Article
Engineering, Electrical & Electronic
Jiarui Lin, Jialei Sun, Linghui Yang, Rao Zhang, Yongjie Ren
Summary: This article presents how the rotary laser surface model can be optimized for large-scale optoelectronic measurement systems and validates the effectiveness of the optimization models through experimental results.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Engineering, Multidisciplinary
Jie Li, Lin Chen, Wenbo Yang, Dun Liu, Xuedong Cao
Summary: This paper proposes a scanning distance measuring interferometer method for accurately measuring the transmitted wavefront of large-aperture flat optics. The feasibility of this method is validated by comparing the results with a phase measuring interferometer.
Article
Chemistry, Medicinal
Sohvi Luukkonen, Erik Meijer, Giovanni A. Tricarico, Johan Hofmans, Pieter F. W. Stouten, Gerard J. P. van Westen, Eelke B. Lenselink
Summary: Protein kinases are a protein family with important roles in complex diseases. They have ATP binding sites that can be targeted to create multitarget drugs. Multitask deep learning models outperform single-task deep learning and tree-based models for protein kinase activity prediction.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2023)
Article
Chemistry, Multidisciplinary
Yijie Liu, Xuexuan Li, Yuliang Zhang, Lin Ge, Yingchun Guan, Zhen Zhang
Summary: This study presents a novel approach, the ultra-large scale stitchless atomic force microscopy (ULSS-AFM), which combines a compliant nano-manipulator (CNM) to enable high-throughput characterization of an area of up to 1 x 1 mm(2). Experimental results demonstrate the effectiveness of the ULSS-AFM in different scanning ranges, modes, resolutions, and frequencies. Compared to conventional AFMs, this approach allows for the characterization of ultra-large scale samples without stitching or bow errors, expanding the scanning area of traditional AFMs by two orders of magnitude and opening up important avenues for cross-scale scientific research and industrial applications in nano- and microscale.
Article
Computer Science, Information Systems
Thiago Guarnieri, Idilio Drago, Italo Cunha, Breno Almeida, Jussara M. Almeida, Alex B. Vieira
Summary: This paper presents an in-depth characterization and modeling of client behavior during the live streaming of the 2018 FIFA World Soccer Cup, based on logs covering more than 60 million streaming sessions. By proposing a simple hierarchical model and employing non-supervised clustering, the study shows that specialized models are more accurate in describing the diversity of client behavior patterns compared to a single general model.
MULTIMEDIA SYSTEMS
(2021)
Article
Geosciences, Multidisciplinary
Taewon Cho, Julianne Chung, Scot M. Miller, Arvind K. Saibaba
Summary: Atmospheric inverse modeling is the process of estimating greenhouse gas fluxes or air pollution emissions at the Earth's surface using observations of these gases in the atmosphere. This article discusses computationally efficient methods for large-scale atmospheric inverse modeling and addresses challenges in computation and practicality. The study develops generalized hybrid projection methods that are efficient, robust, automatic, and flexible. The benefits of these methods are demonstrated with a case study from NASA's OCO-2 satellite.
GEOSCIENTIFIC MODEL DEVELOPMENT
(2022)
Article
Geography, Physical
Ziwei Wang, Sijie Yan, Long Wu, Xiaojian Zhang, BinJiang Chen
Summary: This article introduces a method for addressing the registration problem of noisy and featureless 3D point clouds. The method escapes from local minima and restrains sliding by introducing point-to-point l(p) distance constraints and a weighted enhanced l(p) distance error metric. Experimental results show that the method effectively handles outliers and noisy point clouds.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2022)
Article
Environmental Sciences
Elena Crestani, Matteo Camporese, Enrica Belluco, Abderrezak Bouchedda, Erwan Gloaguen, Paolo Salandin
Summary: Salt-water intrusion is a global problem that is increasingly affecting coastal aquifers. Researchers have conducted experiments using physical and numerical modeling to better predict the evolution of the salt-water wedge and design suitable countermeasures. The laboratory facility designed in this study provides valuable benchmarks for future studies of salt-water intrusion and can be used for evaluating salt-water evolution using electrical resistivity tomography. The agreement between observed data, numerical simulations, and ERT results demonstrates the effectiveness of the laboratory facility.
