20005 research outputs found
Sort by
A graded elastic modulus concept to eliminate stress or strain energy density singularity at sharp notches and cracks, with consequent elimination of size-scale effect on strength
No full textIt has been recently suggested by the author that in the classical problem of a sharp wedge or crack loaded in plane (mode I and/or mode II), the stress singularity can be removed by grading the elastic properties of the underlying material from the notch tip by using a power law, E∼rβ. While the treatment is extended to the case of mode III (antiplane shear) which permits closed form results, we also discuss two ways to deal with the likely effect of material's grading on strength. In one, already explored in the previous paper, the strength is a power law of the modulus, and we suggest an “optimal” design by keeping the dominant stress constantly equal to the strength. In a second method, we propose to cancel the singularity in the strain energy density, which requires a much stronger grading, and we also possibly take into account that the critical strain energy density is a power law of the modulus. Noticing that only in the presence of a singularity a length scale can be defined experimentally by testing a very large notch and a very small one, according to the Theory of Critical Distances (TCD), the effect of cancelling singularity also implies independence on size/scale and constant strength. It is concluded that the technique is much more powerful than drilling a hole or rounding the tip of the notch/crack. Moreover, if a “smart” material could be designed to damage itself as to reduce its modulus when near a high stress concentration according to our prescriptions, it would naturally self-heal, opening up interesting applications
AI and responsibility : no gap, but abundance
No full textThe best-performing AI systems, such as deep neural networks, tend to be the ones that are most difficult to control and understand. For this reason, scholars worry that the use of AI would lead to so-called responsibility gaps, that is, situations in which no one is morally responsible for the harm caused by AI, because no one satisfies the so-called control condition and epistemic condition of moral responsibility. In this article, I acknowledge that there is a significant challenge around responsibility and AI. Yet I don't think that this challenge is best captured in terms of a responsibility gap. Instead, I argue for the opposite view, namely that there is responsibility abundance, that is, a situation in which numerous agents are responsible for the harm caused by AI, and that the challenge comes from the difficulties of dealing with such abundance in practice. I conclude by arguing that reframing the challenge in this way offers distinct dialectic and theoretical advantages, promising to help overcome some obstacles in the current debate surrounding ‘responsibility gaps’
A polynomial chaos approach to stochastic LQ optimal control: error bounds and infinite-horizon results
No full textThe stochastic linear–quadratic regulator problem subject to Gaussian disturbances is well known and usually addressed via a moment-based reformulation. Here, we leverage polynomial chaos expansions, which model random variables via series expansions in a suitable L2 probability space, to tackle the non-Gaussian case. We present the optimal solutions for finite and infinite horizons and we analyze the infinite-horizon asymptotics. We show that the limit of the optimal state-input trajectory is the unique solution to a corresponding stochastic stationary optimization problem in the sense of probability measures. Moreover, we provide a constructive error analysis for finite-dimensional polynomial chaos approximations of the optimal solutions and of the optimal stationary pair in non-Gaussian settings. A numerical example illustrates our findings
Chitin analysis in insect-based feed ingredients and mixed feed: development of a cost-effective and practical method
No full textInsects are used as an alternative sustainable, protein-rich ingredient in fish, pet, pig and poultry diets. The significant difference between insect meals and common protein sources is the content of chitin. The nitrogen contained in chitin, which makes up 6.89% of the chitin mass, is detected as crude protein in the analysis and, therefore, deludes the crude protein content in a higher range. In this work, we developed a chitin analysis method that does not require expensive and specialized equipment within insect production and processing industries. The method is based on classical chemical methods such as crude fibre and nitrogen content, making it easily implementable within existing feed analysis. In the process of method validation, a recovery rate of over 95% for chitin in the presence of protein and a standard deviation of 10% at concentrations as low as 2% is possible. The method was used to determine the chitin content in various products derived from insect breeding and processing. The chitin content was determined in four insect species (Hermetia Illucens; Tenebrio molitor; Acheta domesticus; Bombyx mori) and different developmental stages of the yellow mealworm (T. molitor), including larvae, pupae and beetles, as well as in commercial pet food. These results also allow for an estimation of the insect protein content, provided that the raw material is known
