504 results
Filters
Settings
Now showing1 - 10 of 504
- Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Approximation of the Basset force in the Maxey-Riley-Gatignol equations via universal differential equations(Institute of Physics (IOP), 2026-10-27); ; ; The Maxey-Riley-Gatignol equations (MaRGE) model the motion of spherical inertial particles in a fluid. They contain the Basset force, an integral term which models history effects due to the formation of wakes and boundary layer effects. This causes the force that acts on a particle to depend on its past trajectory and complicates the numerical solution of MaRGE. Therefore, the Basset force is often neglected, despite substantial evidence that it has both quantitative and qualitative impact on the movement patterns of modelled particles. Using the concept of universal differential equations, we propose an approximation of the history term via neural networks which approximates MaRGE by a system of ordinary differential equations that can be solved with standard numerical solvers like Runge-Kutta methods. In particular, for long-term and high-resolution simulations of trajectories, where the number of required time-steps is large, the UDE approach can reduce time-to-solution.Publicationtype: Journal Article54 1 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Conditional neural field-based reduced order model for dynamic ditching load prediction(American Institute of Physics (AIP), 2026-10-22); ; ; Grid-based neural networks such as convolutional autoencoders are widely used in dimension reduction-based surrogate models for computational fluid dynamics. Recently, the use of coordinate-based approaches like conditional neural fields (CNFs) has emerged. Their independence of the spatial discretization is an advantageous feature and a key motivation for computational fluid dynamic applications. This paper discusses the spatiotemporal prediction of aircraft ditching loads using a CNF approach. The model employs a Fourier feature mapping of input coordinates to capture high frequency content. Based on feature-wise linear modulation, the output is conditioned on a latent vector, which is learned by autodecoding. The model is evaluated using two datasets for the dynamic loads on the fuselage of a German Aerospace Center (DLR) D150 aircraft. The first dataset relates to a single fixed spatial discretization, and the second dataset includes data from different discretizations. On the first dataset, the model is compared to a grid-dependent convolutional autoencoder-based model, and both approaches achieve similar reconstruction accuracy, but the CNF requires significantly fewer parameters. When paired with a long short-term memory network in the latent space, both approaches achieve similar spatiotemporal prediction accuracy. Results for the second dataset demonstrate the ability of the CNF-based approach to reconstruct ditching loads accurately for heterogeneous spatial discretizations. In particular, the model can generalize to discretizations not used for training. This allows for flexible use of training datasets generated for different geometries and/or discretizations and the use of the surrogate model to predict loads for different configurations.Publicationtype: Journal Article16 4 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Tracking in-silico Lagrangian sensors in a lab-scale stirred tank reactor.(Elsevier, 2026-09-28); ; ; ; Lagrangian sensors have shown promise to improve operator awareness of conditions inside a chemical reactor but three-dimensional tracking remains a mostly unsolved challenge. We explore a setup where in-silico sensors, based on a recently proposed real-world design, are tracked using data from an accelerometer and magnetometer available from a built-in inertial measurement unit. Filtering algorithms, using a bespoke dynamical model, are used to process these readings into position estimates. We compare tracking performance of an extended Kalman filter, a particle filter and the unscented Kalman filter implemented in the pykalman library. Our numerical experiments track in-silico particles moving in an analytically given threedimensional vortex as well as in the experimentally measured flow-field of a lab-scale stirred tank reactor. Using the Maxey-Riley-Gatignol equations for the movement of inertial particles as ground-truth, we demonstrate that the extended Kalman filter and the particle filter reconstruct trajectories from noisy synthetic data with mean relative errors below 3%.Publicationtype: Journal Article52 27 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Millimeter-scale Lagrangian sensor spheres for local state monitoring in SMART reactors: towards energy-autonomous operation(American Chemical Society (ACS), 2026-09-18); ; ; ; Flow-following sensor particles are a promising route to obtaining spatially resolved measurements in (bio)chemical reactors. However, their practical deployment is limited by miniaturization, power supply, and packaging. This work presents a millimeter-scale Lagrangian sensor-node platform that integrates an inertial measurement unit (IMU) and a low-power microcontroller unit (MCU) on a high-density interconnect (HDI) printed circuit board (PCB), illustrating the evolution from a first demonstrator (V1) toward a reduced-footprint design (V2) intended for spherical encapsulation. The main contribution is a quantitative system-level feasibility boundary that links miniaturized sensor-node design, duty-cycled energy demand, candidate energy-input routes, and spherical packaging constraints at the targeted millimeter scale. We analyze the baseline power demand of the current duty-cycled implementation, including a cycle-based charge and energy budget, and relate this demand to two experimentally investigated energy-input routes: photovoltaic harvesting under controlled optical illumination and resonant inductive power transfer at 13.56 MHz. The harvesting experiments are performed on dedicated evaluation platforms and are used to quantify candidate energy-input routes. A closed energy-autonomous sensor-in-flow demonstration is not yet claimed. The results show that the platform concept, compact HDI integration, low-power duty cycling, optical data transmission, and a glass-based encapsulation route are technically feasible at the targeted size. At the same time, neither harvesting route yet closes the gap to robust energy-autonomous operation under realistic reactor conditions. Photovoltaic powering appears strongly constrained by alignment, optical access, and geometric coupling. Inductive transfer is more promising but remains highly sensitive to orientation, field distribution, and RF-to-DC conversion efficiency. The presented measurements therefore define both