Showing papers on "System identification published in 1998"

Fault detection and diagnosis in engineering systems

Abstract: Introduction to fault detection and diagnosis discrete linear systems random variables parameter estimation fundamentals analytical redundancy concepts parity equation implementation of residual generators design for structured residuals design for directional residuals residual generation for parametric faults robustness in residual generation statistical testing of residuals model identification for the diagnosis of additive faults diagnosing multiplicative faults by parameter estimation

Multichannel blind identification: from subspace to maximum likelihood methods

Abstract: A review of blind channel estimation algorithms is presented. From the (second-order) moment-based methods to the maximum likelihood approaches, under both statistical and deterministic signal models. We outline basic ideas behind several new developments, the assumptions and identifiability conditions required by these approaches, and the algorithm characteristics and their performance. This review serves as an introductory reference for this currently active research area.

Parametric and nonparametric identification of linear systems in the presence of nonlinear distortions-a frequency domain approach

Abstract: This paper studies the asymptotic behavior of nonparametric and parametric frequency domain identification methods to model linear dynamic systems in the presence of nonlinear distortions under some general conditions for random multisine excitations. In the first part, a related linear dynamic system (RLDS) approximation to the nonlinear system (NLS) is defined, and it is shown that the differences between the NLS and the RLDS can be modeled as stochastic variables with known properties. In the second part a parametric model for the RLDS is identified. Convergence in probability of this model to the RLDS is proven. A function of dependency is defined to detect and separate the presence of unmodeled dynamics and nonlinear distortions and to bound the bias error on the transfer function estimate.

Learning Nonlinear Dynamical Systems Using an EM Algorithm

Abstract: The Expectation-Maximization (EM) algorithm is an iterative procedure for maximum likelihood parameter estimation from data sets with missing or hidden variables [2]. It has been applied to system identification in linear stochastic state-space models, where the state variables are hidden from the observer and both the state and the parameters of the model have to be estimated simultaneously [9]. We present a generalization of the EM algorithm for parameter estimation in nonlinear dynamical systems. The "expectation" step makes use of Extended Kalman Smoothing to estimate the state, while the "maximization" step re-estimates the parameters using these uncertain state estimates. In general, the nonlinear maximization step is difficult because it requires integrating out the uncertainty in the states. However, if Gaussian radial basis function (RBF) approximators are used to model the nonlinearities, the integrals become tractable and the maximization step can be solved via systems of linear equations.

Identification of RC networks by deconvolution: chances and limits

Abstract: This paper deals with the identification of RC networks from their time- or frequency-domain responses. A new method is presented based on a recent approach of the network description where all response functions are calculated by convolution integrals. The identification is carried out by deconvolution (NID method). This paper discusses the practical details of the method. Special attention is paid to the identification and modeling of distributed RC networks, like the problems of conductive heat-flow. A number of examples make it easy to understand the operation and the capabilities of the NID method. Comparative considerations are given concerning the accuracy and expenses of the NID and the popular AWE (momentum-matching) methods.

Comparative study of damage identification algorithms applied to a bridge: II. Numerical study

Abstract: This paper extends the study of damage identification algorithms summarized in the accompanying paper `Comparative study of damage identification algorithms: I. Experiment' to numerical examples. A finite element model of a continuous three-span portion of the I-40 bridges, which once crossed the Rio Grande in Albuquerque, NM, was constructed. Dynamic properties (resonant frequencies and mode shapes) of the undamaged and damaged bridge that were predicted by the numerical models were then correlated with experimental modal analysis results. Once correlated with the experimental results, eight new damage scenarios were introduced into the numerical model including a multiple damage case. Also, results from two undamaged cases were used to study the possibility that the damage identification methods would produce false-positive readings. In all cases analytical modal parameters were extracted from time-history analyses using signal processing techniques similar to those used in the experimental investigation. This study provides further comparisons of the relative accuracy of these different damage identification methods when they are applied to a set of standard numerical problems.