Optimizing temperature and pressure in PEM electrolyzers: A model-based approach to enhanced efficiency in integrated energy systems
No full textHydrogen stands as a promising energy carrier within the ongoing energy supply transformation, yet its production via electrolyzers remains prohibitively costly. To address this challenge, this paper proposes an advanced equation-oriented process model for a PEM (Polymer-Electrolyte-Membrane) electrolysis system, including the electrolyzer and downstream hydrogen compression, aimed at optimizing the interaction of its operating parameters (i.e., current density, temperature, pressure). Initially, the model is utilized to analyze the isolated performance of the electrolysis system through operational flowsheet optimizations, followed by its integration into a broader energy system for operational planning optimization. The study reveals several key findings: optimizing operational parameters, rather than using fixed values at the maximum, improves peak system efficiency by approximately 5 %pt. and shifts this peak to lower current densities, thus expanding the range of high-efficiency operation. Each current density has an optimal pair of temperature and pressure, with maximum temperatures only advantageous at loads above 40%, while maximum operating pressure is suboptimal across the entire load range. The analysis indicates that incorporating operating parameter optimization within the operational planning of the electrolysis system reduces energy consumption by 4% and operating costs by 7% in the evaluated energy system. Additionally, the study distinguishes between optimizing the electrolyzer's operating parameters for maximizing its own efficiency and for system efficiency (i.e., including hydrogen compression). It demonstrates that maximum system efficiency is achievable only when the electrolyzer considers hydrogen compression in its operation mode, accepting some efficiency losses individually but yielding greater efficiency gains in the context of hydrogen compression. In summary, the findings of this paper suggest that continuously operating a PEM electrolyzer at maximum temperature and pressure may not be the most efficient approach. Instead, dynamic adjustments based on current density improve operational efficiency, thereby reducing electricity consumption and operating costs. Evaluating the electrolyzer within the broader energy system context – and accepting minor efficiency losses at the electrolyzer level – can yield significant overall benefits and savings. These results underscore the importance of comprehensive, context-aware strategies in advancing cost-effective green hydrogen production
Transient wave propagation in a 1-D gradient model with material nonlinearity
No full textA novel nonlinear 1-D gradient model has been previously proposed by the authors, combining (i) the higher-order gradient terms that capture the influence of material micro-structure and (ii) a nonlinear softening material behavior through the use of a hyperbolic constitutive model. While the previous study focused on the existence and properties of solitary-type waves, the current study focuses on the characteristics of the transient wave propagation in the proposed model. Findings show that as nonlinearity increases, the bulk of the wave slows down, and its shape becomes more distorted in comparison to the response of the linear system. The energy analysis reveals that, unlike the linear system, the nonlinear one continuously exchanges energy, in which the kinetic energy decreases over time while the potential one increases. Furthermore, the spectral (wavenumber) energy density of the nonlinear-elastic system presents peaks at large wavenumbers. However, these are eliminated when a small amount of linear viscous damping is added indicating that they are not physically relevant. A notable feature that persists despite the presence of damping is the formation of small-amplitude waves traveling in the opposite direction to the main wave. Generalized continua, like gradient elasticity models, miss the small energy scatter by the micro-structure. This study shows that adding material nonlinearity to a homogeneous generalized continuum can capture reverse energy propagation, though at much smaller magnitudes than the main wave. These findings shed light on the characteristics of the transient wave propagation predicted by the proposed nonlinear 1-D gradient model and its applicability in, for example, predicting the seismic site response
Machine learning approaches for intentional materials engineering
