Showing papers on "System identification published in 2005"

Digital Control Systems: Design, Identification and Implementation

Abstract: Continuous Control Systems: A Review -- Computer Control Systems -- Robust Digital Controller Design Methods -- Design of Digital Controllers in the Presence of Random Disturbances -- System Identification: The Bases -- System Identification Methods -- Practical Aspects of System Identification -- Practical Aspects of Digital Control -- Identification in Closed Loop -- Reduction of Controller Complexity.

A new class of wavelet networks for nonlinear system identification

Abstract: A new class of wavelet networks (WNs) is proposed for nonlinear system identification. In the new networks, the model structure for a high-dimensional system is chosen to be a superimposition of a number of functions with fewer variables. By expanding each function using truncated wavelet decompositions, the multivariate nonlinear networks can be converted into linear-in-the-parameter regressions, which can be solved using least-squares type methods. An efficient model term selection approach based upon a forward orthogonal least squares (OLS) algorithm and the error reduction ratio (ERR) is applied to solve the linear-in-the-parameters problem in the present study. The main advantage of the new WN is that it exploits the attractive features of multiscale wavelet decompositions and the capability of traditional neural networks. By adopting the analysis of variance (ANOVA) expansion, WNs can now handle nonlinear identification problems in high dimensions.

A bounded-error approach to piecewise affine system identification

Abstract: This paper proposes a three-stage procedure for parametric identification of piecewise affine autoregressive exogenous (PWARX) models. The first stage simultaneously classifies the data points and estimates the number of submodels and the corresponding parameters by solving the partition into a minimum number of feasible subsystems (MIN PFS) problem for a suitable set of linear complementary inequalities derived from data. Second, a refinement procedure reduces misclassifications and improves parameter estimates. The third stage determines a polyhedral partition of the regressor set via two-class or multiclass linear separation techniques. As a main feature, the algorithm imposes that the identification error is bounded by a quantity /spl delta/. Such a bound is a useful tuning parameter to trade off between quality of fit and model complexity. The performance of the proposed PWA system identification procedure is demonstrated via numerical examples and on experimental data from an electronic component placement process in a pick-and-place machine.

Model identification for hydrological forecasting under uncertainty

Abstract: Methods for the identification of models for hydrological forecasting have to consider the specific nature of these models and the uncertainties present in the modeling process. Current approaches fail to fully incorporate these two aspects. In this paper we review the nature of hydrological models and the consequences of this nature for the task of model identification. We then continue to discuss the history (“The need for more POWER‘’), the current state (“Learning from other fields”) and the future (“Towards a general framework”) of model identification. The discussion closes with a list of desirable features for an identification framework under uncertainty and open research questions in need of answers before such a framework can be implemented.

Adaptive identification and control of hysteresis in smart materials

Abstract: Hysteresis hinders the effective use of smart materials in sensors and actuators. This paper addresses recursive identification and adaptive inverse control of hysteresis in smart material actuators, where hysteresis is modeled by a Preisach operator with a piecewise uniform density function. Two classes of identification schemes are proposed and compared, one based on the hysteresis output, the other based on the time-difference of the output. Conditions for parameter convergence are presented in terms of the input to the Preisach operator. An adaptive inverse control scheme is developed by updating the Preisach operator (and thus its inverse) with the output-based identification method. The asymptotic tracking property of this scheme is established, and for periodic reference trajectories, the parameter convergence behavior is characterized. Practical issues in the implementation of the adaptive identification and inverse control methods are also investigated. Simulation and experimental results based on a magnetostrictive actuator are provided to illustrate the proposed approach.

System identification of power converters with digital control through cross-correlation methods

Abstract: For digitally controlled switching power converters, on-line system identification can be used to assess the system dynamic responses and stability margins. This paper presents a modified correlation method for system identification of power converters with digital control. By injecting a multiperiod pseudo random binary signal (PRBS) to the control input of a power converter, the system frequency response can be derived by cross-correlation of the input signal and the sensed output signal. Compared to the conventional cross-correlation method, averaging the cross-correlation over multiple periods of the injected PRBS can significantly improve the identification results in the presence of PRBS-induced artifacts, switching and quantization noises. An experimental digitally controlled forward converter with an FPGA-based controller is used to demonstrate accurate and effective identification of the converter control-to-output response.