Neural predictive controller for insulin delivery using the subcutaneous route

Abstract: A neural predictive controller for closed-loop control of glucose using subcutaneous (s.c.) tissue glucose measurement and s.c. infusion of monomeric insulin analogs was developed and evaluated in a simulation study. The proposed control strategy is based on off-line system identification using neural networks (NNs) and nonlinear model predictive controller design. The system identification framework combines the concept of nonlinear autoregressive model with exogenous inputs (NARX) system representation, regularization approach for constructing radial basis function NNs, and validation methods for nonlinear systems. Numerical studies on system identification and closed-loop control of glucose were carried out using a comprehensive model of glucose regulation and a pharmacokinetic model for the absorption of monomeric insulin analogs from the s.c. depot. The system identification procedure enabled construction of a parsimonious network from the simulated data, and consequently, design of a controller using multiple-step-ahead predictions of the previously identified model. According to the simulation results, stable control is achievable in the presence of large noise levels, for unknown or variable time delays as well as for slow time variations of the controlled process. However, the control limitations due to the s.c. insulin administration makes additional action from the patient at meal time necessary.

Robust nonlinear system identification using neural-network models

Abstract: We study the problem of identification for nonlinear systems in the presence of unknown driving noise, using both feedforward multilayer neural network and radial basis function network models. Our objective is to resolve the difficulty associated with the persistency of excitation condition inherent to the standard schemes in the neural identification literature. This difficulty is circumvented here by a novel formulation and by using a new class of identification algorithms recently obtained by Didinsky et al. (1995). We present a class of identifiers which secure a good approximant for the system nonlinearity provided that some global optimization technique is used. Subsequently, we address the same problem under a third, worst case L/sup /spl infin// criterion for an RBF modeling. We present a neural-network version of an H/sup /spl infin//-based identification algorithm from Didinsky et al., and show how it leads to satisfaction of a relevant persistency of excitation condition, and thereby to robust identification of the nonlinearity.

Local dynamic modeling with self-organizing maps and applications to nonlinear system identification and control

Abstract: The technique of local linear models is appealing for modeling complex time series due to the weak assumptions required and its intrinsic simplicity. Here, instead of deriving the local models from the data, we propose to estimate them directly from the weights of a self-organizing map (SOM), which functions as a dynamic preserving model of the dynamics. We introduce one modification to the Kohonen learning to ensure good representation of the dynamics and use weighted least squares to ensure continuity among the local models. The proposed scheme is tested using synthetic chaotic time series and real-world data. The practicality of the method is illustrated in the identification and control of the NASA Langley wind tunnel during aerodynamic tests of model aircraft. Modeling the dynamics with an SOM lends to a predictive multiple model control strategy. Comparison of the new controller against the existing controller in test runs shows the superiority of our method.

Preisach model identification of a two-wire SMA actuator

Abstract: In recent years, the Preisach hysteresis model (1935) has emerged as the model of choice for the behaviour of many smart materials, such as shape memory alloys (SMA). This research treats the identification of Preisach models for a differential SMA actuator. The traditional identification technique is applied, and several models are derived for the actuator. It is seen that the classical Preisach model is able to model its behaviour. However, the results raise questions concerning the robustness of the traditional identification technique, as well as the conservative nature of current passivity-based control results for the Preisach model.

Parameter identification of induction motor model using genetic algorithms

Abstract: The paper deals with methods of identification of the parameters of an induction motor model using genetic algorithms. It is supposed that the inverter supplying the motor is directly accessible for control of the conduction sequences of its power switches. This makes it possible to carry out a test consisting of a transient from standstill to steady-state operation at a given frequency and successive free motion to standstill. During this test, data are acquired referring to stator voltages, and currents and speed. Then, a genetic algorithm is employed with the aim of determining the mechanical and electrical parameters of the model, so as to reproduce the input-output behaviour of a real open-loop system.

A signal processing framework based on dynamic neural networks with application to problems in adaptation, filtering, and classification

Abstract: We present a coherent neural net based framework for solving various signal processing problems. It relies on the assertion that time-lagged recurrent networks possess the necessary representational capabilities to act as universal approximators of nonlinear dynamical systems. This applies to system identification, time-series prediction, nonlinear filtering, adaptive filtering, and temporal pattern classification. We address the development of models of nonlinear dynamical systems, in the form of time-lagged recurrent neural nets, which can be used without further training. We employ a weight update procedure based on the extended Kalman filter (EKF). Against the tendency for a net to forget earlier learning as it processes new examples, we develop a technique called multistream training. We demonstrate our framework by applying it to 4 problems. First, we show that a single time-lagged recurrent net can be trained to produce excellent one-time-step predictions for two different time series and also to be robust to severe errors in the input sequence. Second, we model stably a complex system containing significant process noise. The remaining two problems are drawn from real-world automotive applications. One involves input-output modeling of the dynamic behavior of a catalyst-sensor system which is exposed to an operating engine's exhaust stream, the other the real-time and continuous detection of engine misfire.