No full textThe development of nanoporous metals and metallic composites through dealloying processes presents significant opportunities in materials engineering. However, designing multicomponent precursor alloys and establishing corresponding processing methods that yield predictable compositions and nanostructures remain a complex challenge. This article explores how machine learning (ML)-augmented computational and experimental methodologies can tackle these challenges by predicting precursor alloy compositions, final nanoporous structures, and mechanical properties, while integrating ML-enabled autonomous experimentation for material design and quantification. We highlight recent advancements in applying ML to nanostructured materials design via dealloying and discuss how techniques from other nanomaterial designs can be adapted for improved control over morphological and compositional outcomes in nanoporous and nanocomposite materials. Furthermore, we explore the role of ML in autonomous synchrotron x-ray experimentation, enabling real-time feedback between modeling and experimental setups. ML-driven approaches to microstructure characterization and mechanical property prediction are also examined, with a focus on modeling and advanced imaging techniques such as three-dimensional nanotomography. Finally, this article outlines future directions for ML-enhanced materials science, emphasizing the exploration of high-dimensional parameter spaces and the incorporation of materials kinetics into processing and property evaluation, ultimately advancing the design of nanoporous structures and materials science
Variety and costs - the effects of product variety on costs and costing system accuracy
No full textFollowing the microeconomic objective of profit maximization, firms offer their products on heterogeneous markets characterized by individual customer needs, different legal restrictions, competitor products, or cultural factors. Limited in influencing the external complexity, as it is outside their control, firms can decide how to respond by offering a specific product variety. While a large product variety has the advantage of addressing customer needs more precisely, it comes with downsides for firms, such as increased costs, higher lead times, poorer quality, or higher errors in product costing systems. Although research investigated these effects, existing studies are either on a conceptual level or investigate a single cost effect in isolation. Therefore, the latest studies call for operationalizing the various and complicated cause-effect relationships between internal complexity and its economic consequences. Following these calls, this work investigates the effects of internal complexity on total costs and product costing system accuracy on an operational level. Numerical experiments are chosen as a research method as they allow for creating observations that are difficult to gain in practice, such as the product family design or full information environments. A simulation model for creating product family designs, estimating costs, and combining those designs with various product costing systems is introduced. This model comes with a set of validated measures, operationalizing internal complexity. Large-scale numerical experiments reveal the moderating effect of component commonality and overdesign on costs. Overdesign and, as a result, the increased component commonali-ty are levers to offer product variety cost-efficiently. However, too much overdesign increases costs as the increased economies of scale cannot compensate for additional material costs. Therefore, overdesign and costs form a U-shaped curve where the apex indicates the optimal degree of overdesign (component commonality). This work further finds that the cause-effect relationships are much more complicated, as suggested by conceptual studies. As a result, local minima exist along the U-shaped cost curve. A second experiment investigates the effects of internal complexity on product costing accu-racy. It is observed that component commonality is not only a lever to reduce costs but also a lever to increase costing system accuracy since it leads to more homogenous resource consumption patterns. A product-level analysis highlights characteristics of products with highly biased product costs under different product costing systems. This work is placed between engineering design and cost accounting, aiming to strengthen interdisciplinary research. By using numerical experiments, empirical observed cause-effect relationships are generalized. This work is a further step toward an increased understanding of design decisions’ economic consequences, which is crucial for decision-making within firms. In practice, these consequences are difficult to observe since cost assignment and cost occurring differ, and the cost advantages of increased commonality mainly affect the indirect costs.Dem betriebswirtschaftlichen Ziel der Gewinnmaximierung folgend, bieten Unternehmen ihre Produkte auf heterogenen Märkten an, die durch individuelle Kundenbedürfnisse, unterschiedliche rechtliche Beschränkungen, Konkurrenzprodukte oder kulturelle Faktoren gekennzeichnet sind. Daraus ergibt sich eine externe Komplexität, die Unternehmen nur begrenzt beeinflussen können. Sehr wohl können diese aber entscheiden, wie sie darauf reagieren, indem sie eine bestimmte Produktvielfalt anbieten. Eine große Produktvielfalt hat zwar den Vorteil, dass sie den Kundenbedürfnissen besser gerecht wird, bringt aber auch Nachteile für die Unternehmen mit sich, wie z. B. höhere Kosten, längere Lieferzeiten, schlechtere Qualität , mehr Fehler in den Produktkostenrechnungssystemen und resultiert in einer internen Komplexität. Obwohl