05 Fakultät Informatik, Elektrotechnik und Informationstechnik
Permanent URI for this collectionhttps://elib.uni-stuttgart.de/handle/11682/6
Browse
15 results
Search Results
Item Open Access Precise voltage measurement for power electronics with high switching frequencies(2018) Nitzsche, Maximilian; Zehelein, Matthias; Tröster, Nathan; Roth-Stielow, JörgIn this paper different approaches in precise measurement of gate voltages as well as drain-source voltages of modern SiC and GaN transistors are compared. An approach to calculate the necessary bandwidth of a voltage probe to reproduce the voltage slope is presented. Furthermore, state-of-the-art voltage probes are compared in means of bandwidth, common mode reduction and response on EMI.Item Open Access Optimal transistor dimensioning in T-type topology for reduced quasi-2-level switching loss(2025) Söllner, Adrian; Jie, Chengcong; Mönch, StefanA quasi-2-level switching T-type topology reduces hard-switching loss compared to half-bridges, but requires more semiconductor area. This work shows that the middle transistor can be dimensioned smaller than the high/low-side transistors, which further reduces both the switch node capacitance and switching loss. The paper also presents a scalable transistor model, which is used in simulations of inductive-load hard-switching to determine switching losses and reveal a loss-optimal transistor dimensioning. Furthermore a double pulse setup (600 V-rated GaN HEMTs in a T-type topology) with 2 ground referenced shunts is proposed to determine switching energy of middle and low side transistors simultaneously. To verify the concept of loss-optimal transistor dimensioning in Q2L T-type topology, switching energy was measured at 200 V and 1 A, with the middle transistors area reduced by half compared to high/low side, resulting in a measured reduction from 4.44 µJ to 2.18 µJ (-51%) which is similar to the simulated reduction (2.39 µJ to 1.43 µJ, -40%). This method allows reduction of Q2L switching-loss with optimal transistor area and can be used for a wide range of applications.Item Open Access Einfluss der Abschlussimpedanz von Hochvoltkabeln auf Funkstörgrößen in elektrisch angetriebenen Kraftfahrzeugen(2012) Reuter, Martin; Waible, Manuel; Tenbohlen, Stefan; Köhler, WolfgangIn diesem Beitrag wird die Frage untersucht, welche Auswirkung die Fehlanpassung von Kfz-Hochvoltkabeln auf EMV-Störgrößen in der Komponentenmessung nach CISPR 25 hat.Item Open Access Voltage balancing of a Pi-type multilevel converter for charging electrocaloric capacitors(2025) Bennour, Ines; Mönch, StefanElectrocaloric solid state heat pumps provide a sustainable solution for cooling and heating systems powered by electricity. The electrocaloric effect is a reversible temperature change when the electric field changes in multilayer capacitors made of special dielectric functional materials. It can be transferred into an efficient solid-state heat pump system but requires highly efficient power electronics. High efficiency has already been achieved with multilevel converters with ideal capacitive loads. However, when operating with real, lossy and non-linear electrocaloric components as a load, the internal buffer voltages drift away due to uneven charge distribution across the capacitors. The main idea of this paper is to investigate and solve the unbalancing problem with a purely software-based control technique: Intentional unsymmetrical level-skipping can effectively charge or discharge the buffer voltage. The new balancing operation is experimentally demonstrated with a four-level Pi-type GaN-based multilevel converter, hysteretic current control for zero voltage switching, and an electrocaloric ceramic capacitive load (10µF, 120V, 0.5Hz system frequency). The 4 level balancing operation reduced the power loss by 24% (compared to 2 level), and enabled continuous operation with stabilized buffer voltages.Item Open Access Informationsmodelle mit intelligenter Auswertung für den Digitalen Zwilling(2020) Müller, Manuel; Ashtari Talkhestani, Behrang; Jazdi, Nasser; Rosen, Roland; Wehrstedt, Jan Christoph; Weyrich, MichaelDie zunehmende Komplexität hochautomatisierter Systeme bringt neue Herausforderungen bei