Adaptive H^∞ filter: Its application to structural identification

Abstract: By adding the function of memory fading for past observation data to the \iH\d∞ filter, the adaptive \iH\d∞ filter was developed for identifying structural systems with nonstationary dynamic characteristics. Identification algorithms are proposed for time-varying structural systems in which the velocity and displacement of each floor are available for observation, as well as for the case when only the velocity and displacement of some floors are available. The Akaike-Bayes information criterion is used to determine the optimal forgetting factor. Identification algorithms that use the adaptive \iH\d∞ and Kalman filters are applied to a five-degree of freedom (DOF) linear system with nonstationary dynamic characteristics and to a five-DOF nonlinear structural system. Digital simulation results show that the adaptive \iH\d∞ filter efficiently traces the time-varying properties of structural systems. The behavior of the adaptive \iH\d∞ filter is better than that of the adaptive Kalman filter for identifying a structural system with time-varying dynamic characteristics. The former is more efficient and robust for identifying structural systems with nonstationary dynamic characteristics.

Modeling and Parameter Identification of Harmonic Drive Systems

Abstract: The unique performance features of harmonic drives, such as high gear ratios and high torque capacities in a compact geometry, justify their widespreaa industrial application. However, harmonic drive can exhibit surprisingly more complex dynamic behavior than conventional gear transmission. In this paper a systematic way to capture and rationalize the dynamic behavior of the harmonic drive systems is developed. Simple and accurate models for compliance, hysteresis, and friction are proposed, and the model parameters are estimated using least-squares approximation for linear and nonlinear regression models. A statistical measure of variation is defined, by which the reliability of the estimated parameter for different operating condition, as well as the accuracy and integrity of the proposed model is quantified. By these means, it is shown that a linear stiffness model best captures the behavior of the system when combined with a good model for hysteresis. Moreover, the frictional losses of harmonic drive are modeled at both low and high velocities, The model performance is assessed by comparing simulations with the experimental results on two different harmonic drives. Finally, the significance of individual components of the nonlinear model is assessed by a parameter sensitivity study using simulations.

Runge-Kutta neural network for identification of dynamical systems in high accuracy

Abstract: This paper proposes Runge-Kutta neural networks (RKNNs) for identification of unknown dynamical systems described by ordinary differential equations (i.e., ordinary differential equation or ODE systems) with high accuracy. These networks are constructed according to the Runge-Kutta approximation method. The main attraction of the RKNNs is that they precisely estimate the changing rates of system states (i.e., the right-hand side of the ODE x/spl dot/=f(x)) directly in their subnetworks based on the space-domain interpolation within one sampling interval such that they can do long-term prediction of system state trajectories. We show theoretically the superior generalization and long-term prediction capability of the RKNNs over the normal neural networks. Two types of learning algorithms are investigated for the RKNNs, gradient-and nonlinear recursive least-squares-based algorithms. Convergence analysis of the learning algorithms is done theoretically. Computer simulations demonstrate the proved properties of the RKNNs.

Using neural networks in the identification of Preisach-type hysteresis models

Abstract: The identification process of the classical Preisach-type hysteresis model reduces to the determination of the weight function of elementary hysteresis operators upon which the model is built. It is well known that the classical Preisach model can exactly represent hysteretic nonlinearities which exhibit wiping-out and congruency properties. In that case, the model identification can be analytically and systematically accomplished by using first-order reversal curves. If the congruency property is not exactly valid, the Preisach model can only be used as an approximation. It is possible to improve the model accuracy in this situation by incorporating more appropriate experimental data during the identification stage. However, performing this process using the traditional systematic techniques becomes almost impossible. In this paper, the machinery of neural networks is proposed as a tool to accomplish this identification task. The suggested identification approach has been numerically implemented and carried out for a magnetic tape sample that does not possess the congruency property. A comparison between measured data and model predictions suggests that the proposed identification approach yields more accurate results.

Application of functional link neural network to HVAC thermal dynamic system identification

Abstract: Recent efforts to incorporate aspects of artificial intelligence into the design and operation of automatic control systems have focused attention on techniques such as fuzzy logic, artificial neural networks and expert systems. The use of computers for direct digital control highlights the recent trend toward more effective and efficient heating, ventilating and air-conditioning (HVAC) control methodologies. Researchers in the HVAC field have stressed the importance of self-learning in building control systems and have encouraged further studies in the integration of optimal control and other advanced techniques into the formulation of such systems. Artificial neural networks can also be used to emulate the plant dynamics, in order to estimate future plant outputs and obtain plant input/output sensitivity information for online neural control adaptation. This paper describes a functional link neural network approach to performing the HVAC thermal dynamic system identification. Methodologies to reduce inputs of the functional link network to reduce the complexity and speed up the training speed are presented. Analysis and comparison between the functional link network approach and the conventional network approach for the HVAC thermal modeling are also presented.