Subspace methods for system identification

Abstract: Introduction Part I: Preliminaries Linear Algebra and Preliminaries Disctrete-time Linear Systems Stochastic Processes Kalman Filter Part II: Realization Theory Realization of Deterministic Problems Stochastic Realization Theory I Stochastic Realization Theory II Part III: Subspace Identification Subspace Identification I: ORT Subspace Identification II: CCA Identification of Closed-loop System Appendices Least-squares Method Input Signals for System Identification Overlapping Parametrization Matlab(R) Programs Solutions to Problems

A fast nonlinear model identification method

Abstract: The identification of nonlinear dynamic systems using linear-in-the-parameters models is studied. A fast recursive algorithm (FRA) is proposed to select both the model structure and to estimate the model parameters. Unlike orthogonal least squares (OLS) method, FRA solves the least-squares problem recursively over the model order without requiring matrix decomposition. The computational complexity of both algorithms is analyzed, along with their numerical stability. The new method is shown to require much less computational effort and is also numerically more stable than OLS.

Dynamic systems identification with Gaussian processes

Abstract: This paper describes the identification of nonlinear dynamic systems with a Gaussian process (GP) prior model. This model is an example of the use of a probabilistic non-parametric modelling approach. GPs are flexible models capable of modelling complex nonlinear systems. Also, an attractive feature of this model is that the variance associated with the model response is readily obtained, and it can be used to highlight areas of the input space where prediction quality is poor, owing to the lack of data or complexity (high variance). We illustrate the GP modelling technique on a simulated example of a nonlinear system.

Input design via LMIs admitting frequency-wise model specifications in confidence regions

Abstract: A framework for reformulating input design problems in prediction error identification as convex optimization problems is presented For linear time-invariant single input/single output systems, this framework unifies and extends existing results on open-loop input design that are based on the finite dimensional asymptotic covariance matrix of the parameter estimates Basic methods for parametrizing the input spectrum are provided and conditions on these parametrizations that guarantee that all possible covariance matrices for the asymptotic distribution of the parameter estimates can be generated are provided A wide range of model quality constraints can be handled In particular, different frequency-by-frequency constraints can be used This opens up new applications of input design in areas such as robust control Furthermore, quality specifications can be imposed on all models in a confidence region Thus, allowing for statements such as "with at least 99% probability the model quality specifications will be satisfied"

Identification of Nonlinear Aeroelastic Systems Based on the Volterra Theory: Progress and Opportunities

Abstract: The identification of nonlinear aeroelastic systems based on the Volterra theory of nonlinear systems is presented. Recent applications of the theory to problems in computational and experimental aeroelasticity are reviewed. Computational results include the development of computationally efficient reduced-order models (ROMs) using an Euler/Navier–Stokes flow solver and the analytical derivation of Volterra kernels for a nonlinear aeroelastic system. Experimental results include the identification of aerodynamic impulse responses, the application of higher-order spectra (HOS) to wind-tunnel flutter data, and the identification of nonlinear aeroelastic phenomena from flight flutter test data of the active aeroelastic wing (AAW) aircraft.

Hierarchical identification of lifted state-space models for general dual-rate systems

Abstract: This paper is motivated by practical consideration that the input updating and output sampling rates are often limited due to sensor and actuator speed constraints. In particular, for general dual-rate systems with different updating and sampling periods, we derive the lifted state-space models (mapping relations between available dual-rate input-output data), and, by using a hierarchical identification principle, present combined parameter and state estimation algorithms for identifying the canonical lifted models based on the given dual-rate input-output data, taking into account the causality constraints of the lifted systems. Finally, we give an illustrative example to indicate that the proposed algorithm is effective.

Dynamic Wavelet Neural Network for Nonlinear Identification of Highrise Buildings

Abstract: : Recently, the authors presented a multiparadigm dynamic time-delay fuzzy wavelet neural network (WNN) model for nonparametric identification of structures using the nonlinear autoregressive moving average with exogenous inputs. Compared with conventional neural networks, training of a dynamic neural network for system identification of large-scale structures is substantially more complicated and time consuming because both input and output of the network are not single valued but involve thousands of time steps. In this article, an adaptive Levenberg–Marquardt least-squares algorithm with a backtracking inexact linear search scheme is presented for training of the dynamic fuzzy WNN model. The approach avoids the second-order differentiation required in the Gauss–Newton algorithm and overcomes the numerical instabilities encountered in the steepest descent algorithm with improved learning convergence rate and high computational efficiency. The model is applied to two highrise moment-resisting building structures, taking into account their geometric nonlinearities. Validation results demonstrate that the new methodology provides an efficient and accurate tool for nonlinear system identification of high-rising buildings.

