Individual information

A Hybrid Energy Storage System (HESS) uses DC-DC converters to couple its energy sources. However, this device represents a "dead weight" in the system and must be reduced to a minimum in order to maximize the HESS' performance. This work proposes a new coupling architecture to reduce the converter's volume and mass. Not yet addressed in the literature, this architecture is based on a series coupling of the sources. In this case, a DC-DC converter is used to control the current difference between the two sources. If this difference is zero, so is the power processed by the converter. By reducing the power processed by the converter, its mass and volume can be reduced. Simulation and experimental tests were carried out to validate the architecture concept. For the latter, power supplies were used to emulate the batteries and the load, and a 2 kg, 3.3 kW evaluation board served as the DC-DC converter. The results show that, compared to a conventional solution that is usually adopted in the literature, with the series architecture, it is possible to reduce the converter sizing power by almost 3.7.

High-frequency and high-power density converters are essential for many applications such as automotive and more electric aircraft. To improve power density, planar magnetic components are increasingly used in power electronic converters owing to their advantages, such as less high-frequency losses, repeatability, and better thermal performance compared to conventional wound components. Considerable attention has been paid to their electrical performance. However, the thermal aspects are essential for enhancing completely their performance. Currently, there is an increasing interest in these critical aspects of high-density power converters. This paper presents a literature review of studies on the thermal modeling of planar magnetics. The papers were organized according to the model type and analyzed to highlight their merits and limitations. A multicriteria comparison of the studied papers was performed. Moreover, the key points and issues related to the thermal modeling are addressed to guide designers in enhancing the thermal aspects of planar magnetic components.

Conceiving planar magnetic components for power electronic converters is very constraining, especially in the case of prototype development. Indeed, such making requires skills, specific appliances as well as human time for setting up the machine tools and the fabrication process. With the emergence of Fabrication Laboratory (FabLab), conceiving of planar copper foil prototypes becomes more feasible in a shortened time process for engineers and researchers. This paper presents a methodology and process for conceiving power planar transformers with the help of machines and tools that can be found in the usual FabLab.

Thermal performance of power converters is a key issue for the power integration. Temperatures inside active and passive devices can be determined using thermal models. Modelling the temperature distribution of high frequency magnetic components is quite complex due to diversity of their geometries and used materials. This paper presents a thermal modelling method based on lumped elements thermal network model, applied to planar magnetic components made of EE and E/PLT cores. The 3D model is automatically generated from the component’s geometry. The computation enables to obtain 3D temperature distribution inside windings and core of planar transformers or inductors, in steady state or in transient case. The paper details the proposed modelling method as well as the automated tool including the problem definition and the solving process. The obtained temperature distributions are compared with Finite Element simulation results and measurements on different planar transformers.

The More Electric Aircraft (MEA) has motivated aircraft manufacturers since few decades. Indeed, their investigations lead to the increase of electric power in airplanes. The challenge is to decrease the weight of embedded systems and therefore the fuel consumption. This is possible thanks to new efficient power electronic converters made of new components. As magnetic components represent a great proportion of their weight, planar components are an interesting solution to increase the power density of some switching mode power supplies. This paper presents the benefits and drawbacks of high frequency planar transformers in DC/DC converter, different models developed for their design and different issues in MEA context related to planar’s specific geometry and technology.

Loss values are key parameters for designing high performance high frequency (HF) magnetic components for power electronics (PE) converters. With the increase of PE switching frequencies, copper losses have to be precisely quantified, ideally until some megahertz. In the literature, many 2-D numerical simulations based on finite element analysis (FEA) are performed for such computations. 3-D FEA studies of planar components are still limited because of modeling problems, computational resources and computing time. In this paper, quantitative comparisons between 2-D and 3-D simulation results for planar inductors are achieved focusing on copper loss computation. Results are compared in terms of simulation performances and accuracy. The aim of the paper is to highlight benefits of 2-D and 3-D FEA simulations in order to choose the appropriate model according to the studied problem.