the current feasibility range and the remaining engineering requirements for future energy-autonomous sensor spheres.Publicationtype: Journal Article42 22 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Multilinearization of polynomial models by linear state transformation(Elsevier, 2026-09-05); ; ; Many problems in engineering involve the analysis and model-based control of nonlinear systems. Often a tradeoff exists between the fidelity of the model and the computational complexity of analysis and design algorithms. Multilinear models offer significant advantages in this regard since they can be represented as tensors which enables efficient representation and computation, even in high-dimensional settings. This paper deals with the multilinearization of polynomial models through linear transformation. We are able to provide a comprehensive analysis for the case of two variables. For the higher dimensional case, we propose partial solutions which exploit our results for the two-dimensional case. We perform theoretical runtime analyses of the resulting algorithms which we also illustrate through numerical results.Publicationtype: Journal Article17 7 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Constructions for positional games and applications to domination games(Australian National University, 2026-08-28); ; ; We present constructions regarding the general behaviour of biased positional games, and amongst others show that the outcome of such a game can differ in an arbitrary way depending on which player starts the game, and that fair biased games can behave highly non-monotonic. We construct a gadget that helps to transfer such results to Maker-Breaker domination games, and by this we extend a recent result by Gledel, Iršič, and Klavžar, regarding the length of such games. Additionally, we introduce Waiter-Client dominations games, give tight results when they are played on trees or cycles, and using our transference gadget we show that in general the length of such games can differ arbitrarily from the length of their Maker-Breaker analogue.Publicationtype: Journal Article9 1 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , Resilience against soft faults through adaptivity in spectral deferred correction(Springer, 2026-08-06); ; ; ; As supercomputers grow in hardware complexity, their susceptibility to faults increases, and measures need to be taken to ensure the correctness of results. Some numerical algorithms have certain characteristics that allow them to recover from some types of faults. It has been demonstrated that adaptive Runge–Kutta methods provide resilience against transient faults without adding computational cost. Using recent advances in adaptive step size selection for spectral deferred correction (SDC), an iterative numerical time stepping scheme that can achieve arbitrary order, we show that adaptive SDC can also detect and correct transient faults. Its performance is found to be comparable to that of the dedicated resilience strategy Hot Rod.Publicationtype: Conference Paper42 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , On linearisation, existence and uniqueness of preduals: the isometric caseWe study the problem of existence and uniqueness of isometric Banach preduals of a Banach space. We derive necessary and sufficient conditions for the existence of an isometric Banach predual of a Banach space X. Then we focus on the case that X = F(Ω) is a Banach space of scalar-valued functions on a non-empty set Ω and describe those spaces which admit a special isometric Banach predual, namely a strong isometric Banach linearisation, i.e. there are a Banach space Y , a map δ∶Ω → Y and an isometric isomorphism T∶F(Ω) → Y* such that T(f)○δ = f for all f ∈ F(Ω). Finally, we give necessary and sufficient conditions for Banach spaces F(Ω) with a strong isometric Banach linearisation to have a (strongly) unique isometric Banach predual.Publicationtype: Journal Article17 8 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Generalization of LSTM and CNN autoencoders for anomaly detection across orthogonal and longitudinal turningMachine learning models enable automated process monitoring in manufacturing and support adaptive quality assurance and predictive maintenance systems. Autoencoders have emerged as a promising approach to unsupervised anomaly detection in processes. However, their application to turning processes, particularly their generalizability across different process variants such as varying machine kinematics, cutting geometries, and process variability, remains largely unexplored. Existing studies mainly focus on controlled laboratory environments using single machine tools and standardized parameters, limiting industrial scalability. This work addresses this gap by systematically comparing long short-term memory (LSTM) and convolutional (CNN) based autoencoders developed in collaboration with two research projects. The models are trained and evaluated on two harmonized datasets derived from orthogonal and longitudinal turning processes. The datasets differ with respect to machine tools, kinematics, and cutting tools. Each model is trained on both its native process data and on the respective other dataset to evaluate cross-domain performance. The results on normalized reconstruction errors show that LSTM-based autoencoders yield higher temporal sensitivity, whereas CNN-based autoencoders exhibit greater reconstruction stability but reduced sensitivity to transient dynamics. Moreover, relative reconstruction errors are additionally evaluated in the original physical force units to improve process interpretability. This analysis reveals larger relative deviations for LSTM models due to their heightened sensitivity. Furthermore, the variability of the training dataset is identified as a key factor influencing out-of-distribution generalization performance. The findings demonstrate the generalizability of autoencoder based monitoring approaches across different cutting conditions and provide insights for data-driven and grey-box modeling frameworks in manufacturing.Publicationtype: Journal Article32 7 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , Fourier neural operators for Rayleigh–Bénard convection(Springer, 2026-06); ; ; ; We propose an improved Fourier Neural Operator (FNO) for modeling two-dimensional Rayleigh–Bénard convection by predict-ing time increments instead of full solutions, achieving higher accuracy than a standard FNO baseline. The resulting model is compact (314k parameters, 1.26 MB) and fast (7 ms inference), while maintainingsim-ilaraccuracyasdemonstratedinpreviousbenchmarks.WeshowthatalthoughFNOsgeneralizetofinermeshes,accuracyremainslimitedbytheresolutionofthetrainingdata.Publicationtype: Conference Paper46