Andreas Ruf

Loss minimizing control strategy for electrical machines considering iron loss distribution

Iron losses have a large share in the overall losses of high power density electrical machines operating as variable speed drives. Therefore, commonly used control strategies such as Maximum Torque per Ampere or Maximum Torque per Voltage aiming at minimizing the copper losses do not select the best direct- and quadrature currents to maximize the efficiency or minimize the overall losses at each operating point. This paper elaborates a loss minimizing control strategy considering the iron loss distribution at all operating points, comparing different iron loss models.

Analysis and determination of mechanical bearing load caused by unbalanced magnetic pull

One common mechanical fault in electrical drives applied to industrial processes are related to bearing damage [1]. These faults can be brought forward by mechanical and thermal stress during the operation. This paper focuses on the radial electromagnetic forces, which are known as static and dynamic unbalanced magnetic pull (UMP), caused by rotor eccentricities. In particular in this paper the interaction of the rotor position and the amplitude of the static and dynamic eccentricity, the slot harmonics and saturation effects in the entire operational range of a permanent magnet synchronous machine (PMSM) are studied.

A methodology to identify electrical ageing of winding insulation systems

Forced by the high requirements on modern variable speed controlled drives, power electronics are used to achieve the specific requirements on accuracy and dynamics. These properties are realized by fast switching converters, which result in voltages and currents with high slew rates. As a consequence, these rise times or slew rates lead to high frequency phenomena throughout the electrical drive. High frequency electrical traveling waves superimpose in the electrical machine, in the supply cables and in the power inverter. Based on this superimposition, high overvoltage occurs at different locations in the electrical drive systems [1]. These high voltage waves result in high electrical fields, which load the insulation by various ageing mechanisms [2]. Therefore, ageing of insulation in low voltage machines is influenced by a combination of the dominant factors: thermal, mechanical, electrical ageing and the interaction of their impacts. For this reason, high frequency phenomena in electrical drives lead to failures, e.g. in the winding system of electrical machines. This paper presents a methodology to identify the electrical load of insulation systems of machine windings. The methodology is based on the use of a high voltage generator. Here, a voltage generator is presented, which is used to identify the electrical ageing procedure for different winding topologies.

Influence of non-linear frequency-dependent material properties on the operation of rotating electrical machines

Purpose – The purpose of this paper is to focus on the frequency-dependent non-linear magnetization behaviour of the soft magnetic material, which influences both the energy loss and the performance of the electrical machine. The applied approach is based on measured material characteristics for various frequencies and magnetic flux densities. These are varied during the simulation according to the operational conditions of the rotating electrical machine. Therewith, the fault being committed neglecting the frequency-dependent magnetization behaviour of the magnetic material is examined in detail. Design/methodology/approach – The influence of non-linear frequency-dependent material properties is studied by variation of the frequency-dependent magnetization characteristics. Two different non-oriented electrical steel grades having the same nominal losses at 1.5 T and 50 Hz, but different thickness, classified as M330-35A and M330-50A are studied in detail. Both have slightly different magnetization and lo...

Lebensdauermodellierung von nicht-teilentladungsresistenten Isoliersystemen elektrischer Maschinen in dynamischen Lastkollektiven

ZusammenfassungIn diesem Beitrag wird ein Verfahren zur Lebensdauermodellierung von nicht-teilentladungsresistenten Isoliersystemen, die in dynamischen Lastkollektiven betrieben werden, an einem Beispiel diskutiert und untersucht. Ziel dieser Untersuchungen ist die Parametrierung eines Lebensdauermodells, welches anwendungsspezifische Belastungen durch Temperatur, Vibration oder Umweltbedingungen, aber auch normative Anforderungen an Prüfspannungen und Teilentladungsfreiheit berücksichtigt. Hierfür werden zunächst beschleunigte Alterungsversuche motiviert und ihre Modellgrenzen diskutiert. Darauf aufbauend werden Probekörper bzw. Motoretten ausgelegt, die das Isoliersystem einer elektrischen Maschine vollständig nachbilden, um diese einer beschleunigten Alterung auszusetzen und schließlich zyklisch hinsichtlich ihrer Teilentladungseigenschaften zu untersuchen. Die Ergebnisse der Lebensdaueruntersuchungen werden mit Hilfe von Zuverlässigkeitskenngrößen und drei verschiedenen Regressionsverfahren ausgewertet, um als Resultat eine belastungsabhängige Wahrscheinlichkeitsverteilung der prognostizierten Lebensdauer zu erhalten.AbstractThis article discusses and examines a process of lifetime modeling of non-partial discharge resistant insulation systems operating in dynamic load collectives. The aim of these investigations is the parameterization of a lifetime model, which considers application specific loads due to temperature, vibration or environmental conditions, but also normative requirements for test voltages and partial discharges. For this purpose, accelerated aging tests are motivated and their model limitations are discussed. Based on this, specimens resp. motorettes are designed which completely reproduce the insulation system of an electrical machine. These are used to study their partial discharge properties during cyclical durability tests. The results of the lifetime investigations are evaluated using reliability characteristics and different regression methods in order to obtain a load-dependent probability distribution of the predicted lifetime.