Article
Computer Science, Information Systems
Boyu Li, Ting Guo, Ruimin Li, Yang Wang, Amir H. Gandomi, Fang Chen
Summary: Intelligent transportation system (ITS) is an important symbol of smart cities, aiming to provide sustainable and efficient services to residents. Railway systems, playing a vital role in ITS, have integrated with multiple IoT devices to monitor real-time inbound passenger flows and ensure pedestrian safety. However, the consolidation of real-time information from IoT sources and accurate estimation of future flow face challenges such as coarse-grained data and dynamic interchanged passengers. To overcome these challenges, a two-stage self-adaptive model is proposed for accurately and timely predicting passenger flow in metropolitan railway systems. The model includes a self-attention-based prediction model and a real-time fine-tuning model that combine offline deep learning and real-time allocation.
IEEE INTERNET OF THINGS JOURNAL
(2023)
Article
Astronomy & Astrophysics
Thomas M. Cross, David M. Benoit, Marco Pignatari, Brad K. Gibson
Summary: This work presents a new approach implemented in the Prometheus code to model synthetic rovibrational spectra for all molecules of astrophysical interest. The study evaluates the accuracy of this method by analyzing four diatomic molecules and shows that the simple model achieves better approximation of real spectra. The results are compared with high-resolution spectral data, revealing a decrease in modeling accuracy for rovibrational transitions away from the band origin, highlighting the need for further adaptation of the theory.
ASTROPHYSICAL JOURNAL
(2022)
Article
Computer Science, Software Engineering
Zeqiang Lai, Kaixuan Wei, Ying Fu, Philipp Hartel, Felix Heide
Summary: This article introduces del-Prox, a domain-specific modeling language and compiler for large-scale optimization problems using differentiable proximal algorithms. del-Prox's core feature is full differentiability, supporting hybrid model- and learning-based solvers that integrate proximal optimization with neural network pipelines. With just a few lines of code, del-Prox can generate performant solvers for various image optimization problems, including end-to-end computational optics, image deraining, and compressive magnetic resonance imaging. It can also be used in completely different domains such as energy system planning, outperforming state-of-the-art CVXPY and commercial Gurobi solvers.
ACM TRANSACTIONS ON GRAPHICS
(2023)
Article
Energy & Fuels
Asif Hamid, Danish Rafiq, Shahkar Ahmad Nahvi, Mohammad Abid Bazaz
Summary: The model order reduction (MOR) enterprise has shown unprecedented applications in power systems by allowing realistic simulations of complex power grids. However, existing methods often approximate poorly and have limited accuracy due to the truncation of higher-order modes and linear projection. This paper proposes a dimensionality reduction framework using machine learning techniques, which learns a low-dimensional nonlinear trial manifold using autoencoder (AE) network and obtains the evolution of reduced dynamics using long short-term memory (LSTM) networks. The resulting reduced order models (ROMs) are compact and significantly more accurate than linear projection-based methods.
SUSTAINABLE ENERGY GRIDS & NETWORKS
(2023)
Article
Green & Sustainable Science & Technology
Asmae Berrada
Summary: This study models and assesses the financial performance of a novel energy storage system called gravity energy storage (GES), and compares it with other large-scale energy storage systems. The results show that GES has good performance metrics and is cost-effective compared to its competitors in terms of energy cost.
Article
Meteorology & Atmospheric Sciences
Pratiman Patel, Sajad Jamshidi, Raghu Nadimpalli, Daniel G. Aliaga, Gerald Mills, Fei Chen, Matthias Demuzere, Dev Niyogi
Summary: This study evaluates the impact of land surface models and urban heterogeneity on air temperature simulated by the Weather Research and Forecasting model during a regional extreme event. The results show that implementing local climate zones significantly improves the temperature simulations, altering the surface energy balance and affecting temperatures beyond the urban regions. The impact of land surface model selection is more significant than the inclusion of local climate zones.
JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES
(2022)
Article
Geochemistry & Geophysics
Jianmei Zhou, Kailiang Lu, Xiu Li, Wentao Liu, Zhipeng Qi, Yanfu Qi
Summary: In this article, a restarting polynomial Krylov method is presented for modeling 3-D large-scale TEM responses. The method utilizes the mimetic finite volume method for spatial discretization and does not require solving large-scale linear equations. It achieves high accuracy and uses limited memory.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Materials Science, Characterization & Testing
B. C. F. Oliveira, A. A. Seibert, V. K. Borges, A. Albertazzi, R. H. Schmitt
Summary: Carbon fibre reinforced plastics (CFRPs) are replacing metals in fields like aerospace for their high mechanical strength and low weight. This study introduces a transfer learning approach using a U-Net neural network for segmenting OLT images of CFRP plates with impact damages. The results demonstrate that U-Net is the most reliable method for assessing defective areas, ensuring higher safety in maintenance tasks.