Adaptive feedback cancellation in hearing aids with linear prediction of the desired signal

Abstract: The standard continuous adaptation feedback cancellation algorithm for feedback suppression in hearing aids suffers from a large model error or bias if the received sound signal is spectrally colored. To reduce the bias in the feedback path estimate, we propose adaptive feedback cancellation techniques that are based on a closed-loop identification of the feedback path as well as the (auto-regressive) modeling of the desired signal. In general, both models are not simultaneously identifiable in the closed-loop system at hand. We show that-under certain conditions, e.g., if a delay is inserted in the forward path-identification of both models is indeed possible. Two classes of adaptive procedures for identifying the desired signal model and the feedback path are derived: a two-channel identification method as well as a prediction error method. In contrast to the two-channel identification method, the prediction error method allows use of different adaptation schemes for the feedback path and for the desired signal model and, hence, is found to be preferable for highly nonstationary sound signals. Simulation results demonstrate that the proposed techniques outperform the standard continuous adaptation algorithm if the conditions for identifiability are satisfied.

Subspace identification of Hammerstein systems using least squares support vector machines

Abstract: This paper presents a method for the identification of multiple-input-multiple-output (MIMO) Hammerstein systems for the goal of prediction. The method extends the numerical algorithms for subspace state space system identification (N4SID), mainly by rewriting the oblique projection in the N4SID algorithm as a set of componentwise least squares support vector machines (LS-SVMs) regression problems. The linear model and static nonlinearities follow from a low-rank approximation of a matrix obtained from this regression problem.

