scispace - formally typeset
Search or ask a question

Showing papers by "Hadi Y. Kanaan published in 2002"


Proceedings ArticleDOI
07 Aug 2002
TL;DR: A comparative evaluation of three simulation tools for modeling and numerically implementing switch mode power converters is presented, based on the use of a state-space averaged model of the converter and the Power System Blockset tool of Matlab/Simulink.
Abstract: In this paper, a comparative evaluation of three simulation tools for modeling and numerically implementing switch mode power converters is presented. The first simulation technique is based on the use of a state-space averaged model of the converter, the second approach makes use of the switching function concept, whereas the third one consists of using the Power System Blockset tool of Matlab/Simulink. The three modeling techniques are applied to a same DC-to-DC boost topology operating in a continuous conduction mode, with a fixed switching frequency. These techniques are compared on the basis of their rapidity, precision and limitations.

28 citations


Proceedings ArticleDOI
05 Nov 2002
TL;DR: The derivation of the proposed control law is based on a fourth-order low-frequency time-invariant nonlinear state model of the converter, expressed in a rotating frame, which shows a nearly zero current harmonic distortion, a unity displacement factor, a low output voltage ripple, and a high robustness to DC load unbalance.
Abstract: In this paper, a new nonlinear control system is developed and applied to a three-phase, three-switch, three-level AC-to-DC converter. The derivation of the proposed control law is based on a fourth-order low-frequency time-invariant nonlinear state model of the converter. This mathematical model, expressed in a rotating frame, is obtained using averaging techniques over a double time base. The proposed control law is designed on the basis of a multiple-loops control strategy, where an exact input-output feedback linearization is applied to the inner and outer loops. The regulators are designed in order to ensure the shaping of the line-currents, the regulation of the DC output voltage and the compensation of the DC load unbalance. The performances of the proposed control law in steady state and transient regimes are then analyzed by numerical simulations. In this purpose, a digital version of the converter is implemented in the Matlab/Simulink framework, using the switching function approach. The obtained simulation results show a nearly zero current harmonic distortion, a unity displacement factor, a low output voltage ripple, and a high robustness to DC load unbalance.

18 citations


Proceedings ArticleDOI
10 Dec 2002
TL;DR: In this paper, a linear decoupling strategy for a three-phase current injection rectifier is designed for a pulse-width modulation technique and an elaboration of a small-signal mathematical model that represents the converter in the frequency domain.
Abstract: In this paper, a new control law, based on the linear decoupling strategy, is designed for a three-phase current injection rectifier. The proposed control scheme is based on the pulse-width-modulation technique and on the elaboration of a small-signal mathematical model that represents the converter in the frequency domain. The basic model of the converter is derive from the application of the state-space-averaging technique. The control feedback loops are designed on two levels. Firstly, an inner multivariable feedback subsystem is established in order to ensure line current shaping. The decoupling principle has been used in order to avoid the cross-dependency between the variables and to ensure, consequently, a high dynamic performance. Then, an outer loop is constructed for the regulation of the output voltage. Both current and voltage loops are elaborated using optimizing criteria discussed in the linear control theory. The performance characteristics of the new proposed control approach are highlighted through numerical simulations carried out using Matlab/Simulink. Moreover, the efficiency of the proposed control approach is emphasized through a comparative evaluation with respect to other control laws yet applied to the topology considered in the paper.

5 citations


14 Mar 2002
TL;DR: In this paper, the authors present a comparative analysis of techniques of commande, reglage, and modelisation of redresseurs, in terms of performance, performance dynamiques, impedance d'entree et sa susceptibilite.
Abstract: L'utilisation des redresseurs triphases non polluants pour la conversion alternatif-continu de l'energie electrique est actuellement en pleine croissance. Bien que ces dispositifs soient tres performants en termes de qualite d'onde du reseau et de regulation de tension cote continu, un probleme se pose au niveau de leur commande. La technique des bascules a hysteresis est assez repandue en pratique, ceci etant du surtout a sa simplicite d'implantation. Toutefois, le fait que la frequence de commutation propre a ce type de commande depend considerablement des parametres du convertisseur, de ses conditions d'operation et du temps rend cette commande moins preferable vis-a-vis d'une autre technique dite a Modulation de Largeur d'Impulsions (MLI). Celle-ci se caracterise, en effet, par une frequence de commutation fixe, et assure ainsi au convertisseur un haut degre de fiabilite et de performance en terme de qualite d'onde. La mise en oeuvre d'un systeme de reglage base sur une commande de type MLI necessite, en principe, la connaissance d'un modele mathematique assez precis du convertisseur. De tels modeles sont actuellement disponibles pour des classes particulieres de convertisseurs, notamment les hacheurs et les redresseurs bidirectionnels. Dans le present recueil, la modelisation d'une autre famille de redresseurs, de type unidirectionnel, est proposee. Deux topologies sont choisies a cet egard. Il s'agit, d'une part, du redresseur a injection de courant et, d'autre part, du redresseur a trois interrupteurs et trois niveaux, dit de Vienne. Ces topologies se distinguent, toutes les deux, par un nombre reduit de composants actifs, une simplicite de commande, un faible cout, une robustesse et un rendement energetique eleves. Le procede de modelisation utilise l'approche de la moyenne a l'echelle d'une periode commutation. Les modeles moyens ainsi obtenus sont ensuite utilises dans l'elaboration systematique des lois de reglage appropriees. La validite des modeles et des lois commandes associees est verifiee numeriquement a l'aide d'outils de simulation. Le present recueil est divise en cinq chapitres. Dans le chapitre 1, une revue bibliographique des principales topologies de redresseurs non polluants est effectuee, et l'avantage des structures triphasees unidirectionnelles est souligne. Dans le chapitre 2, une analyse comparative des techniques de commande, de reglage et de modelisation est presentee. Dans le chapitre 3, un systeme de reglage lineaire associe a un redresseur monophase a deux etages est concu. Les performances dynamiques du systeme, ainsi que son impedance d'entree et sa susceptibilite, sont egalement analysees. Le chapitre 4 est consacre a la modelisation et au reglage du redresseur a injection de courant. Le modele du convertisseur est valide numeriquement et experimentalement. Finalement, une modelisation et un reglage non lineaire du redresseur de Vienne sont proposes au chapitre 5. Les performances du systeme sont verifiees numeriquement.

4 citations



Proceedings ArticleDOI
07 Aug 2002
TL;DR: In this paper, a linear control strategy applied to a three-phase current injection AC-DC converter is presented based on the elaboration of the converter small-signal mathematical model, derived from the application of the state-space-averaging technique.
Abstract: In this paper, a new linear control strategy applied to a three-phase current injection AC-DC converter is presented. The proposed control scheme is based on the elaboration of the converter small-signal mathematical model, derived from the application of the state-space-averaging technique. The control design methodology of both current and voltage loops uses linear control theory. More precisely, each regulator's parameters are chosen in order to have an optimal second-order behavior of the corresponding closed-loop system. A simulation experiment using Matlab/Simulink highlights the efficiency of the proposed control approach. Experimental results, obtained from a 2 kW-laboratory prototype, are also presented for validation purposes.