NONDESTRUCTIVE TESTING AND EVALUATION
(2021)
Article
Biology
Tobias Piotrowski, Oliver Rippel, Andreas Elanzew, Bastian Niessing, Sebastian Stucken, Sven Jung, Niels Koenig, Simone Haupt, Laura Stappert, Oliver Bruestle, Robert Schmitt, Stephan Jonas
Summary: This paper presents a method for fully automating cell state recognition using phase contrast microscopy and deep learning, for in-process control during automated hiPSC cultivation. The algorithm is capable of accurately segmenting important parameters of hiPSC colony formation and discriminating between different classes of cells. It provides localized information about the cell state and enables well-based treatment of the cell culture in automated processes.
COMPUTERS IN BIOLOGY AND MEDICINE
(2021)
Article
Engineering, Chemical
Bastian Niessing, Raphael Kiesel, Laura Herbst, Robert H. Schmitt
Summary: Induced pluripotent stem cells (iPSC) offer a unique perspective for manufacturing cell products for drug development and regenerative medicine. Automated production of iPSC can increase throughput, improve quality, and provide economic advantages compared to manual production. This paper presents the biological and technological basics for automated iPSC production, profitability calculation, and a comparison of profitability between manual and automated production in different scenarios.
Article
Engineering, Multidisciplinary
Kevin Nikolai Kostyszyn, Tobias Claus Brandstaetter, Thomas Vollmer, Robert Schmitt
Summary: ISO 7870-8 standardizes the application of charting techniques for short runs and small mixed batches, highlighting the importance of similar distributions in grouped processes. However, differences in distribution parameters can impact control chart performances. Proposing a statistical test for monitoring grouped processes, simulative testing demonstrates the effectiveness of this method in testing for control chart performances.
QUALITY AND RELIABILITY ENGINEERING INTERNATIONAL
(2021)
Article
Engineering, Chemical
Jelena Ochs, Ferdinand Biermann, Tobias Piotrowski, Frederik Erkens, Bastian Niessing, Laura Herbst, Niels Koenig, Robert H. Schmitt
Summary: Laboratory automation is crucial in biotechnology and plays a key role in personalized therapies, enabling cost efficiency and widespread availability of tailored treatments. StemCellDiscovery is a fully automated robotic laboratory for cultivating human mesenchymal stem cells, providing automated confluence estimation and expansion of cell cultures. Simulation modeling shows that StemCellDiscovery is capable of handling over 95 cell culture plates per day under high-throughput conditions.
Article
Computer Science, Artificial Intelligence
Amon Goeppert, Lea Grahn, Jonas Rachner, Dennis Grunert, Simon Hort, Robert H. Schmitt
Summary: The demand for individualized products drives the development of manufacturing systems towards greater adaptability and flexibility, with digital twins serving as a fully connected digital model to physical and digital assets. Standardization and structured modeling are crucial in the creation and deployment of digital twins, along with communication standards and protocols for data exchange. However, there is a lack of consistent workflow from ontology-driven definition to standardized modeling. This paper aims to design an end-to-end digital twin pipeline and automate the process of establishing communication connections. A line-less assembly system with manual stations and a mobile robot is used as an example to explain the digital twin pipeline transparently.
JOURNAL OF INTELLIGENT MANUFACTURING
(2023)
Article
Engineering, Multidisciplinary
Jonas Grosseheide, Kilian Geiger, Anderson Schmidt, Calvin Buetow, Benjamin Montavon, Robert H. Schmitt
Summary: The reconstruction of surface geometry of x-ray computed tomography (XCT) measurement data is a challenge. This study presents a reliable surface geometry determination method based on previously acquired fringe projection (FP) data, which can capture surface geometry at a sub-voxel resolution.
MEASUREMENT SCIENCE AND TECHNOLOGY
(2022)
Article
Management
Tobias Mueller, Alexander Segin, Christoph Weigand, Robert H. Schmitt
Summary: This study aims to reduce the effort of modeling in determining measurement uncertainty by using feature selection methods. Two feature selection methods were identified based on their stability, universality, and complexity, which can reliably identify relevant and irrelevant influencing quantities for a measurement model.
