Search

State estimators for induction motors are generally designed based on standard simplified models that suppose the machine magnetic characteristic to be linear. Since this characteristic is actually nonlinear, especially for high power machines, the mentioned state estimators are likely not able to achieve the estimation accuracy they have been designed for. In this paper, a new state estimator is developed for AC machine based on a model that appropriately accounts for the saturation feature in the magnetic characteristic. The proposed estimator is a full-order nonlinear observer. The gains of the latter are tuned, using Lyapunov stability tools, so that the state estimation error asymptotically vanishes. Supremacy of the new observer over standard ones is illustratedby simulation using a 7.5 KW machine.

In this paper, we consider a class of discrete-time nonlinear systems for which the control design is based on singular perturbation and sliding-mode techniques. In addition, a discrete-time observer is designed to estimate the nonmeasurable states required in the control law. The proposed controller–observer scheme is applied to a small synchronous generator. Simulations are carried out to show the performance of the controller–observer scheme.

In this paper, we propose an observer-based controller for synchronous generators' models. The proposed observer and controller designs are simple and systematic, and are based on the construction of similarity transformations which are used in finding canonical forms for linear observable and controllable systems. Moreover, it is shown, via a simulation example dealing with a typical synchronous generator, that the proposed observer-based controller is robust under parameter uncertainties. A comparative study of a PID controller with respect to the proposed controller is also made in order to show the performance of the latter.

In this paper, we propose an observer-based position tracking control for a synchronous generator. The controller design is based on the sliding-mode and singular perturbation techniques. The observer design is based on a recent approach which does not require the solution of a differential equation to update its gain.

In this paper, we give a new observer design methodology for single-output uniformly observable systems. In fact, the proposed observer is an improvement of the high gain observer given by Gauthier et al. (1992). The improvement lies basically in the fact that part of the non-linearity of the system is incorporated in the observer design strategy. Consequently, unlike the observer given in Gauthier et al. (1992), the gain of the improved observer is function of the state estimate. A comparison of the proposed observer to that given by Gauthier et al. (1992), is made via a simulation example dealing with a flexible joint mechanism and it is shown that the improved observer is more robust with regard to measurement noise and present less transient oscillations. An extension of the design strategy is also presented for a more general class of non-linear systems.

A control-observer structure for a class of nonlinear singularly perturbed systems is presented. The structure is based on a two-time scale sliding-mode control and an output injection observer. Also, an stability analysis is carried out to study the stability properties of the resultant closed-loop system using the stability characteristics of the observer and the closed-loopsystem without observer.

An exponential observer is constructed for a multi-output observable Bilinear System Class (BSC) in the Diop-Fliess' Observability sense. A differential algebraic approach is proposed for the estimation of the state of a class of bilinear system. A result on Multi-output Translated Fliess' Generalized Observability Canonical Form (MTGOCF) is given.

International audience; In this paper, an adaptive observer is proposed to solve the problem of simultaneous parameter identification and state estimation for a class of cascade state affine systems. Sufficient conditions are given in order to guarantee the exponential convergence of the proposed observer. Furthermore, simulation results are given illustrating the performance of the proposed observer when it is applied in the synchronization and identification problem of Rossler's chaotic system.

International audience; In this paper, the problem of adaptive observer design for the class of state affine systems is discussed. The discussion is based on recent results on adaptive observer with exponential rate of convergence obtained for multi-input/multi-output linear time varying systems on the one hand, and the well-known Kalman-like design for state affine systems on the other hand. In particular the relationship between both designs is emphasized, showing how they can even be equivalent. The interest of such an adaptive observer for state affine systems is illustrated by the example of state and parameter estimation for the Lorenz chaotic system. The observer performances are illustrated via simulation.