Planar magnetic components are promising solutions for the integration of power electronic systems. The leakage inductance of such components plays an essential role in power converters. In this paper, an analytical modeling method for leakage inductance computation is developped for planar components with plasto-ferrite leakage layers. This method is based on the solution of Poisson’s equations for magneto-static using multilayered Green’s functions. The obtained formulations are general and precise and have been validated by numerical tests. Experimental characterizations have been performed on two magnetic components: A planar LLC and planar common mode choke. The obtained results show that with the described method, the static leakage inductance of planar components can be accurately estimated.

Predetermination of losses and inductance values in the design phase, is necessary for the development of high-performance magnetic components for power electronics. Numerical modeling, based on the Finite Element Method (FEM) can be used to determine the characteristics of a particular component with a complex geometry in high frequency (HF). These models are very accurate but the computation time required is high compared to analytical models. The Model Order Reduction (MOR) methods can be applied to reduce the computation time while maintaining high accuracy. Nowadays, the Proper Orthogonal Decomposition (POD) is the most popular of MOR approaches. This technique has been applied to study problems in many elds of engineering. In this paper, the POD method is developed to solve magneto-harmonic problems in order to study a planar magnetic inductor.

Two current probes are designed with the appropriate magnetic material in order to make impedance measurements in High Frequency (HF) using the Current Injection Method (CIM). These probes are then used to measure the common-mode impedance of power converters in real-operating conditions. The characterization of this impedance is required for a correct EMI filter design. In this paper, a simple formulation of the probe transfer impedance, based on S-parameters, is proposed. These probes allow improving the accuracy of impedance measurements in a wide frequency range, up to 100 MHz. The measured impedances of the power converter are then applied in the design of the common mode EMI filter under real operation conditions. The obtained insertion losses of the filter are finally compared with those measured under 50Ω-50Ω

Power converters with high switching frequency generate conducted electromagnetic interference (EMI) noise. EMI filters are thus widely used to reduce these conducted noises for the compliance with electromagnetic compatibility standards. In this paper, a high-frequency (HF) equivalent circuit model for common mode (CM) chokes used in EMI filters is proposed together with its parameter extraction procedure. This procedure is based on impedance measurements and it incorporates an iterative rational function approximation fitting algorithm to extract the parameters in the model. The proposed model and procedure is applied to a planar CM choke which is used to realize an EMI filter. The simulated results of the filter show good agreement with the experimental ones. This extraction procedure is quite general and it can also be extended to identify the HF model of other passive components.

This paper presents an analytical method for calculating the parasitic capacitances of planar components with 3 dimensions. The calculation is based on the Electric Field Decomposition (EFD) between multi-conductors. 2D formulations used in micro-electronics have been extended to enable the applications on the cases of planar magnetic components. The 3D calculation is then based on an energy approach with the consideration of the magnetic core effects. Prototypes of inductors and coupled inductors allow the validation of this fully analytical approach.

Abstract--- Electromagnetic interference (EMI) filters are main solutions to suppress the conducted emissions from static power converters. For some years, integration techniques for EMI filters have been widely investigated to realize more compact systems. In this letter, a common mode (CM) choke with toroid-EQ mixed structure is presented. This structure is constituted of a toroidal CM choke embedded into an EQ core. The proposed toroid-EQ CM choke presents three-fold advantages. First, it effectively increases the leakage inductance of the toroidal CM choke and hence increases the DM inductance. Secondly, it reduces the parasitic coupling between the choke and the filter capacitors. Finally, the component can be easily fabricated with low cost. Experimental verifications have been carried out to validate the effectiveness of the proposed toroid-EQ CM choke.