An adaptive run-to-run optimizing controller for linear and nonlinear semiconductor processes

Abstract: This paper presents a new run-to-run (R2R) multiple-input-multiple-output controller for semiconductor manufacturing processes. The controller, termed optimizing adaptive quality controller (OAQC), can act both as an optimizer-in case equipment models are not available-or as a controller for given models. The main components of the OAQC are shown and a study of its performance is presented. The controller allows one to specify input and output constraints and weights, and input resolutions. A multivariate control chart can be applied either as a deadband on the controller or simply to provide out of control alarms. Experimental designs can be utilized for on-line (recursive) model identification in the optimization phase. For testing purposes, two chemical mechanical planarization processes were simulated based on real equipment models. It is shown that the OAQC allows one to keep adequate control even if the input-output transfer function is severely nonlinear. Software implementation including the integration of the OAQC with the University of Michigan's Generic Cell Controller (GCC) is briefly discussed.

Optimal instrumentation of uncertain structural systems subject to earthquake ground motions

Abstract: A criterion is proposed for making decisions regarding the optimal location of a given number of sensors to record the seismic response of a structure for identification purposes. The optimal location of the sensors is selected so that the expected value of a Bayesian loss function, expressed in terms of the Fisher information in the recordings, is minimized. The criterion is applied to the case of multi-degree-of-freedom systems with uncertain structural properties subjected to earthquake ground motions modelled as stationary stochastic processes. The use and capabilities of the criterion are thoroughly illustrated by means of an example. Results are used to assess the influence of record duration, recording noise, and ground motion frequency content and amplitude, on the optimal location of accelerometers as well as on the reduction of prior uncertainty about the structural parameters. © 1998 John Wiley & Sons, Ltd.

Model-based control of strip temperature for the heating furnace in continuous annealing

Abstract: Presents the model of the heating furnace in continuous annealing processes for use in design of self-tuning control systems. A simplified mathematical model is derived from first principles. The model parameters are recursively estimated with an algorithm called recursive parameter estimation with a vector-type variable forgetting factor (REVVF). The REVVF algorithm was developed for such cases where some knowledge on parameter variability can be obtained beforehand. The control system of strip temperature presented here is hierarchical. The upper level is called "optimal preview control", which performs preset control. It previews the approaching setup change, which is the change of strip size or reference temperature, and optimizes the line speed and the strip temperature trajectory. Next, the lower level is called "temperature tracking control", which performs closed-loop control using the above trajectory as the control target. At this level, the generalized pole-placement self-tuning control was first employed; and later, the generalized predictive self-tuning control was introduced. These control methods were applied with some practical modifications and with the above mentioned REVVF. The control has been working successfully in several real plants.

Neural network modeling and identification of nonlinear channels with memory: algorithms, applications, and analytic models

Abstract: This paper proposes a neural network (NN) approach for modeling nonlinear channels with memory. Two main examples are given: (1) modeling digital satellite channels and (2) modeling solid-state power amplifiers (SSPAs). NN models provide good generalization performance (in terms of output signal-to-error ratio). NN modeling of digital satellite channels allows the characterization of each channel component. Neural net models represent the SSPA as a system composed of a linear complex filter followed by a nonlinear memoryless neural net followed by a linear complex filter. If the new algorithms are to be used in real systems, it is important that the algorithm designer understands their learning behavior and performance capabilities. Some simplified neural net models are analyzed in support of the simulation results. The analysis provides some theoretical basis for the usefulness of NNs for modeling satellite channels and amplifiers. The analysis of the simplified adaptive models explains the simulation results qualitatively but not quantitatively. The analysis proceeds in several steps and involves several novel ideas to avoid solving the more difficult general nonlinear problem.

Support Vector Machines for System Identification

Abstract: Support vector machines (SVM) are used for system identification of both linear and nonlinear dynamic systems. Discrete time linear models are used to illustrate parameter estimation and nonlinear models demonstrate model structure identification. The VC-dimension of a trained SVM indicates the model accuracy without using separate validation data. We conclude that SVM have potential in the field of dynamic system identification, but that there are a number of significant issues to be addressed.