A proposed adaptive observer for time-varying linear systems [Q. Zhang, 2002] is revisited on the basis of the well- known Kalman-like design for state affine systems [H. Hammouri et al., 1990], [G. Besancon et al., 1996]. This approach in particular allows to emphasize the possible arbitrary rate of convergence in the design. The corresponding observer is applied to a problem of state and parameter estimation for a synchronous machine connected to an infinite bus, and its performances are illustrated through simulation.

In this paper, we consider a class of nonlinear systems which are discretized via an Euler discretization procedure. A control design based on sliding-mode techniques is proposed. Furthermore, a discrete-time nonlinear observer is used. The proposed controller-observer scheme is applied to a synchronous generator connected to an infinite bus. Simulations are carried out to show the performance of the controller-observer scheme.

This paper is concerned with the design of a controller-observer scheme for the exponential stabilization of a class of singularly perturbed nonlinear systems based on sliding-modes techniques and nonlinear observers. Under the assumption that only the fast state is available for measurement, an observer design is presented in order to estimate the slow variables. Furthermore, a stability analysis of the closed-loop system is also made to provide sufficient conditions for the exponential stability of the full order closed-loop system when the slow state is estimated by means of the observer. The proposed scheme is applied to the model of a permanent magnet stepper motor in order to illustrate the methodology.

We propose a controller design for a class of nonlinear systems using passive and Hamiltonian design techniques. The result is applied to a synchronous generator in order to synthesize an excitation control which permits stabilisation of the generator at its equilibrium position. The main characteristic of this proposition is that neither the equilibrium point of the system must be known nor the state of the generator must be available for measurement.

The authors present an observer for a class of nonlinear systems which are affine in the state and composed of subsystems interconnected in series. This observer works for the class of regularly persistent inputs. They recall main assumptions and restate the results proved by H. Hamouri and J. De Leon Morales (1990). The authors' result is stated and proved. >

Some recent results about nonlinear regularly persistent observers applied to some class of nonlinear systems are presented. The authors propose an observer based on a Kalman-like observer, which is formed with a copy of the model of the system plus a corrective term K(t). This term is being computed by minimizing some criterion depending on the output error. Following this criterion, an observer is obtained which works for a class of admissible inputs. >

The authors consider mechanical systems for which a stabilizing control law from the minimization of a Lyapunov function has to be constructed. A stabilizing control law for an overhead crane is proposed. In particular, a high-gain observer is used to estimate the momentum of the mechanical process. Its gain is adaptive and depends on the estimation error. >

We present a digital implementation of an observer-based controller, using a simple forward Euler discretization scheme, for a rigid robot. The observer used here is the one given in Busawon, Saif and Farza (1997) with a slight change in the gain so as to incorporate part of the nonlinearity. So doing, the high gain property of the observer is lessened to some extent. A comparison of the convergence is presented when this modified observer and a purely high gain observer is utilized. It is shown that the transient oscillations when using a purely high gain observer is considerably diminished by using the modified observer.

In this paper, observer-based controllers are designed using the singular perturbations theory and differential algebra, for position and speed control of a single flexible joint robot manipulator. Both schemes are compared in simulation.

Techniques of differential algebra are used to derive an observer-based controller applied to a flexible manipulator. A dynamical linearizing feedback controller is derived on the basis of Fliess generalized observability canonical form (GOCF). Our controller actually achieves both stability and tracking by linearising the tracking error. The controller is then of error-driven type; the dynamics of the error are estimated by an observer in the closed-loop. An analysis of the stability of the closed-loop system is given. Simulation results are presented for illustrating the performance of this observed based controller scheme, when it is applied to a mathematical model of a robot with flexible joints.

In this paper, a controller-observer scheme for a class of nonlinear singularly perturbed systems is designed and applied to the model of a synchronous generator. The controller design is based on the sliding-mode technique and allows it to have state reference tracking. The observer design is based on a recent approach that uses state transformation.