Consideration of the manufacturing influence in standardized material characterizations using machine measurements

Iron losses have a significant contribution to the overall losses of high power density electrical machines operating as variable speed drives. During production, mechanical stress is applied to the soft magnetic material resulting in local magnetic deterioration and hence rising iron losses. Dependent on the cutting technique, geometrical sizes of tooth and yoke width or external loads from housing and shaft, the resulting local iron losses increase significantly. Hence, standardized Epstein or single sheet measurements under ideal sinusoidal condition underestimate the resulting machines iron losses as the geometrical specimen sizes are to large to quantify the manufacturing influences. This paper applies a semi-physical approach to test bench machine measurements. The approach is derived from the parameter identification of an iron loss formula in due consideration of the different frequency and flux density dependencies of the various iron loss components. Thereby, a calibration of the used iron loss formulation considering manufacturing influences is presented. This allows an a-priori assessment of realistic iron losses during the design stage.

Einfluss von parasitären Effekten und Fertigungsabweichungen auf die Kräfte in elektrischen MaschinenInfluence of parasitic effects and manufacturing deviations on forces in electrical machines

ZusammenfassungIm Fertigungsprozess von elektrischen Maschinen treten unweigerlich Abweichungen auf. Wie bei allen Fertigungsprozessen ist nur eine endliche Fertigungsgenauigkeit erreichbar, so dass jedes gefertigte Bauteil eine Abweichung von seinen idealen Eigenschaften aufweist. In mehrstufigen Fertigungsverfahren akkumulieren sich diese Abweichungen der einzelnen Bauteile. Für die Konstruktion und Fertigung einer elektrischen Maschine werden die Bauteile zwar entsprechend toleriert, während der elektromagnetischen Auslegung hingegen wird weitestgehend von idealen Abmessungen und geometrischen Symmetrien ausgegangen. Eine übliche Asymmetrie beispielsweise ist die Exzentrizität, eine nicht konzentrische Ausrichtung zwischen Rotor und Stator.Bei der Auslegung und der Berechnung der elektrischen Maschine werden in der Regel auch isotrope und homogene Materialeigenschaften angenommen. In permanentmagneterregten elektrischen Maschinen unterliegen die Permanentmagnete beispielsweise je nach Güte einer deutlichen lokalen Abweichung in ihrer Magnetisierungsamplitude und -richtung.All diese Abweichungen haben Einfluss auf die magnetische Flussdichteverteilung in der Maschine und somit auch auf die auftretenden lokalen Kräfte. Summenkräfte, Drehmoment, Verluste, akustisches Verhalten oder Lebensdauer nicht idealer elektrischer Maschinen können somit deutlich von denen in idealen Maschinen abweichen. Wird die Maschine in einem geregelten Antriebsstrang betrieben, sind auch parasitäre Einflüsse durch Antriebsstrangkomponenten wie Leistungselektronik und Regelung von Bedeutung.In diesem Beitrag wird der Einfluss von parasitären Effekten und Fertigungsabweichungen auf die Kräfte in elektrischen Maschinen untersucht. Durch statistische Toleranzrechnung werden die Wahrscheinlichkeiten für ausgewählte Abweichungen bestimmt. Mittels eines zeiteffizienten analytischen Rechenverfahrens werden die durch die Abweichungen zusätzlich entstehenden Kräfte über den gesamten Betriebsbereich der elektrischen Maschine bestimmt und charakterisiert.AbstractIn the production process of electrical machines, various stochastic manufacturing deviations occur. Like all production processes, just a limited accuracy is achievable. Every manufactured component deviates from its ideal properties. During the manufacturing steps, deviations of the individual parts are accumulated. The components of electrical machines are designed and produced within a tolerance range. During the electromagnetic design, ideal dimensions and geometric symmetries are usually assumed. This does not reflect the reality. An example of a typical asymmetric problem is eccentricity, a non-concentric orientation between the rotor and the stator.Furthermore, in the design and simulation of electrical machines, the material properties are normally assumed to be isotropic and homogeneous. Due to the manufacturing process, the material properties can deviate from their ideal characteristics. In electrical machines with permanent magnets, e.g. the magnets have considerably local deviations in magnitude and direction of the magnetization.All deviations influence the flux density distribution in an electrical machine. This leads to the deviation of the calculated local forces. Therefore, the behavior of a real electrical machine, i.e. sum forces, torque, losses, acoustic behavior or operating lifetime, can differ significantly from an ideal machine. For machines operating in controlled drive trains, parasitic effects from the drive train components, such as power electronics and control, have significant roles.In this article, the influences of manufacturing deviations and parasitic effects on forces in electrical machines are presented. With statistic tolerance calculation, the probabilities for relevant manufacturing deviations are determined. With a time efficient analytical analysis tool, the forces originated by manufacturing deviations are calculated. The forces are determined and distinguished for the whole operating area of the machine.