Ambient Vibration Monitoring

Abstract: PREFACE. ACKNOWLEDGEMENTS. SUMMARY. 1 INTRODUCTION. 1.1 Scope of Applications. 1.2 Laws and Regulations. 1.3 Theories on the Development of the AVM. 2 OBJECTIVES OF APPLICATIONS. 2.1 System Identification. 2.1.1 Eigenfrequencies and Mode Shapes. 2.1.2 Damping. 2.1.3 Deformations and Displacements. 2.1.4 Vibration Intensity. 2.1.5 Trend Cards. 2.2 Stress Test. 2.2.1 Determination of Static and Dynamic Stresses. 2.2.2 Determination of the Vibration Elements. 2.2.3 Stress of Individual Structural Members. 2.2.4 Determination of Forces in Tendons and Cables. 2.3 Assessment of Stresses. 2.3.1 Structural Safety. 2.3.2 Structural Member Safety. 2.3.3 Maintenance Requirements and Intervals. 2.3.4 Remaining Operational Lifetime. 2.4 Load Observation (Determination of External Influences). 2.4.1 Load Collective. 2.4.2 Stress Characteristic. 2.4.3 Verification of Load Models. 2.4.4 Determination of Environmental Influences. 2.4.5 Determination of Specific Measures. 2.4.6 Check on the Success of Rehabilitation Measures. 2.4.7 Dynamic Effects on Cables and Tendons. 2.4.8 Parametric Excitation. 2.5 Monitoring of the Condition of Structures. 2.5.1 Assessment of Individual Objects. 2.5.2 Periodic Monitoring. 2.5.3 BRIMOS- Recorder. 2.5.4 Permanent Monitoring. 2.5.5 Subsequent Measures. 2.6 Application of Ambient Vibration Testing to Structures for Railways. 2.6.1 Sleepers. 2.6.2 Noise and Vibration Problems. 2.7 Limitations. 2.7.1 Limits of Measuring Technology. 2.7.2 Limits of Application. 2.7.3 Limits of Analysis. 2.7.4 Perspectives. References. 3 FEEDBACK FROM MONITORING TO BRIDGE DESIGN. 3.1 Economic Background. 3.2 Lessons Learned. 3.2.1 Conservative Design. 3.2.2 External versus Internal Pre-stressing. 3.2.3 Influence of Temperature. 3.2.4 Displacement. 3.2.5 Large Bridges versus Small Bridges. 3.2.6 Vibration Intensities. 3.2.7 Damping Values of New Composite Bridges. 3.2.8 Value of Patterns. 3.2.9 Understanding of Behaviour. 3.2.10 Dynamic Factors. References. 4 PRACTICAL MEASURING METHODS. 4.1 Execution of Measuring. 4.1.1 Test Planning. 4.1.2 Levelling of the Sensors. 4.1.3 Measuring the Structure. 4.2 Dynamic Analysis. 4.2.1 Calculation Models. 4.2.2 State of the Art. 4.3 Measuring System. 4.3.1 BRIMOS(r). 4.3.2 Sensors. 4.3.3 Data-Logger. 4.3.4 Additional Measuring Devices and Methods. 4.4 Environmental Influence. 4.5 Calibration and Reliability. 4.6 Remaining Operational Lifetime. 4.6.1 Rainflow Algorithm. 4.6.2 Calculation of Stresses by FEM. 4.6.3 S-N Approach and Damage Accumulation. 4.6.4 Remaining Service Lifetime by Means of Existing Traffic Data and Additional Forward and Backward Extrapolation. 4.6.5 Conclusions and Future Work. References. 5 PRACTICAL EVALUATION METHODS. 5.1 Plausibility of Raw Data. 5.2 AVM Analysis. 5.2.1 Recording. 5.2.2 Data Reduction. 5.2.3 Data Selection. 5.2.4 Frequency Analysis, ANPSD (Averaged Normalized Power Spectral Density). 5.2.5 Mode Shapes. 5.2.6 Damping. 5.2.7 Deformations. 5.2.8 Vibration Coefficients. 5.2.9 Counting of Events. 5.3 Stochastic Subspace Identification Method. 5.3.1 The Stochastic Subspace Identification (SSI) Method. 5.3.2 Application to Bridge Z24. 5.4 Use of Modal Data in Structural Health Monitoring. 5.4.1 Finite Element Model Updating Method. 5.4.2 Application to Bridge Z24. 5.4.3 Conclusions. 5.5 External Tendons and Stay Cables. 5.5.1 General Information. 5.5.2 Theoretical Bases. 5.5.3 Practical Implementation. 5.5.4 State of the Art. 5.5.5 Rain-Wind Induced Vibrations of Stay Cables. 5.5.6 Assessment. 5.6 Damage Identification and Localization. 5.6.1 Motivation for SHM. 5.6.2 Current Practice. 5.6.3 Condition and Damage Indices. 5.6.4 Basic Philosophy of SHM. 5.7 Damage Prognosis. 5.7.1 Sensing Developments. 5.7.2 Data Interrogation Procedure for Damage Prognosis. 5.7.3 Predictive Modelling of Damage Evolution. 5.8 Animation and the Modal Assurance Criterion (MAC). 5.8.1 Representation of the Calculated Mode Shapes. 5.8.2 General Requirements. 5.8.3 Correlation of Measurement and Calculation (MAC). 5.8.4 Varying Number of Eigenvectors. 5.8.5 Complex Eigenvector Measurement. 5.8.6 Selection of Suitable Check Points using the MAC. 5.9 Ambient Vibration Derivatives (AVD(r)). 5.9.1 Aerodynamic Derivatives. 5.9.2 Applications of the AVM. 5.9.3 Practical Implementation. References. 6 THEORETICAL BASES. 6.1 General Survey on the Dynamic Calculation Method. 6.2 Short Description of Analytical Modal Analysis. 6.3 Equation of Motion of Linear Structures. 6.3.1 SDOF System. 6.3.2 MDOF System. 6.3.3 Influence of Damping. 6.4 Dynamic Calculation Method for the AVM. 6.5 Practical Evaluation of Measurements. 6.5.1 Eigenfrequencies. 6.5.2 Mode Shapes. 6.5.3 Damping. 6.6 Theory on Cable Force Determination. 6.6.1 Frequencies of Cables as a Function of the Inherent Tensile Force. 6.6.2 Influence of the Bending Stiffness. 6.6.3 Influence of the Support Conditions. 6.6.4 Comparison of the Defined Cases with Experimental Results. 6.6.5 Measurement Data Adjustment for Exact Cable Force Determination. 6.7 Transfer Functions Analysis. 6.7.1 Mathematical Backgrounds. 6.7.2 Transfer Functions in the Vibration Analysis. 6.7.3 Applications (Examples). 6.8 Stochastic Subspace Identification. 6.8.1 Stochastic State-Space Models. 6.8.2 Stochastic System Identification. References. 7 OUTLOOK. 7.1 Decision Support Systems. 7.2 Sensor Technology and Sensor Networks. 7.2.1 State-of-the-Art Sensor Technology. 7.3 Research Gaps and Opportunities. 7.4 International Collaboration. 7.4.1 Collaboration Framework. 7.4.2 Activities. 8 EXAMPLES FOR APPLICATION. 8.1 Aitertal Bridge, Post-tensional T-beam (1956). 8.2 Donaustadt Bridge, Cable-Stayed Bridge in Steel (1996). 8.3 F9 Viaduct Donnergraben, Continuous Box Girder (1979). 8.4 Europa Bridge, Continuous Steel Box Girder (1961). 8.5 Gasthofalm Bridge, Composite Bridge (1979). 8.6 Kao Ping Hsi Bridge, Cable-Stayed Bridge (2000). 8.7 Inn Bridge Roppen, Concrete Bridge (1936). 8.8 Slope Bridge Saag, Bridge Rehabilitation (1998). 8.9 Flyover St Marx, Permanent Monitoring. 8.10 Mur Bridge in St Michael, Bridge Rehabilitation. 8.11 Rosen Bridge in Tulln, Concrete Cable-Stayed Bridge (1995). 8.12 VOEST Bridge, Steel Cable-Stayed Bridge (1966). 8.13 Taichung Bridge, Cable-Stayed Bridge. APPENDIX. Nomenclature. INDEX.