der Verwaltung ihrer Modelle entlang des gesamten Lebenszyklus des Systems mit sich - von der Kundenakquise über Engineering und Rekonfiguration bis hin zum Systemrecycling. Der Digitale Zwilling ist ein Konzept, welches über den gesamten Lebenszyklus eines Assets hinweg das Management dieser Modelle sicherstellen kann. Es unterstützt jedoch nicht die automatisierte Modellerweiterung. Hier setzt diese Arbeit an. Die Anreicherung des Digitalen Zwillings um Modellverständnis und KI-Algorithmen zur eigenständigen Modellerweiterung bildet die Grundlager des vorgestellten Konzepts. Über die intelligente Auswertung der Informationsmodelle -angereichert mit aktuellen Prozessdaten- erkennt der Digitale Zwilling, wenn Modelle an ihre Grenzen stoßen. Zwei mögliche Ursachen für diesen Sachverhalt werden genauer betrachtet: (1) es fehlt eine Fähigkeit oder Information (2) der Gültigkeitsbereich des Modells wurde verlassen. Für beide Zustände wird ein Verfahren vorgeschlagen, welches auf Basis kooperativer Information aus dem Wertschöpfungsnetzwerk automatisiert eine Lösung findet. Die Evaluierung des Konzepts anhand eines Szenarios aus der Logistik und aus der Produktion liefert vielversprechende Ergebnisse.Item Open Access Systematic construction of deadlock-free routing for NoC using integer linear programming(2023) Liu, Shuang; Radetzki, MartinNetwork-on-Chip (NoC) presents a promising solution for on-chip communication in highly integrated System-on-Chips (SoCs). This work addresses critical challenges in NoC design, including routing construction, application mapping, and particularly the issue of deadlocks in the widely-used wormhole routing method. In this paper, an Integer Linear Programming (ILP) approach for deadlock-free routing is proposed, applicable to arbitrary network topologies. We systematically analyze deadlock-free routing construction for mesh and torus topologies under uniform random traffic and provide alternative solutions to turn models. In the context of application-specific NoCs, application mapping, and deadlock-free routing are integrated within a single ILP. Through evaluation with several benchmark applications, it is demonstrated that the ILP method consistently delivers optimal solutions and could obtain better results than various heuristic methods within an acceptable time. Fault tolerance is also explored and existing techniques are incorporated into the ILP approach. As an illustrative example, application mapping and a 1-link-fault-tolerant deadlock-free routing for the MP3 application on a mesh network is performed.Item Open Access New mitigation methods for transient overvoltages in gas insulated substations(2012) Burow, Simon; Riechert, Uwe; Köhler, Wolfgang; Tenbohlen, StefanDuring switching of disconnectors (DS) in GIS a varying number of pre-strikes and re-strikes occur. Due to the very short duration of the voltage collapse, traveling surges are generated in the busbar duct. These very fast transient overvoltages (VFTO) can become the limiting dielectric stress which defines the dimensions at UHV voltage levels. The decision shall be based on the maximum VFTO peak value that occurs with reference to the rated lightning impulse withstand voltage (LIWV) of the equipment. If the maximum VFTO is higher than the LIWV, it is necessary to consider the VFTO level as dimensioning criteria or to suppress VFTO by suitable measures. The main challenges are the reduction in VFTO amplitudes and finally the reduction of the effects of VFTO on the equipment. For the different sources of VFTO and for the different equipment different mitigation methods are known. The damping of VFTO by integration of a damping resistor is a well proven technology. The way to overcome the drawback of such unwieldy designs is to use other internal damping measures. Several methods have been proposed and examined in the past, such as ferrite material or high frequency (RF) resonators. The VFTO damping solution utilizing ferrite rings has been analysed and tested and will be described here. The measurements show that a damping effect can be achieved, but with an important drawback: the magnetic material goes easily into saturation, which complicates the design and reduces