Publication Server of Hochschule Darmstadt
Not a member yet
259 research outputs found
Sort by
University and Practice as Actors in Child Protection
No full textProfessionalität von Sozialarbeiter_innen erfordert die Bezugnahme auf vielfältige Wissensbestände und eine einzelfallbezogene Transformation und Relationierung ebendieser, womit – insbesondere vor dem Hintergrund des Abstraktionsgrades wissenschaftlichen Wissens – komplexe Anforderungen an die (angehenden) Fachkräfte einhergehen. Um den Grundstein für eine so verstandene Professionalität legen zu können, müssen im Rahmen des Studiums Erfahrungs‑, Erprobungs- und Reflexionsräume für diese Transformations- und Relationierungsprozesse bereitgestellt werden. Der Beitrag gibt Einblicke in ein innovatives Lehrkonzept, das diese Anliegen aufgreift, skizziert hier gewonnene Erfahrungen und Erkenntnisse sowie Impulse für eine professionalitätsfördernde Qualifizierung von Sozialarbeiter_innen.The professionalism of social workers requires them to refer to a wide range of knowledge and to transform and relate this knowledge on a case-by-case basis, which—especially against the background of the degree of abstraction of scientific knowledge—places complex demands on (prospective) professionals. In order to be able to lay the foundation for professionalism understood in this way, the degree course must provide spaces for experience, testing and reflection for these transformation and relationalization processes. This article provides insights into an innovative teaching concept that addresses these concerns, outlines the experiences and insights gained here and provides impulses for a qualification of social workers that promotes professionalism
A generalized quantification method for bi-directional laser-induced fluorescence
No full textQuantitative measurements of minor species are essential for understanding flame propagation and emission formation, and for validation of chemical kinetic models. Laser-induced fluorescence-based methods are widely employed due to their ability to selectively excite specific species and achieve high signal-to-noise ratios. However, these techniques are inherently susceptible to collisional quenching, which complicates signal quantification. Bi-directional laser-induced fluorescence (BD-LIF) was proposed decades ago as a promising approach to obtain absolute species concentrations while preserving spatial resolution. Despite its potential, initial measurements showed deviations of 50 – 60 % compared to equilibrium calculations and 1D simulations. We present a generalized quantification strategy for BD-LIF based on the general form of Beer’s law that explicitly accounts for wavenumber-dependent absorption and the resulting spatial evolution of the overlap between laser and absorption line, due to the stronger absorption near the line center. The method is demonstrated by measuring hydroxyl (OH) radicals following excitation in the A–X(1,0) system in the post-flame region of laminar CH4-air flames. The results show very good agreement with simulated OH concentrations, underscoring the robustness of the generalized approach and its potential for broader application in combustion diagnostics
Post-Disaster Building Damage Assessment: Multi-Class Object Detection vs. Object Localization and Classification
No full textHighlights
What are the main findings? Separating building localization from damage classification outperforms multi-class detection on unmanned aerial vehicle (UAV) orthomosaics of post-cyclone Mozambique and provides stronger single-class localization. Transfer learning remains consistently beneficial despite the domain gap between COCO/ImageNet and UAV disaster imagery, indicating that generic mid-level features transfer well to this task.
What are the implications of the main findings?
A modular two-stage pipeline improves robustness and adaptability for real-world deployments, enabling independent optimization, rapid component swaps, and straightforward extension to new damage taxonomies, geographies, or sensing conditions without full retraining. By strengthening localization and dedicating classification, the approach better handles challenging, imbalanced classes (e.g., destroyed buildings) and shows improved robustness on non-Western, underrepresented built environments—facilitating more reliable, scalable post-disaster assessment across diverse global contexts.
