This thesis deals with the analysis of magnetic losses in electrical machines and the possibilities of their measurement. Magnetically soft and hard materials are very prone to changing magnetic properties. They can be changed simply by changing the temperature of the material or different stresses induced in the material, resulting in different results. These changes are important to keep in mind when an electric machine is being designed. The original parameters of the affected materials can be restored by annealing or grinding. These methods release the induced stresses within the material. Due to these effects, it is also important to know how to measure magnetic parameters. Each way has its own specifics and has a certain error of the measurement. When the machines for higher efficiency, rpm or higher temperatures are designed, it is advisable to know how their magnetic properties changed. This thesis deals with the properties of different materials, their measurements and finally simulation of the chosen electric machine with the application of the measured results.

As the major electricity consumer, electrical machines play a key role for global energy savings. Machine manufacturers put considerable efforts into the development of more efficient electrical machines for loss reduction and higher power density achievements. A consolidated knowledge of the occurring losses in electrical machines is a basic requirement for efficiency improvements. This thesis deals with iron losses in electrical machines. The major focus is on the influences of the stator core magnetic material due to the machine manufacturing process, temperature influences, and the impact of inverter operation. The first part of the thesis gives an overview of typical losses in electrical machines, with focus put on iron losses. Typical models for predicting iron losses in magnetic materials are presented in a comprehensive literature study. A broad comparison of magnetic materials and the introduction of a new material selection tool conclude this part. Next to the typically used silicon-iron lamination alloys for electrical machines, this thesis investigates also cobalt-iron and nickel-iron lamination sheets. These materials have superior magnetic properties in terms of saturation magnetization and hysteresis losses compared to silicon-iron alloys. The second and major part of the thesis introduces the developed measurement system of this project and presents experimental iron loss investigations. Influences due to machine manufacturing changes are studied, including punching, stacking and welding effects. Furthermore, the effect of pulse-width modulation schemes on the iron losses and machine performance is examined experimentally and with finite-element method simulations. For nickel-iron lamination sheets, a special focus is put on the temperature dependency, since the magnetic characteristics and iron losses change considerably with increasing temperature. Furthermore, thermal stress-relief processes (annealing) are examined for cobalt-iron and nickel-iron alloys by magnetic measurements and microscopic analysis. A thermal method for local iron loss measurements is presented in the last part of the thesis, together with experimental validation on an outer-rotor permanent magnet synchronous machine. / <p>QC 20140516</p>

Le transport et la distribution de l'énergie électrique sont traditionnellement réalisés en alternatif (50 Hz ou 60 Hz), un des éléments-clés de ces infrastructures est le transformateur de puissance. Ce dernier est utilisé depuis plus d'un siècle et donc sa conception est maîtrisée (avec des rendements très élevés, supérieurs à 99 %). Depuis quelques années, la part des énergies renouvelables est en constante augmentation. Bien souvent, la production des énergies renouvelables est éloignée des centres de consommation. Or, le transport en courant continu sous haute tension (HVDC) sur de grandes distances est plus rentable. Dans ce cas, nous avons besoin de convertisseurs de puissance fonctionnant pour certains avec des Transformateurs Moyennes Fréquences (TMF) entre 1 kHz et quelques dizaines de kilohertz. Dans ces applications, la recherche du rendement maximal est primordiale. L'augmentation de la fréquence de fonctionnement a pour effet bénéfique de diminuer l'encombrement d'un transformateur. Cependant un certain nombre de problèmes vont apparaître avec cette augmentation. Nous pouvons citer : les pertes dans les conducteurs et dans le circuit magnétique sont liées à la fréquence ; le type de bobinages (fil de Litz et feuillard) et les matériaux magnétiques (ferrites et nanocristallins) en moyennes fréquences sont différents de ceux utilisés en 50 Hz ; le refroidissement est plus complexe car la densité de puissance volumique est plus élevée... Ainsi dans cette thèse, nous avons mis en place une méthodologie de conception afin de maîtriser au mieux le dimensionnement d'un TMF avec un compromis précision et coût de calculs. Nous avons identifié les modèles (analytiques et numériques) susceptibles d'être utilisés pour estimer les performances d'un TMF. Deux TMF d'une puissance de 180 kVA et de 1 kVA ont été dimensionnés, fabriqués et testés afin de mettre en évidence le domaine de validité et d'ajuster les différents modèles. Ce travail nous a permis de mettre en place une méthodologie de conception allant des spécifications du convertisseur jusqu'à la simulation de celui-ci avec le modèle du transformateur dimensionné. Nous avons mis en évidence : l'influence de paramètres technologiques sur l'élévation de la résistance pour des bobinages de type feuillard et l'influence de paramètres technologiques sur les propriétés magnétiques des matériaux nanocristallins. Ce travail de thèse a été réalisé avec le groupe « Matériaux du Génie Electrique » du laboratoire Ampère et financé par l'institut pour la transition énergétique SuperGrid Institute / The transmission and distribution of electric power is normally made by ac networks (50 Hz or 60 Hz), where one of the key elements of this infrastructure is the power transformer; used for more than a century, its design is very well understood, with a level of operating efficiency normally greater than 99%. In recent years, the share of renewable energy has been increasing. Often times the energy generated from renewable sources is produced far from consumption centers, and so transportation in the form of high voltage direct current (HVDC) over long distances is more profitable, due to the lower losses seen than with HVAC after a certain length of transmission line. In this case, we need power converters operating with Medium Frequency Transformers (MFT) from 1 kHz to tens of kilohertz. For these applications, the research of their maximum efficiency in operation is paramount. Increasing the transformer operating frequency has the beneficial effect of reducing its size. However, a number of problems will appear with this frequency increase, such as: the increase in the losses in the conductors and the magnetic circuit that are related to the frequency; the less well understood winding type (Litz wire and foil) and magnetic materials (ferrites and nanocrystalline) in the MF that are different from those used at 50 Hz; the cooling is more complex because the power density is higher, etc. In this thesis, a design methodology was developed in order to optimise the design of MFTs with respect to the compromise between accuracy and the length of calculations. In addition, analytical and numerical models were identified that can be used to accurately estimate the performance of an MFT. Furthermore, two MFTs (apparent power: 180 kVA and 1 kVA, respectively) were sized, manufactured and tested in order to demonstrate the domain of validity of the models, and also for optimisation of the different models. This work has enabled the development of a design methodology using the converter specifications and build a simulation with complete model of the transformer, which can then be used to validate an MFT design. We have highlighted: the influence of the technological parameters on the rise of resistance in the foil coils and the influence of the technological parameters on the magnetic properties of nanocrystalline materials. This work was performed with the group "Materials for Electrical Engineering" Ampère laboratory and funded by the Institute for Energy Transition SuperGrid Institute.