Einfluss von parasitären Effekten und Fertigungsabweichungen auf die Kräfte in elektrischen Maschinen

ZusammenfassungIm Fertigungsprozess von elektrischen Maschinen treten unweigerlich Abweichungen auf. Wie bei allen Fertigungsprozessen ist nur eine endliche Fertigungsgenauigkeit erreichbar, so dass jedes gefertigte Bauteil eine Abweichung von seinen idealen Eigenschaften aufweist. In mehrstufigen Fertigungsverfahren akkumulieren sich diese Abweichungen der einzelnen Bauteile. Für die Konstruktion und Fertigung einer elektrischen Maschine werden die Bauteile zwar entsprechend toleriert, während der elektromagnetischen Auslegung hingegen wird weitestgehend von idealen Abmessungen und geometrischen Symmetrien ausgegangen. Eine übliche Asymmetrie beispielsweise ist die Exzentrizität, eine nicht konzentrische Ausrichtung zwischen Rotor und Stator.Bei der Auslegung und der Berechnung der elektrischen Maschine werden in der Regel auch isotrope und homogene Materialeigenschaften angenommen. In permanentmagneterregten elektrischen Maschinen unterliegen die Permanentmagnete beispielsweise je nach Güte einer deutlichen lokalen Abweichung in ihrer Magnetisierungsamplitude und -richtung.All diese Abweichungen haben Einfluss auf die magnetische Flussdichteverteilung in der Maschine und somit auch auf die auftretenden lokalen Kräfte. Summenkräfte, Drehmoment, Verluste, akustisches Verhalten oder Lebensdauer nicht idealer elektrischer Maschinen können somit deutlich von denen in idealen Maschinen abweichen. Wird die Maschine in einem geregelten Antriebsstrang betrieben, sind auch parasitäre Einflüsse durch Antriebsstrangkomponenten wie Leistungselektronik und Regelung von Bedeutung.In diesem Beitrag wird der Einfluss von parasitären Effekten und Fertigungsabweichungen auf die Kräfte in elektrischen Maschinen untersucht. Durch statistische Toleranzrechnung werden die Wahrscheinlichkeiten für ausgewählte Abweichungen bestimmt. Mittels eines zeiteffizienten analytischen Rechenverfahrens werden die durch die Abweichungen zusätzlich entstehenden Kräfte über den gesamten Betriebsbereich der elektrischen Maschine bestimmt und charakterisiert.AbstractIn the production process of electrical machines, various stochastic manufacturing deviations occur. Like all production processes, just a limited accuracy is achievable. Every manufactured component deviates from its ideal properties. During the manufacturing steps, deviations of the individual parts are accumulated. The components of electrical machines are designed and produced within a tolerance range. During the electromagnetic design, ideal dimensions and geometric symmetries are usually assumed. This does not reflect the reality. An example of a typical asymmetric problem is eccentricity, a non-concentric orientation between the rotor and the stator.Furthermore, in the design and simulation of electrical machines, the material properties are normally assumed to be isotropic and homogeneous. Due to the manufacturing process, the material properties can deviate from their ideal characteristics. In electrical machines with permanent magnets, e.g. the magnets have considerably local deviations in magnitude and direction of the magnetization.All deviations influence the flux density distribution in an electrical machine. This leads to the deviation of the calculated local forces. Therefore, the behavior of a real electrical machine, i.e. sum forces, torque, losses, acoustic behavior or operating lifetime, can differ significantly from an ideal machine. For machines operating in controlled drive trains, parasitic effects from the drive train components, such as power electronics and control, have significant roles.In this article, the influences of manufacturing deviations and parasitic effects on forces in electrical machines are presented. With statistic tolerance calculation, the probabilities for relevant manufacturing deviations are determined. With a time efficient analytical analysis tool, the forces originated by manufacturing deviations are calculated. The forces are determined and distinguished for the whole operating area of the machine.

Stator Current Vector Determination Under Consideration of Local Iron Loss Distribution for Partial Load Operation of PMSM

Iron losses have a large share in the overall losses of high power density electrical machines operating as variable speed drives. Especially in the partial load area, the ratio of iron losses is dominant with respect to the copper losses. Therefore, commonly used control strategies, such as maximum torque per ampere or maximum torque per voltage, aiming at minimizing the current for given restrictions as the maximum voltage do not select the best direct and quadrature currents to maximize the efficiency or minimize the overall losses at each operating point. This paper elaborates a stator current vector determination strategy considering the iron loss distribution at all operating points, comparing different iron loss models in a machine with distributed windings. Depending on the operating point and the operational conditions of the machine, the overall losses can be reduced up to 7%.