its general applicability and robustness. A new approach for damping is to implement compact electromagnetic high-frequency resonators with low quality factor specially designed to cover a wider frequency range. The novelty of this idea is not only to design the resonators but also to dissipate the VFTO energy. The VFTO damping effect of the developed RF resonator tuned to the dominant harmonic component was confirmed by experiments. Rings of a nanocrystalline alloy placed around the GIS conductor were also investigated. Depending on number, material and size of the rings a good mitigation could be achieved.Item Open Access A 2.3 mW 1 MHz digitally isolated hysteretic current control circuit with shunt-based current sense amplifier(2025) Öhrlein, Julius; Söllner, Adrian; Bennour, Ines; Mönch, StefanHysteretic current control (HCC) of the inductor current in switched-mode power converters enables fast and direct control and requires high bandwidth isolated current sensing. However, analog-isolated Hall or TMR sensors consume a high power (typ.: 15 mW at 1.5 MHz). This work presents a digitally isolated HCC circuit using a shunt resistor, differential amplifier and two comparators with selectable reference values. A 1 MHz bandwidth HCC circuit was built and applied to a half-bridge converter, consuming only 2.3 mW during operation (from which 1 mW are from an isolated 5 V power supply).Item Open Access A review of electromagnetic simulation and modelling approaches for the research on axial flux synchronous machines(2024) Schäfer, Adrian; Pecha, Urs; Parspour, Nejila; Kampker, Achim; Born, Henrik; Hartmann, Sebastian; Franke, Jörg; Baader, Marcel; Hahn, RomanExtensive electromagnetic (EMAG) studies are necessary to fully realize the potential of axial flux machines (AFMs). However, the disc-shaped air gap and the complex three-dimensional path of magnetic flux pose challenges in modelling AFMs compared to conventional radial flux machines. This study reviews current research on EMAG modelling and simulation of AFMs, highlighting the need for tools that address AFM-specific effects. Existing approaches are analysed based on the requirements composed by fundamental objectives of EMAG simulations and AFM-specific effects, revealing limitations in flexibility and the ability to capture emerging trends in the field of AFMs. While computationally expensive 3D finite element analysis (FEA) offers comprehensive flexibility in EMAG modelling, it lacks efficiency to carry out extensive studies on such trends. Therefore, there is a need to either further accelerate 3D FEA or to increase the flexibility of existing alternatives to facilitate and thereby promote research in the field of AFM and other 3D flux machines. While the integration of some production-specific effects, such as manufacturing tolerances, already is investigated for EMAG simulations of AFMs the future research on the early estimation of manufacturability based on EMAG simulations is crucial for evaluating designs and anticipating manufacturing influences.Item Open Access Slope compensated hysteretic current control of 800 V 4-level GaN/SiC ANPC converter(2025) Bennour, Ines; Söllner, Adrian; Vijayakumar, Karthikeyan; Basler, Michael; Grieshaber, Daniel; Reiner, Richard; Mönch, StefanThis work presents hysteretic current control (HCC) of a 4-level active neutral point clamped (ANPC) 650 V GaN and 1200 V SiC converter for up to 800 V operation. Compared to PWM control, HCC enables a direct cycle-by-cycle current control, useful for dynamic charging applications such as required for driving dielectric smart materials. HCC suffers from the limited current sensor bandwidth, which is addressed in this work by a software-based cycle-by-cycle peak/valley current compensation method. From measured high-side and low-side switching durations, a peak/valley correction correction is calculated and applied to the next cycle. At 150 V, for 2-level operation, slope compensation increased the efficiency from 99.13% to 99.19%. 4-level operation (cyclic charging/discharging a 50 µF load at 1 Hz) with slope compensation resulted in 99.57% peak efficiency at 400 V, and over 99.4% efficiency over a wide input voltage range of 100-800 V.