This paper deals with a high-frequency modeling method of the coupled inductors used in electromagnetic interference (EMI) filters. These filters are intended to reduce conducted emissions generated by power static converters towards the power grid. To model the EMI filters, it is necessary to identify the various parameters of the passive elements: inductors and capacitors. Because of their major impact on filter efficiency, these elements must be identified with accuracy. In this study, high-frequency model of common-mode-coupled inductors is proposed. The identification of the model parameters is based on the experimental approach. Simulation results of the proposed model are compared to the experimental data obtained using the specific experimental setup. These results made it possible to validate the EMI filter model and its robustness in a frequency range varying from 9 kHz to 30MHz. The proposed high-frequency inductor models will be very helpful for design and optimization of EMI filters, since the high-frequency behavior of the filter mainly depends on magnetic materials used and on the geometrical characteristics of winding.

The predetermination of the leakage inductances of transformers is essential for component designers. Except for special winding disposal, the usual analytical methods of evaluation are ineffective, while the finite-element-method simulation evaluation requires too much time to be suitable for any optimization process. In a previous work, we supplied the analytical expression for vector potential from which we deduced leakage inductance through numerical integration. In this paper, an explicit expression of leakage inductances is given. This expression does not appear as a series, requires shorter computing time, and opens up a lot of future applications relying on optimization software. Moreover, the calculation is not restricted to leakage inductances: It has been extended to all the parameters characterizing the whole leakage behavior of any transformer. The influence of ferrite permeability and thickness is also investigated.

As a consequence of the increase of power needs in low-voltage applications, wiring several windings in parallel to sustain large currents has become common. This choice may have a serious impact on the transformer reliability. In practice, due to additional currents that we call “circulation currents,” highly localized extra losses occur. Although hot points resulting from these currents can destroy the component, the related losses are generally not taken into account by analytical approaches. In planar transformers, windings are made of printed circuit board layers, and circulation currents lead to severe unbalance of current sharing between parallel layers. This paper presents an analytical method enabling the evaluation of the currents in every layer using only a circuit simulation software such as Pspice or PSIM. For a designer, this method is very intuitive and fast compared with the use of finite element method simulations. Several extensions of the modeling method used here are also presented.

High-Frequency (HF) transformer is a central part of isolated HF power converters. Its parameters, in particular leakage inductance and losses, have a significant impact on the overall converter performances. With the increase of switching frequencies linked to the use of SiC and GaN based active devices, the control of the HF transformer parameters becomes essential. A HF transformer is usually designed without considering its surrounding. However, the latter can have a significant impact on the transformer's performance. In this paper, the effect of an aluminum casing surrounding the transformer is studied and quantified for two parameters: The transformer leakage inductance and the supplementary losses. The goal is to consider the casing effect in the early design stage of power converters.

Recent studies have shown that the use of battery-battery coupling in Hybrid Energy Storage Systems (HESS) presents advantages in terms of mass, volume and cost when compared to the battery-supercapacitor coupling. However, the sizing of this type of system is not much studied in the literature. So, in this paper a graphical sizing method using Ragone plots is presented. With this method, a HESS perfectly suited to a given application can be obtained.

When analyzing the characteristics of a Hybrid Energy Storage System in a Ragone plot, it can be seen that the mass of the necessary power electronics has a great influence on the system’s performance. Although many coupling architectures have been studied so far, the size of the power converter is not usually a priority. So, in this paper, a low-volume and high-efficiency converter is proposed. Placed in series between the sources, the converter depends only on the voltage difference between them. In this way, the converter can be reduced, since it does not manage the total power supplied by the accumulators, as a classic converter.

This paper presents a topology optimization process for tuning the leakage inductance of a High Frequency (HF) transformer. The 2D Magnetic Equivalent Circuit (MEC) is used to model the studied transformer. Then, through an optimization problem, primary winding positions ensuring maximal and minimal leakage inductances will be gotten.

This paper presents a methodology to evaluate and analyze the volumetric power density of planar magnetics used in power electronics converters. The power density is computed for various EE and E/PLT cores considering optimal configurations for the planar transformers’ design and for its cooling heatsink. The analysis is performed for three cooling configurations: natural convection without heatsink, single sided cooled component with one heatsink, and double sided cooled with two heatsinks. This study can be very useful for designers to evaluate their design specifications and to adapt their technological choices to achieve the desired planar magnetics’ characteristics.