Abstract
Natural disasters demand swift and accurate impact assessment, yet traditional field-based methods remain prohibitively slow. While semi-automatic techniques leveraging remote sensing and drone imagery have accelerated evaluations, existing datasets predominantly emphasize Western infrastructure, offering limited representation of African contexts. The EDDA dataset (a Mozambique post-disaster building damage dataset developed under the Efficient Humanitarian Aid Through Intelligent Image Analysis project), addresses this critical gap by capturing rural and urban damage patterns in Mozambique following Cyclone Idai. Despite encouraging early results, significant challenges persist due to task complexity, severe class imbalance, and substantial architectural diversity across regions. Building upon EDDA, this study introduces a two-stage building damage assessment pipeline that decouples localization from classification. We employ lightweight You Only Look Once (YOLO)-based detectors—RTMDet, YOLOv7, and YOLOv8—for building localization, followed by dedicated damage severity classification using state-of-the-art architectures including Compact Convolutional Transformers, EfficientNet, and ResNet. This approach tests whether separating feature extraction tasks—assigning detectors solely to localization and specialized classifiers to damage assessment—yields superior performance compared to multi-class detection models that jointly learn both objectives. Comprehensive evaluation across 640+ model combinations demonstrates that our two-stage pipeline achieves competitive performance (mAP 0.478) with enhanced modularity compared to multi-class detection baselines (mAP 0.455), offering improved robustness across diverse building types and imbalanced damage classes
Harnessing AI for Societal and Technological Challenges: Key Findings from the 2024 Annual Meeting of the Research Center for Applied Computer Science
No full textThe 2024 Annual Meeting of the Applied Computer Science Research Center convened experts to explore the application of artificial intelligence (AI) and data science in healthcare, cybersecurity, urban infrastructure, and online misinformation detection. Key discussions highlighted AI's potential to generate actionable insights, improve clinical outcomes, develop hate speech detection tools, and enhance cybersecurity through advanced threat assessment, also considering ethical aspects. The importance of robust data architectures for effective AI deployment was emphasized, particularly in the absence of traditional datasets. This article provides insights into these topics and points out ongoing and future research
Mechanical Behavior of SLS-Printed Parts and Their Structural Simulation
Full text linkThis study aims to assess the mechanical tensile properties of Polyamide produced via selective laser sintering (SLS). The research focuses on the effects of post-processing, positional dependency, anisotropy, and the repeatability of SLS print jobs on material properties. Understanding this anisotropy is crucial for reliable component simulation. A design-appropriate simulation method is developed. A total of 27 identical specimens were fabricated in various orientations and positions within the build chamber, repeated across three print jobs, alongside standard specimens for different post-processing treatments and tempering durations. The mechanical tensile properties were evaluated through tensile tests and compared with simulation outcomes. A new material modeling concept was formulated in the finite element (FE) program ANSYS, employing an orthotropic approach based on linear elastic initial deformation. The Hill Yield Criterion was utilized to model the transition to the plastic region, characterized by a nonlinear strain hardening curve. The print direction was integrated into the FE simulation mesh via a local material coordinate system. Surface treatment via glass bead blasting resulted in slight increases in mechanical response, while tempering had a minor influence. Significant anisotropy was observed, with only the z-position in the build chamber affecting mechanical properties. Successful mapping of anisotropy in structural simulations was achieved. This research did not address optimization of the printing process, recyclate effects, powder aging, or fatigue. The findings provide a comprehensive analysis of the mechanical behavior of SLS-printed specimens, serving as a foundation for treatment methodologies and simulation strategy development