INTERNATIONAL JOURNAL OF QUALITY & RELIABILITY MANAGEMENT
(2023)
Article
Engineering, Industrial
Robert H. Schmitt, Dominik Wolfschlaeger, Evelina Masliankova, Benjamin Montavon
Summary: This paper presents a framework based on generative deep learning methods for interpreting characteristic features extracted from complex inputs and treating them as metrological quantities. The approach is demonstrated in the field of machine vision and shows potential applications in industrial image processing.
CIRP ANNALS-MANUFACTURING TECHNOLOGY
(2022)
Article
Computer Science, Interdisciplinary Applications
Armin F. Buckhorst, Lea Grahn, Robert H. Schmitt
Summary: The Line-less Mobile Assembly System paradigm (LMAS) provides flexibility for large-scale product production. A suitable control system is needed to connect resources and autonomously configure transient assembly stations. This paper introduces a suitable decentralized multi-agent control system approach.
ROBOTICS AND COMPUTER-INTEGRATED MANUFACTURING
(2022)
Article
Materials Science, Characterization & Testing
N. Grozmani, D. Chupina, B. Montavon, R. H. Schmitt
Summary: Computed tomography (CT) is a technology used for inspecting internal and external features of workpieces simultaneously. However, the presence of beam hardening and scattered radiation artifacts due to different material densities and atomic numbers often leads to inaccuracies in CT measurements of multi-material workpieces. Multispectral CT (MSP CT) measurements address this issue by combining multiple CT volumes performed with different x-ray spectra. This paper focuses on sinogram interpretability in the CT radon space and proposes a CT simulation method to determine the optimal x-ray spectra for artifact reduction.
NONDESTRUCTIVE TESTING AND EVALUATION
(2022)
Article
Physics, Applied
Andreas Ulm, Mirza Tareq Ahmed, Robert Schmitt
Summary: We propose an optical setup based on a spatial light modulator (SLM) for rapid micro structuring such as laser lithography. The system addresses beam shaping and mitigates common issues of SLMs. Separating the zero-order focal plane from the first image plane and applying Fourier filtering improves image quality and achieves a resolution of approximately 11 μm.
JOURNAL OF PHYSICS D-APPLIED PHYSICS
(2023)
Article
Materials Science, Multidisciplinary
Lukas Conrads, Natalie Honne, Andreas Ulm, Andreas Hessler, Robert Schmitt, Matthias Wuttig, Thomas Taubner
Summary: This study demonstrates flexible encoding of different absorption/emission properties within a metasurface. By patterning cm-sized stripe gratings on an adaptable grating absorber metasurface using a commercial direct laser writing setup, the plasmonic phase-change material In3SbTe2 (IST) is locally switched between an amorphous and crystalline state to achieve control over the emissivity. The laser power and IST stripe width can be modified to encode different polarization-sensitive patterns with nearly perfect absorption. The results pave the way for low-cost, large-area, and adaptable metasurfaces with wavelength and polarization-selective perfect absorption for applications such as enhanced thermal detection, infrared camouflage, or encoding anti-counterfeiting symbols.
ADVANCED OPTICAL MATERIALS
(2023)
Article
Multidisciplinary Sciences
Zhipeng Ma, Marco Kemmerling, Daniel Buschmann, Chrismarie Enslin, Daniel Luetticke, Robert H. Schmitt
Summary: Causal inference is a fundamental research topic for discovering cause-effect relationships. Time series data provides a good basis for inferring causal relationships. This publication proposes a new data-driven two-phase multi-split causal ensemble model to combine the strengths of different causality base algorithms.
Proceedings Paper
Computer Science, Theory & Methods
Sebastian Beckschulte, Junjie Liang, Robin Guenther, Robert H. Schmitt
Summary: This paper introduces and tests a system-theoretical model that considers the complex interactions between operational production goals and reactive failure management. The model utilizes five modules to represent all production chains and uses stock and flow diagrams for visualization. Testing on the final simulation model highlights the importance of well-structured failure elimination processes for achieving production goals.
INTERNATIONAL CONFERENCE ON ELECTRICAL, COMPUTER AND ENERGY TECHNOLOGIES (ICECET 2021)
(2021)