In this paper, the thermal resistance of planar magnetic components in natural convection is studied in order to obtain analytical formula taking into account effects of ambient temperature as well as power dissipated inside the component. The analytical model is deduced from Computational Fluid Dynamic simulations and is validated with measurements on prototypes.

Planar technology is an interesting solution for high frequency (HF) magnetic components in power electronic converters. Over the past few decades, the trend is increasingly towards these integrated magnetic components with high power density, especially for embedded systems, automotive and avionics devices. In order to design optimized planar inductances or transformers any aspect of the component’s geometry is important. This paper is focused on one of these aspects: the track shape of copper conductors made of Printed Circuit Board (PCB). Three designs of conventional tracks are simulated by 3D Finite Element Analysis (FEA) to calculate the dynamic (AC) resistances and their resistive factor. Different combinations of these elementary tracks are tested and the influence of the turn shape is highlighted.

Predetermination of inductances and losses are needed to effectively design magnetic components used in power electronic converters. Their behaviours change with frequency and have to be determined accurately over a wide frequency range, commonly up to 30MHz to account for EMI aspects. In many papers, 2D finite element simulations are widely performed for the calculations of inductances and losses of planar components. These 2D numerical models are generally taken as references to develop analytical models, including the specific case of High Frequency copper losses. 3D studies of such planar components are still limited because they leads to problems related to modeling, computing resources and computing time. In this paper comparisons in terms of performance and precision for 2D and 3D models of planar inductors is presented. The obtained results are compared with impedance measurements on prototypes.

This paper presents a method to improve EMI filters performances, based on the optimization of the component layout. It’s well known that the EMI filter insertion losses are highly degraded by the high frequency stray couplings between filter passive components. The latter can be reduced by acting on these components’ placement. This paper proposes a method that allow to determine the components’ positions for minimizing the parasitic coupling between them. Their placement is deduced from the study of current distributions and magnetic field induced by these components. This method examines more specifically the parasitic inductive couplings between inductors and capacitors. The choice of the placement that minimizes couplings in 3D is determined to achieve better performance. Some prototypes are realized ​​and characterized to validate the proposed method.

Two current probes, based on the current injection method, are designed with appropriated high frequency magnetic material. These probes are used to measure the common-mode impedance of power converters in real-operating conditions. The characterization of this impedance is of importance for the EMI filter design. In this paper, a simple formulation of the probe’s transfer impedance using S-parameters is proposed. The proposed probe allows improving the accuracy of the impedance measurement in a wide frequency band up to 100 MHz. Different measurements of common-mode configurations are detailed and discussed.

EMI filters are essential parts in power converters to reduce the conducted EMI emissions in order to meet with EMC standards. However, the parasitic capacitances of Common Mode (CM) inductors of filters limit their performances on high frequencies. This paper proposes a complete procedure for modeling the parasitic capacitances of CM inductors in planar technology. The proposed method uses analytical multilayered Green function to determine the capacitance matrix in 2D configuration. Then an energetic approach is applied for calculating the equivalent parasitic capacitances. The proposed procedure is validated by measurements on several prototypes. The calculated results show good agreement with the measured ones.

The passive magnetic ferrite components used in EMI filters have a complex magnetic permeability (CMP) which varies with frequency. The CMP in frequency range from 150 kHz to 30MHz is not given by manufactures. Moreover, measurement techniques require some specific size and shape of sample to measure CMP of ferrites. Common application of these passives components are EMI filters, where toroid ferrite cores need to be characterized to determine the insertion loss function. In this paper a CMP measure procedure is proposed. The leakage inductance and parasitic capacitance of an inductor are determined. Then, a rational function approximation (RTA) is applied to represent the HF equivalent impedance of CMP.