diese Auswirkungen in der Forschung untersucht wurden, beschränken sich die vorhandenen Studien entweder auf eine konzeptionelle Ebene oder untersuchen isoliert einen einzelnen Kosteneffekt. Neuesten Studien fordern daher, die verschiedenen und komplizierten Ursache-Wirkungs-Beziehungen zwischen interner Komplexität und ihren wirtschaftlichen Folgen zu operationalisieren. Diese Arbeit untersucht die Auswirkungen der internen Komplexität auf die Gesamtkosten sowie die Genauigkeit von Produktkostenrechnungssystemen auf operationaler Ebene. Dazu werden numerische Experimente als Forschungsmethode gewählt. Diese ermöglichen es Szenarien zu untersuchen, die so in der Praxis nur schwer zu beobachten wären - wie z.B. verschiedene Produktfamiliendesigns oder Produktkostenrechnungssysteme unter vollständigen Informationen. Im Rahmen der Arbeit wird ein Simulationsmodell zur Generierung von Produktfamiliendesigns, zur Simulation der Kosten sowie zur Kombination der Designs mit verschiedenen Produktkostenrechnungssystemen entwickelt. Das Modell enthält darüber hinaus validierte Metriken zur Operationalisierung der interne Komplexität. Die Experimente zeigen den moderierenden Effekt von Komponentenwiederverwendung und Überdimensionierung auf die Kosten. Überdimensionierung und Wiederverwendung weisen einen U-förmigen Kostenverlauf auf, deren Scheitelpunkt ein Optimum markiert. So ist eine geringe Überdimensionierung zwar durch geringe Materialkosten gekennzeichnet, nutzt allerdings nicht Skaleneffekte aus, die weitere Kostensenkungspotentiale beinhalten. Diese Arbeit zeigt außerdem, dass die Ursache-Wirkungs-Beziehungen viel komplexer sind, als es bisherige Studien nahelegen. Infolgedessen gibt es lokale Minima entlang der U-förmigen Kostenkurve. In einem zweiten Experiment werden die Auswirkungen der internen Komplexität auf die Genauigkeit von Produktkostenrechnungssystemen untersucht. Es wird gezeigt, dass Komponentenwiederverwendung nicht nur ein Hebel zur Kostensenkung sondern auch ein Hebel zur Erhöhung der Genauigkeit in der Produktkostenrechnung ist, da sie zu homogeneren Ressourcenverbrauchsmustern führt. Eine Analyse auf Produktebene hebt Merkmale von Produkten mit stark verzerrten Produktkosten unter verschiedenen Produktkalkulationssystemen hervor. Diese Arbeit positioniert sich an der Schnittstelle zwischen Produktentwicklung und Kostenrechnung und zielt darauf ab, die interdisziplinäre Forschung zu stärken. Durch den Einsatz von numerischen Experimenten werden empirisch beobachtete Ursache-Wirkungs-Beziehungen generalisiert. Damit ist diese Arbeit ein weiterer Schritt zu einem besseren Verständnis der wirtschaftlichen Konsequenzen von Entscheidungen in frühen Phasen der Produktentstehung. Die Folgen gerade dieser Entscheidungen sind in der Praxis bisher nur schwer zu bewerten. Ein Grund ist, dass Kostenfestlegung und Kostenanfall eine zeitliche Differenz aufweisen. Ein weiterer Grund ist, dass sich Einsparungen durch Komponentenwiederverwendung vorrangig in den Gemeinkosten wiederfinden
Distortion Minimization in the Machining of AM-Structures
No full textDie additive Fertigung bietet durch ihr Potenzial der endkonturnahen Bauweise und der damit verbundenen Ressourcenschonung zahlreiche Vorteile, bringt jedoch auch Herausforderungen wie Eigenspannungen mit sich. Diese führen in der spanenden Bearbeitung zu Verzug, Lage- und Maßabweichungen und damit zu Nacharbeit oder Ausschuss. Um dem entgegenzuwirken, werden in diesem Beitrag verschiedene Frässtrategien verglichen, um aufzuzeigen, dass sich durch eine geeignete Prozessauslegung die resultierenden Fehler um mehr als 90 Prozent reduzieren lassen.Additive manufacturing offers numerous advantages due to its near-net-shape design and the associated resources savings, but also poses challenges such as residual stresses. These lead to distortion, positional- and dimensional-errors, resulting in scrap or further post-processing. To counteract this, the article compares various alternating milling strategies with conventional approaches to show that the errors can be reduced by over 90 % with a suitable process design
Exploring the influence of PBF-LB/M process parameters in multi-material AM: a single-track study on CuCrZr deposited onto IN718 substrate
No full textIn the context of multi-material additive manufacturing (MMAM), Cu-based alloys present inherent chemical and physical properties mismatch with other alloys that often lead to cracking, dimensional mismatch and poor bond performance at the interface. This, combined with the lack of optimized laser parameters for the desired arrangement of materials in the 3D-space present a challenge to achieve functionally distinct regions. IN718, which has been used in the past together with Cu-based alloys, benefits from great chemical affinity of Ni with Cu and, and contrary to 316L, has been rarely explored with Cu alloys in PBF-LB/M. The present work focuses on the feasibility study of printing CuCrZr onto IN718 substrate and the influence of PBF-LB/M process parameters on the morphology and the composition of the meltpool. CuCrZr single tracks were printed onto IN718 buildplate for different set of process parameters including laser power, scan speed and layer thickness