Iterative approach to model identification of biological networks

Abstract: Recent advances in molecular biology techniques provide an opportunity for developing detailed mathematical models of biological processes. An iterative scheme is introduced for model identification using available system knowledge and experimental measurements. The scheme includes a state regulator algorithm that provides estimates of all system unknowns (concentrations of the system components and the reaction rates of their inter-conversion). The full system information is used for estimation of the model parameters. An optimal experiment design using the parameter identifiability and D-optimality criteria is formulated to provide "rich" experimental data for maximizing the accuracy of the parameter estimates in subsequent iterations. The importance of model identifiability tests for optimal measurement selection is also considered. The iterative scheme is tested on a model for the caspase function in apoptosis where it is demonstrated that model accuracy improves with each iteration. Optimal experiment design was determined to be critical for model identification. The proposed algorithm has general application to modeling a wide range of cellular processes, which include gene regulation networks, signal transduction and metabolic networks.

Spectral-approximation-based intelligent modeling for distributed thermal processes

Abstract: A spectral-approximation-based intelligent modeling approach is proposed for the distributed thermal processing of the snap curing oven that is used in semiconductor packaging industry. The snap curing oven can be described by a nonlinear parabolic distributed parameter system (DPS) in the time-space domain. After finding a proper approximation of the complex boundary conditions of the system, the spectral methods can be applied to time-space separation and model reduction, and neural networks (NNs) can be used for state estimation and system identification. With the help of model reduction techniques, the dynamics of the curing process derived from physical laws can be described by a model of low-order nonlinear ordinary differential equations with a few uncertain parameters and unknown nonlinearities. A neural observer can then be designed to estimate the states of the ordinary differential equation model from measurements taken at specified locations in the field. Using the estimated states, a hybrid general regression NN is trained to be a nonlinear model of the curing process in state-space formulation, which is suitable for the further application of traditional control techniques. Real-time experiments on the snap curing oven show that the proposed modeling method is effective. This modeling methodology can be applied to a class of nonlinear DPSs in industrial thermal processing.

Identification of Parametric Variations of Structures Based on Least Squares Estimation and Adaptive Tracking Technique

Abstract: An important objective of health monitoring systems for civil infrastructures is to identify the state of the structure and to detect the damage when it occurs. System identification and damage detection, based on measured vibration data, have received considerable attention recently. Frequently, the damage of a structure may be reflected by a change of some parameters in structural elements, such as a degradation of the stiffness. Hence it is important to develop data analysis techniques that are capable of detecting the parametric changes of structural elements during a severe event, such as the earthquake. In this paper, we propose a new adaptive tracking technique, based on the least-squares estimation approach, to identify the time-varying structural parameters. In particular, the new technique proposed is capable of tracking the abrupt changes of system parameters from which the event and the severity of the structural damage may be detected. The proposed technique is applied to linear structures, including the Phase I ASCE structural health monitoring benchmark building, and a nonlinear elastic structure to demonstrate its performance and advantages. Simulation results demonstrate that the proposed technique is capable of tracking the parametric change of structures due to damages.