Power semiconductor components with high switching speed are widely used in static converters. However, they pro-duce conducted electromagnetic interferences in high frequencies. Filters are one solution for reducing the conducted emissions. However, the parasitic elements of the passive components in the EMI filter deteriorate its performances. In this paper, we propose to study a differential mode (DM) inductor in planar technology. The goal is to reduce the parasitic capacitance of the planar DM inductor via an improved parasitic capacitance cancellation technique. The technique is based on the results of an analytical method using Electric Field Decomposition and energy based approach. The cancellation is then realized through the structural parasitic capacitances under an optimal geometry configuration. The efficiency of the proposed cancellation technique is validated by measurements.

This paper presents a new course on lighting systems in a French non-specialized engineer school. This course was conceived to increase student attractiveness for power electronics. After two years of unfolding, the conclusions are interesting. A survey has shown that students were not really more interested in power electronics. The only students' plebiscite is the lighting design using DIALux software. The module is described and placed into the context of a non-specialized engineer school. Students' opinion is then analyzed and capitalized to find why power electronics does not look as an interesting subject for these specific engineers.

Impedance measurements are widely used to characterize the behavior of n-windings magnetic components. The identification process is realized manually and becomes more and more difficult as the number of equivalent circuit parameters increases. This paper deals with an automatic identification method based on Rational Function Approximation (RFA) which enables to obtain these parameter values very quickly and leads to reliable models. The effectiveness of such method is demonstrated with the characterization of a planar common mode inductor and a toroidal coupled inductor.

Active learning in high engineering school is a real opportunity for students. Indeed, it enables to discover some subject deeper than a traditional lectures classroom. In this paper a project example, mainly based on power electronics, has been described. This project has taken place in a non-specialized French engineer school with the financial partnership of Renault. Each actor gives its advice at the end of the project (student, professor and industrial partner) about the learning process in power electronics but also about the global project activity.

This paper analyzes the electrical parameters of differential-mode (DM) and common-mode (CM) inductors used in EMI filters. To design an EMI filter, it is necessary to identify the various parameters of the passive elements: inductors and capacitors. Because of their major impact on filter efficiency, these elements must be identified with accuracy. In this study, a high frequency model of coupled inductors is proposed. Simulation results of the proposed model are compared to the experimental data obtained using a specific experimental setup. These results made it possible to validate the filter model in frequency range varying from 9 kHz to 30 MHz. The proposed high frequency coupled inductors model will be very helpful for EMI filter optimization design.

This paper deals with a high frequency modeling method of the common and differential mode inductors used in EMI filters. These filters are intended to reduce conducted emissions generated by power static converters towards power network. To model the EMI filters, it is necessary to identify the various parameters of the passive elements: inductors and capacitors. Because of their major impact on filter efficiency, these elements must be identified with accuracy. In this study, high frequency model of common mode coupled inductors is proposed. Simulation results of the proposed model are compared to the experimental data obtained using the specific experimental setup. These results made it possible to validate the proposed filter model and its robustness in a frequency range varying from 9 kHz to 30 MHz. The proposed high frequency inductor models will be very helpful for design optimization of EMI filters, since the high frequency behaviour of the filter mainly depends on magnetic materials used and on the geometrical characteristics of winding.

Impedance analyzers are efficient in acquiring wide range impedance measurements in a wide frequency range. However, when measuring low impedances, short circuits used for compensation process can’t be considered as ideal and analogue problem appears for open circuit compensation. Fortunately, because impedance matrix of a two-port passive circuit is symmetrical, characterizing this circuit by four independent impedance measurements introduces redundancy. That property supplies a useful consistency test we call "confidence factor". Looking at this factor, several causes of measurement inaccuracy can be detected. Alternatively, extra equation supplied by redundancy can also be used to add one unknown impedance to the three computed. This way, impact of the impedance of a real imperfect short-circuit can be removed. In this paper, we first present the interest of the confidence factor and then we show how to lower errors related to imperfect shortcircuit. Finally, we show how to use these advances to better characterize HF power transformers and to measure weak wiring inductances

Lumped element equivalent circuits that accurately represent the electrical behavior of H.F. n-winding transformers are now available. These circuits suit for time domain simulation but evaluation of their parameters relies on measurements. In order to avoid prototype manufacturing, engineers need to deduce same parameters from geometrical and physical data and, to carry out optimization, they prefer analytical approach to fem one. This paper shows how to do this for planar transformers which have simple geometry. Comparison of results with experimental data is given and analyzed.