In this paper, we present an improvement of the classical high gain observer given in Gauthier et al (1992) for single-output uniformly observable systems. The improvement lies basically in the fact that the proposed observer incorporates the information of the nonlinearity of the system in the design strategy. Consequently, unlike the classical observer, the gain of the improved observer is function of the state estimate. A comparison between the proposed observer and the classical high gain one is made via a simulation example dealing with a flexible joint mechanism. It is shown that the improved observer is more robust with regards to measurement noise and present less transients oscillations. An extension of the design is also presented for a more general class of nonlinear systems.

We deal with the observer and controller design for Euler discretized form of nonlinear systems. Busawon et al. (1993) treated the problem of digital implementation of the high gain observer given by Gauthier et al. (1992) via a simple forward Euler discretization scheme. De Leon-Morales et al. (1998) investigated the digital implementation of a high gain controller via a similar forward Euler discretization scheme. It is shown that the result is dual to that given by Busawon-Saif-Farza (1997). In this paper, the observer and controller design based on some of the results from Busawon et al. (1998) for the Euler discretized version of nonlinear system is treated.

Techniques of differential algebra are used to derive an observer-based controller applied to a class of continuous stirred tank reactor (CSTR). A dynamical linearizing feedback controller is obtained on the basis of Fliess' generalized observability canonical form (GOCF) and the Fliess' generalized controllability canonical form (GCCF). Our controller actually achieves both stability and tracking by linearizing the tracking error. The controller is then of error-driven type. Numerical simulations are carried out with the overall control scheme to illustrate the methodology.

In this paper, we propose a new controller design approach for a special class of nonlinear systems. The controller design is simple and systematic and is based on the construction of a similarity transformation which is used in finding canonical forms for linear controllable systems. In addition, the design does not require any coordinate transformation and is directly applied on the original structure of the system. The performance of the proposed controller is shown via a simulation example dealing with a typical synchronous generator

Based on differential algebra techniques, an exponential observer is proposed for the state estimation of a class of nonlinear systems. Simulations illustrating the performance of the proposed observer are included.

Techniques of differential algebra are used to derive an observer-based controller for bilinear systems. The generalized observability canonical form (GOCF) and the generalized controller canonical form (GCCF) introduced by Fliess are the main ingredients in our approach. Our controller actually achieves both stability and tracking by linearizing the tracking error. The controller is then of error-driven type; the dynamics of the error is estimated by an observer in the closed loop.

A differential algebraic approach is proposed for the estimation of the state of a class of bilinear systems. An exponential observer is easily constructed for a single output observable bilinear system class (in the observability sense of Diop and Fliess, 1991). An application to a chemical reactor model is given. >

State estimators for induction motors are generally designed based on standard simplified models, assuming linear magnetic characteristics. Since they are actually nonlinear, especially for high power machines, the mentioned state estimators are likely not able to achieve the estimation accuracy they have been designed for. In this paper, a new state estimator is developed for a (uniform air-gap) AC machine, based on a more accurate model that appropriately accounts for the saturation feature in the magnetic characteristics. The proposed estimator is a high-gain full-order nonlinear observer designed using Lyapunov stability tools. The resulting estimation errors are shown to asymptotically vanish, if their initial values belong to a well defined attraction region. Supremacy of the new observer over standard ones is illustrated by simulation using a 7.5 KW AC machine.

The problem of synthesis of observers for nonlinear systems is considered. An observer for state-affine systems is constructed and it is shown that it depends on the inputs of the system. The inputs for which the observer converges are classified, and some topological properties are given. >

An observer-based controller strategy for a synchronous generator is proposed. Our controller actually achieves both stability and tracking by linearizing the tracking error. Simulation results are given when this scheme is applied to a mathematical model describing the behavior of a synchronous generator.

Based on the differential algebra an exponential nonlinear observer is proposed for the estimation of the state of a class of nonlinear systems. We show an application of this result to a nonlinear model.

A differential algebraic approach is proposed for the estimation of the state of a class of bilinear systems. An exponential observer is easily constructed for a single output observable bilinear system class (in the Diop-Fliess' (1991) observability sense). An application to a chemical reactor model is given.