A Hybrid and Modular Integration Concept for Anomaly Detection in Industrial Control Systems
Full text linkEffective anomaly detection is essential for realizing modern and secure industrial control systems. However, the direct integration of anomaly detection within such a system is complex due to the wide variety of hardware used, different communication protocols, and given industrial requirements. Many components of an industrial control system allow direct integration, while others are designed as closed systems or do not have the required performance. At the same time, the effective usage of available resources and the sustainable use of energy are more important than ever for modern industry. Therefore, in this paper, we present a modular and hybrid concept that enables the integration of efficient and effective anomaly detection while optimising the use of available resources under consideration of industrial requirements. Because of the modular and hybrid properties, many functionalities can be outsourced to the respective devices, and at the same time, additional hardware can be integrated where required. The resulting flexibility allows the seamless integration of complete anomaly detection into existing and legacy systems without the need for expensive centralised or cloud-based solutions. Through a detailed evaluation within an industrial unit, we demonstrate the performance and versatility of our concept
Regenerative Wärmeversorgungsvarianten kommunaler Liegenschaften am Beispiel der Friedrich-Ebert-Schule in Darmstadt
Full text linkDie Bundesregierung hat das Ziel ausgerufen, bis 2045 Treibhausgasneutralität zu erlangen. Diese Vorgabe stellt Kommunen und Energieversorger vor erhebliche Herausforderungen. Neben der elektrischen Energie, die einfach auf erneuerbarer Energieträger umgestellt werden kann, muss in Zukunft auch Wärmeenergie emissionsfrei zur Verfügung gestellt werden. Um dieses Problem anzugehen hat die hessische Landesregierung Ende 2022 das hessische Energiegesetz (HEG) verabschiedet. In diesem werden Kommunen ab 20 000 Einwohnern dazu
verpflichtet eine kommunale Wärmeplanung durchzuführen, die sich in drei Schritten gliedert und an deren Ende ein treibhausgas-neutralen Wärmeplan entstehen soll. Im ersten Schritt soll eine Übersichtsplanung über den aktuellen Stand einzelner Infrastrukturen erfasst werden, sowie Potenziale zu den Themen Energieeffizienz und Nutzung erneuerbarer Wärme ausgewiesen werden. Darauf aufbauend wird für die gesamte Kommune eine Bestands- und Potenzialanalyse durchgeführt. Mithilfe dieser kann im dritten Schritt eine Entwicklung der Wärmestrategie mit verschiedenen Szenarien und konkreten Maßnahmen ausgegeben werden (LEA, 2023).
Die Stadt Darmstadt fällt mit ca. 160 000 Einwohnern unter das HEG. Daher wurde Ende 2023 ein eigener Klimaschutzplan veröffentlicht, der bereits 2035 eine vollständige THG-Neutralität vorsieht. Dieses ambitionierte Ziel erfordert schnelle und weitreichende Planung. Da derzeit ca. 60 % der Wärmeenergie in kommunalen Liegenschaften aus Erdgas bezogen wird, ist hier ein großes Potenzial zur Emissionseinsparung (Wissenschaftsstadt Darmstadt, 2022). In der folgenden Arbeit wird in Zusammenarbeit mit dem Darmstädter Amt für Klimaschutz und Klimaanpassungen eine dieser kommunalen Liegenschaften, die Friedrich-Ebert-Schule, betrachtet. Ziel dabei ist eine Übersicht über den aktuellen Stand der Gebäude- und Energieversorgung
zu erhalten sowie eine technische und wirtschaftliche Vorplanung für eine THGNeutrale
Wärmeversorgung zu entwickeln. Dafür werden im 2. Kapitel vom Energiemanagement
des Eigenbetrieb Immobilienmanagement der Wissenschaftsstadt Darmstadt (IDA) Gegebenheiten der Schule, Grundrisse und der Erdgasverbrauch der gesamten Liegenschafft zur Verfügung gestellt. Mithilfe dieser Grundlagen werden zusätzliche notwendige Kennzahlen ermittelt, um den Energieverbrauch der einzelnen Gebäude zu bestimmen. Darauf aufbauend wird in Kapitel 3 auf das Sanierungspotenzial einzelner Gebäude eingegangen, um ein Reduzierungspotenzial der Schule ausweisen zu können. In Kapitel 4 werden verschiedene Möglichkeiten zur zukünftigen Wärmeversorgung aus technischer Sicht betrachtet, für die in Kapitel 5 die wirtschaftlichen Kenngrößen der einzelnen Szenarien ermittelt werden. Kapitel 6 fasst die Erkenntnisse aus den beiden vorangegangenen Kapiteln zusammen und bewertet diese. In Kapitel 7 wird eine Diskussion über die gewählten Parameter durchgeführt und auf weitere Möglichkeiten der Erweiterungen bzw. Anpassungen gewählter Varianten aus Kapitel 4 und 5 dazulegen. Abschließend wird in Kapitel 8 das Fazit der Arbeit dargestellt