L’inductance de fuite est un paramètre essentiel des transformateurs HF dans les convertisseurs DC/DC isolés. Avec l’utilisation de la technologie planar et des enroulements en PCB, ce paramètre peut être réglé à une valeur prédéterminée en ajustant le positionnement des conducteurs dans la fenêtre des transformateurs. Cet article présente une étude autour de la minimisation, de la maximisation et du réglage de l’inductance de fuite d’un transformateur planar. La méthodologie développée est basée sur des simulations par éléments finis (FEM) couplées avec un algorithme d’optimisation multi-objectifs qui permettent de régler une inductance à une valeur donnée, tout en minimisant les pertes cuivre induites.

Cet article étudie des solutions permettant de remplacer les entrefers dans des inductances planar à base de noyaux E/PLT. En effet, ces entrefers, indispensables pour ajuster la perméabilité globale, engendrent des pertes supplémentaires dans les enroulements PCB. Différentes configurations basées sur l’utilisation des matériaux magnétiques à faible μ sont étudiées par simulations éléments finis 2D et comparées en termes de pertes cuivre HF et de valeur de l’inductance. Une amélioration significative des pertes cuivre HF est démontrée avec une réduction de 7 fois la résistance AC par rapport à la configuration avec entrefer.

Les sources de stockage hybrides (HESS) utilisent des convertisseurs DC-DC pour assurer le couplage des sources d'énergie. Pour des applications où l’espace disponible pour embarquer la batterie est limité, le volume et la masse de l’électronique de puissance doivent être réduites au maximum. C’est le cas des véhicules électriques. Dans ce travail, une nouvelle architecture de couplage permettant de réduire le volume et la masse du convertisseur est proposée. Cette architecture, basée sur un couplage en série des sources, est comparée aux solutions classiques. Elle permet de réduire très fortement la puissance de dimensionnement de l’électronique de puissance du HESS.

La méthode du produit des aires est très répandue pour le dimensionnement des composants magnétiques en électronique de puissance, grâce notamment à sa simplicité de mise en œuvre. Néanmoins, l’utilisation de cette méthode nécessite des itérations pour affiner le choix du noyau et la définition des bobinages afin de respecter les contraintes thermiques des composants. Dans cet article, la méthode du produit des aires est étendue pour pouvoir y incorporer la prise en compte des effets haute fréquence et les contraintes thermiques liées aux composants magnétiques planar. L’expression finale obtenue permet un dimensionnement plus efficace et plus rapide des composants planar en évitant les itérations liées aux contraintes thermiques. La différence entre les deux approches est illustrée au travers d’un exemple de dimensionnement de transformateur planar (100 kHz / 2 kW).

La conception rapide de prototypes de laboratoire est intéressante pour permettre la validation, à moindre frais, de dimensionnements spécifiques de transformateurs planar. Avec l’apparition des Fabrication Laboratory (FabLab), il devient possible, en utilisant les outils et machines disponibles dans ces derniers, de réaliser de manière assez simple, rapide et peu coûteuse, des prototypes de transformateur planar en technologie feuillard. Cet article présente donc une méthodologie permettant de concevoir ce type de composant, pour des puissances de quelques kilowatts, en utilisant le matériel usuel des FabLab.

Cet article présente une méthodologie permettant l’analyse des densités de puissance volumique qui peuvent être obtenues via l’utilisation de composants magnétiques planar au sein des convertisseurs d’électronique de puissance. En se basant sur une configuration optimale pour le transformateur, les densités de puissance volumique maximales atteignables pour différents noyaux magnétiques EE et E/PLT sont étudiées pour trois types de refroidissement : sans refroidisseur, refroidissement simple face et refroidissement double face. Ce type d’analyse peut être utile à un concepteur, lors d’une phase de dimensionnement, en le guidant vers des choix technologiques adaptés.

