysis of input–output identification experiments. GSF are based on information theoretical criteria and provide, as compared to traditional sensitivity analysis, a more accurate picture on the information content of measured outputs on individual model parameters at different times. Case studies are presented on an input–output model and on two structural circulatory and respiratory models. GSF allow the definition of relevant time intervals for the identification of specific parameters and improve the understanding of the role played by specific model parameters in describing experimental data. © 1999 Biomedical Engineering Society.
PAC99: 8710+e, 8719Uv
TL;DR: In this article, a hybrid hysteresis model integrating the classical Preisach model and a neural network is proposed to identify and implement a piezoelectric actuator model for precision machining.