Stiemer, Marcus

Maintaining compliance to electromagnetic compatibility (EMC) standards becomes an increasing challenge for the automotive industry in the course of the ongoing automation of vehicles. Novel extended design procedures and test standards are required to ensure safety of automated driving functions, particularly in an adverse electromagnetic (EM) environment. In order to keep the number of tests within a reasonable and practical limit, an evaluation framework based on virtual design methods and information drawn from legacy experiments and simulations can support the automotive industry. With this digital framework appropriate technological solutions can be identified during the pre-compliance phase and efficient experimental designs can be generated to ensure EMC compliance. Furthermore, such a framework paves the way for digital EMC twins of automated vehicles (AVs) considering the complex interrelations of AV’s (sub-)systems to accurately predict the behaviour of new AV functions in various EM environments. To this purpose, a cross-domain platform is being developed in this work as the backbone of such a virtual framework. It supports the handling, storage and processing of various datasets from EMC test campaigns, including (intentional) electromagnetic interference ((I)EMI) tests, as well as simulations of automotive devices-under-test (DUTs). The platform allows for the establishment of interconnections between various data sources and deeper analyses based on artificial intelligence (AI) methods to deduce EMC information for new developments, whilst maintaining traceability.

Increasing amounts of data and simulations in scientific areas enforce the need of improved software performance. The maintaining scientific staff is often not primarily trained for this purpose or lacks personnel and time to address software performance issues. A particular aim of the dtec.bw-funded project hpc.bw is to tackle some of these shortcomings. A pillar of the hpc.bw agenda is the offer of a low-threshold consultancy and development support focused on performance engineering. This paper provides an insight on our related activities. We illustrate the structure of our annual calls for short-term performance engineering projects, we outline our results at the example of the performance engineering project “benEFIT - Numerical simulation of non-destructive testing in concrete”, and we draw a first conclusion on the current procedure.

Hohe Anforderungen an die elektronischen Komponenten von Systemen der medizinischen Bildgebung können im Rahmen derer Energieversorgung nur durch den Einsatz von moderner Leistungselektronik erfüllt werden. Mit der Verwendung von Wide-Bandgap-Leistungshalbleitern, z. B. in Stromrichtern von Röntgensystemen, gehen damit schnellere Schaltvorgänge sowie hohe Schaltfrequenzen einher. Dies macht eine Untersuchung der Auswirkungen auf die Elektromagnetische Verträglichkeit (EMV) erforderlich. Das betrifft sowohl die Einhaltung von normativen EMV-Grenzwerten als auch die Wechselwirkung der parasitären Eigenschaften von Komponenten innerhalb eines Systems. Im Forschungsprojekt DiMoLEK werden messtechnische Untersuchungen durchgeführt und es wird ein Ansatz zur Modellierung der parasitären Eigenschaften von passiven Komponenten entwickelt. Die Simulation von Stromrichtern mit anschließender Analyse der Wechselwirkungen parasitärer Eigenschaften ermöglicht die Optimierung der (elektromagnetischen) Resilienz. Dies kann beispielsweise durch eine Anpassung des Designs der Komponenten und der Modulationsverfahren erreicht werden.

In recent years, electric vehicles (EVs) are considered to be a promising way to reduce greenhouse gas emissions from the transportation sector. However, the increasing penetration of EVs into the distribution network (DN) raises serious concerns about the network’s safe and reliable operation. The uncontrolled EV charging with random behavior will lead to volatile load peaks on the distribution transformer. In order to obtain more transformer loading capacity available for integration of further EVs, distributed energy resources (DERs) and related devices, such as heat pumps, the transformer loading must be limited to a certain range. For this reason, an intelligent charging management based on model-free Reinforcement Learning (RL) is proposed in this work. The RL management is able to control the charging power of all EVs connected to the network without previous knowledge about the arriving- and leaving time. The needed information for the RL-agent to perceive the current state of the system is formed with cumulated values such as the total energy requirement and the total charging power demand of all EVs. In this paper, the RL algorithm is trained on real-world energy consumption data for a month and on a reference network, created with selected characteristics of a substation network area in the northeast of Hamburg. Comparing with uncontrolled charging, the simulation results show that the RL-based charging management avoids 99 % of threshold violations regarding transformer loading and results in 1% of EV energy requirement is not satisfied. Through sensitivity analysis regarding the state space representation in the employed RL process, the necessity of providing the state of charge (SOC) or the energy requirements of EV users are proven to improve the charging control performance.

Critical frequencies at which electromagnetic fields couple into an electronic device can be identified by comparing the power dissipation of a loaded electromagnetic reverberation chamber to that of an unloaded one. However, working conditions that ensure the required statistical electromagnetic field distribution may also imply a bad signal-to-noise ratio for the measured dissipated power. The purpose of this work is to discuss a measurement procedure and methods for data processing that, nevertheless, allow for gaining significant information from a power spectral analysis in a reverberation chamber. This comprises methods that are based on a fit of suited parametric models for the power absorption of uni- or multi-modal resonators. Their validity has been proven by simple DUTs with previously known characteristics.