Les composants magnétiques planar sont de plus en plus présents dans les convertisseurs de puissance en remplacement des composants bobinés classiques. L’aspect thermique de ces composants est essentiel pour un fonctionnement correct des structures de puissances. Dans cet article, un modèle thermique de transformateurs planar, basé sur un réseau nodal de résistances thermiques (RRT), est développé. Ce réseau est appelé structurel car toutes les résistances thermiques sont directement en lien avec la géométrie du composant. Les résultats de ce modèle seront comparés à des résultats issus de simulations par éléments finis ainsi qu’à des mesures expérimentales sur un prototype.

La technologie planar est une solution intéressante pour les composants magnétiques haute fréquence (HF) dans les alimentations à découpage. La miniaturisation des composants de puissance liée à leur montée en fréquence a permis le développement, depuis quelques décennies, des composants magnétiques planar, notamment dans les systèmes embarqués à forte densité de puissance. Dans l’optique d’optimiser les inductances et transformateurs planar, chaque aspect de leur géométrie est à prendre en considération. Cet article mettra l’accent sur l’un de ces aspects : la forme des angles des pistes de circuit imprimé (PCB). Trois formes standards de pistes avec différents angles seront simulées par éléments finis 3D pour évaluer leurs résistances dynamiques (AC) et leurs facteurs résistifs. Différentes combinaisons de ces formes élémentaires seront ainsi testées et comparées, et l’influence des angles des pistes sera soulignée.

La prédétermination, dans la phase de dimensionnement, des pertes et des valeurs d’inductances est nécessaire à l’élaboration de composants magnétiques performants pour l’électronique de puissance. La modélisation numérique peut être utilisée pour déterminer les caractéristiques propres d’un composant notamment lorsque sa géométrie est "complexe" ou que la fréquence d’utilisation est élevée. Ces modèles sont très précis mais le temps de simulation nécessaire est important par rapport à des modèles analytiques plus simplistes. Les méthodes de réduction de modèles, type POD ou autres, permettent de diminuer le nombre de ces simulations numériques tout en conservant une grande précision.

As part of the More Electric Aircraft (MEA), aeronautics manufacturers are moving towards switching-mode power supplies (SMPS) solutions with high power density (until 15kW/kg for electrical car [3]) to replace pneumatic and hydraulic actuators. These power electronic devices have two main advantages: Firstly, they reduce the mass of conventional equipments and secondly, they also reduce the fuel consumption by a few percent. These improvements were made possible thanks to the evolution of electronic component manufacturing techniques that lead to the miniaturization of electrical equipment in recent decades. In power electronics, whether in automotive or aerospace, the trend is increasingly towards these integrated solutions, compact with high power density. Inside power converter, planar components, made of PCB (Printed Circuit Board) or with copper foil, fit into this context and offer an interesting alternative to conventional High Frequency transformers. They offer excellent electromagnetic and thermal characteristics [2]. They can also be integrated easily inside electrical circuits and have good rigidity, good repeatability and good modularity that make easier their industrialization (automated mass production) [1]. The optimized design of such component (losses, volume and mass) is essential for a better power density of all of the power converters. Based on analytical expressions and finite element models (FEM), this PhD work deals with the development of optimized planar components for aeronautics. Different configurations of planar transformer windings are studied in order to reach some objectives such as the fewest losses (including skin and proximity effects), a controlled and optimized leakage inductance value, a reduction of the parasitic capacitances (linked to transformer’s resonance) and a weak EMC (Electromagnetic Compatibility) impact in a given environment.