Training humans for synthetic face image detection
Full text linkFake identities created using highly realistic synthetic face images have become increasingly prevalent in recent years, driven by advancements in generative neural networks that are readily accessible online and easy to use. These fake identities can be exploited for malicious purposes, such as spreading misinformation or committing fraud. Given the widespread availability of online content and the ease of generating fake online identities, it is desirable that users are able to distinguish real face images from synthetic ones. Additionally, it is important to explore whether specialized training can enhance the ability of individuals to detect synthetically generated face images. In this work, we address these challenges by designing an online experiment to evaluate human detection capabilities and the impact of training on detecting synthetic face images. As part of the experiments, we recruited 184 participants divided into an experimental group and a control group, where the experimental group underwent a tailored training session halfway through the experiment. The study shows that training may moderately enhance human capabilities to detect synthetic face images. Specifically, it was found that the experimental group generally outperformed the control group after training, primarily due to improved abilities in detecting synthetic face images. However, after training, the experimental group showed increased sensitivity and misclassified also more authentic face images, as compared to the control group
Bewaeltigungstrategien Jugendlicher mit Hauptschulabschluss im Übergang in Ausbildung
Full text linkSpätestens am Eintritt in das Übergangssystem wird für Jugendliche mit einem Hauptschulabschluss die von einer früheren Normalität geprägte Vorstellung eines direkten Übergangs in eine Ausbildung oder weiterführende Schulform nach Beenden der allgemeinbildenden Schule infrage gestellt. Zwei Drittel der Jugendlichen im Übergangssystem verfügen über einen Hauptschulabschluss und erleben deutliche Herausforderungen der Bildungs- und Chancenungleichheit, welche wiederkehrend auf Ebene des Alltags der Jugendlichen bewältigt werden müssen. Das Erleben eigener Selbstwirksamkeit erweist sich für die Jugendlichen dabei über alle Bewältigungsstrategien hinweg als zentral. Eine an diesem Selbstwirksamkeitserleben aufgespannte Typologie sowie Theorie der Bewältigungsstrategien stellt das Kernstück der Arbeit dar. Daraus lassen sich Implikationen hinsichtlich einer ganzheitlichen Betrachtung der Jugendlichen, ihrer Eigenschaften, Fähigkeiten sowie Anerkennungs- und Selbstwirksamkeitserfahrungen für die Soziale Arbeit ableiten. Diese sind auf Einzelfall-, Institutions- und Professionsebene angesiedelt und adressieren institutionelle Strukturen ebenso wie interprofessionelle Zusammenarbeit
The State of Material Circularity for Critical Metals
Full text linkThe demand for critical materials continues to increase as technology continues to develop. However, these material reserves and resources are limited. This poses a risk of future material scarcity. This thesis analysed the current state and the future potential of 11 critical materials, including Beryllium, Borates, Fluorine, Gold, Magnesium, Niobium, Phosphorous, Rhenium, Tantalum, Tin and Tungsten. The methodology used in this thesis follows the research conducted by Lundaev et al. (2023), utilizing the growth S-curve and logistic-bell curve models with historical production, reserves and resources data to project future production trends, estimate the peak production and the depletion of material reserves or resources. The results indicated that some materials are approaching their peak production within a few years and most of them in a few decades. After reaching the peak, material production will decline until its reserves or resources depleted near in the future. To reduce or delay the risk of material depletion, circularity strategies are necessary, including increasing material recycling and substituting with alternatives that are less critical