La pré-détermination des valeurs de pertes et d’inductances est nécessaire à l’élaboration de composants magnétiques performants pour l’électronique de puissance. Leurs caractéristiques propres, variant en fonction de la fréquence, doivent être déterminées de manière précise sur une large gamme de fréquences, idéalement jusqu’à 30MHz pour tenir compte des aspects CEM. Dans la littérature, des simulations éléments finis 2D sont largement utilisées pour ce type de calcul. Ces modélisations servent généralement de référence pour le développement de modèles analytiques, notamment pour le cas spécifique des pertes cuivre. L’étude en 3D de ce type de composant est encore très peu développée car elle engendre des problèmes de modélisation, de ressources informatiques et de temps de calcul. Dans cet article la comparaison en termes de performances et de précisions pour des modèles d’inductances planar en 2D et en 3D est présentée. Les résultats obtenus sont comparés à des mesures d’impédances sur des prototypes.

Cet article présente une méthodologie de calcul de capacités parasites pour un composant magnétique de type planar en 3 dimensions. La méthode proposée est basée sur la décomposition du champ électrique entre plusieurs conducteurs (EFD). Des formulations 2D utilisées en micro-électronique ont été étendues pour pouvoir être utilisées dans le cas de composants magnétiques planar. Le calcul en 3D basé sur une approche énergétique, en tenant compte des effets du matériau magnétique, a été proposé. Des prototypes d’inductances et d’inductances couplées ont permis de valider cette approche analytique.

Cet article présente une méthodologie pour l’identification automatique des éléments d’un circuit électrique équivalent à partir de mesures d’impédances. L’algorithme de détermination des paramètres électriques des composants magnétiques est basé sur une décomposition améliorée en fonction rationnelle (IRFA) qui permet de modéliser n’importe quel impédance (ou admittance) avec un degré de précision voulu. Cette décomposition est ensuite transcrite en circuit électrique équivalent de manière systématique en utilisant l’expansion de Foster. Tous ces outils ont été implantés dans une application logicielle pour permettre la caractérisation d’inductances de mode commun d’un filtre CEM. L’approche systématique peut permettre l’extension de cette méthode à tous les composants magnétiques.

L’apprentissage de l’Electronique de Puissance (EP), domaine essentiel à la réalisation de systèmes électriques efficaces, est une tâche délicate, surtout pour un ingénieur généraliste. Dans cet article, un projet étudiant, dont le but est d’exploiter l’énergie solaire afin de réduire les émissions en CO2 de véhicules automobiles, est décrit. Il expose l’approche scientifique, les résultats et comment une telle activité permet un apprentissage actif et efficace de l’EP.

L’idée de cet article est de développer un testeur de modules solaires permettant la caractérisation en courant-tension et donc en puissance de modules solaires photovoltaïques. Ce testeur, simple, portatif et peu coûteux, permet de tester facilement le fonctionnement d’un panneau en traçant l’évolution de ses grandeurs, qu’il soit immobile ou en mouvement. Dans ce dernier cas de figure, l’intérêt est de pouvoir en déduire un modèle comportemental de modules solaires en mouvement afin de pouvoir optimiser les paramètres de réglage d’un algorithme de type MPPT, pour des panneaux embarqués sur des véhicules.

Cet article propose de mettre en pratique la méthode de décomposition du champ électrique pour les conducteurs en technologie PCB, avec pour application, une évaluation analytique de l’effet capacitif des composants planars. En combinant des éléments de décomposition élémentaires, des formules analytiques ont été proposées pour étendre à l’électronique de puissance ces formulations déjà utilisées en micro-électronique. Les résultats obtenus ont été validés à l’aide de simulations FEM et de mesures d’impédances réalisées sur différents prototypes. Ces formules, simples et précises permettent d’envisager, à terme, le calcul analytique des capacités parasites des composants planar.

Les composants magnétiques planar sont de plus en plus présents dans les convertisseurs de puissance en remplacement des composants bobinés classiques. L’aspect thermique de ces composants est essentiel pour un fonctionnement correct des structures de puissances. Dans cet article, un modèle thermique de transformateurs planar, basé sur un réseau nodal de résistances thermiques (RRT), est développé. Ce réseau est appelé structurel car toutes les résistances thermiques sont directement en lien avec la géométrie du composant. Les résultats de ce modèle sont comparés à des résultats issus de simulations par éléments finis ainsi qu’à des mesures expérimentales sur un prototype.