检索结果-内蒙古大学图书馆

Smart data representations: Impact on the accuracy of deep neural networks

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Neumann, Oliver Ludwig, Nicole Turowski, Marian Heidrich, Benedikt Hagenmeyer, Veit Mikut, Ralf Institute for Automation and Applied Informatics Karlsruhe Institute of Technology Hermann-von-Helmholtz-Platz 1 Eggenstein-Leopoldshafen76344 Germany Cluster of Excellence Machine Learning University of Tübingen Maria-von-Linden Str. 6 Tübingen72076 Germany

Deep Neural Networks (DNNs) are able to solve many complex tasks with less engineering effort and better performance. However, these networks often use data for training and evaluation without investigating its representation, i.e. the form of the used data. In the present paper, we analyze the impact of data representations on the performance of DNNs using energy time series forecasting. Based on an overview of exemplary data representations, we select four exemplary data representations and evaluate them using two different DNN architectures and three forecasting horizons on real-world energy time series. The results show that, depending on the forecast horizon, the same data representations can have a positive or negative impact on the accuracy of DNNs. © 2021, CC BY.

关键词： Deep neural networks

Turnpike properties in discrete-time mixed-integer optimal control

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Faulwasser, Timm Murray, Alexander Institute for Energy Systems Energy Efficiency and Energy Economics TU Dortmund University Germany Institute for Automation and Applied Informatics Karlsruhe Institute of Technology Germany

This note discusses properties of parametric discrete-time Mixed-Integer Optimal Control Problems (MIOCPs) as they often arise in mixed-integer NMPC. We argue that in want for a handle on similarity properties of parametric MIOCPs the classical turnpike notion from optimal control is helpful. We provide sufficient turnpike conditions based on a dissipativity notion of MIOCPs, and we show that the turnpike property allows specific and accurate guesses for the integer-valued controls. Moreover, we show how the turnpike property can be used to derive efficient node-weighted branch-and-bound schemes tailored to parametric MIOCPs. We draw upon numerical examples to illustrate our findings. Copyright © 2020, The Authors. All rights reserved.

关键词： Optimal control systems

Cell Segmentation and Tracking using CNN-Based Distance Predictions and a Graph-Based Matching Strategy

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Scherr, Tim Löffler, Katharina Böhland, Moritz Mikut, Ralf Institute for Automation and Applied Informatics Karlsruhe Institute of Technology Eggenstein-Leopoldshafen76344 Germany Institute of Biological and Chemical Systems - Biological Information Processing Karlsruhe Institute of Technology Eggenstein-Leopoldshafen76344 Germany

The accurate segmentation and tracking of cells in microscopy image sequences is an important task in biomedical research, e.g., for studying the development of tissues, organs or entire organisms. However, the segmentation of touching cells in images with a low signal-to-noise-ratio is still a challenging problem. In this paper, we present a method for the segmentation of touching cells in microscopy images. By using a novel representation of cell borders, inspired by distance maps, our method is capable to utilize not only touching cells but also close cells in the training process. Furthermore, this representation is notably robust to annotation errors and shows promising results for the segmentation of microscopy images containing in the training data underrepresented or not included cell types. For the prediction of the proposed neighbor distances, an adapted U-Net convolutional neural network (CNN) with two decoder paths is used. In addition, we adapt a graph-based cell tracking algorithm to evaluate our proposed method on the task of cell tracking. The adapted tracking algorithm includes a movement estimation in the cost function to re-link tracks with missing segmentation masks over short time spans. Our combined tracking by detection method has proven its potential in the IEEE ISBI 2020 Cell Tracking Challenge (http://***/) where we achieved as team KIT-Sch-GE multiple top three rankings including two top performances using a single segmentation model for the diverse data sets. Copyright © 2020, The Authors. All rights reserved.

关键词： Cells

Import options for chemical energy carriers from renewable sources to Germany

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Hampp, Johannes Düren, Michael Brown, Tom Center for International Development and Environmental Research Justus Liebig University Gießen Hesse Germany Department of Digital Transformation in Energy Systems Technische Universität Berlin Berlin Germany Institute for Automation and Applied Informatics Karlsruhe Institute of Technology Baden-Württemberg Germany

Import and export of fossil energy carriers are cornerstones of energy systems world-wide. If energy systems are to become climate neutral and sustainable, fossil carriers need to be substituted with carbon neutral alternatives or electrified if possible. We investigate synthetic chemical energy carriers, hydrogen, methane, methanol, ammonia and Fischer-Tropsch fuels, produced using electricity from Renewable Energy Source (RES) as fossil substitutes. RES potentials are obtained from GIS-analysis and hourly resolved time-series are derived using reanalysis weather data. We model the sourcing of feedstock chemicals, synthesis and transport along nine different Energy Supply Chains to Germany and compare import options for seven locations around the world against each other and with domestically sourced alternatives on the basis of their respective cost per unit of hydrogen and energy delivered. We find that for each type of chemical energy carrier, there is an import option with lower costs compared to domestic production in Germany. No single exporting country or energy carrier has a unique cost advantage, since for each energy carrier and country there are cost-competitive alternatives. This allows exporter and infrastructure decisions to be made based on other criteria than energy and cost. The lowest cost means for importing of energy and hydrogen are by hydrogen pipeline from Denmark, Spain and Western Asia and Northern Africa starting at 36EUR/MWhLHVto 42EUR/MWhLHVor 1.0EUR/kgH2to 1.3EUR/kgH2(in 2050, assuming 5% p.a. capital cost). For complex energy carriers derived from hydrogen like methane, ammonia, methanol or Fischer-Tropsch fuels, imports from Argentina by ship to Germany are lower cost than closer exporters in the European Union or Western Asia and Northern Africa. For meeting hydrogen demand, direct hydrogen imports are more attractive than indirect routes using methane, methanol or ammonia imports and subsequent decomposition to hydrogen because of h

关键词： Methanol

Reflection on boon and bane of water absorbing components in active implants during package testing and operation

学校读者我要写书评

暂无评论

Current Directions in Biomedical Engineering 2018年第1期4卷 319-322页

作者： Koker, Liane Gengenbach, Ulrich Bretthauer, Georg Institute for Automation and Applied Informatics (IAI) Karlsruhe Institute of Technology (KIT) Karlsruhe Germany

For package testing of active implants internal humidity measurements are performed. Thereby, the water absorbency of the measurement components can affect the measurement result, leading to an erroneously prolonged predicted lifetime. Here, the infuence of internal components is discussed using an exemplary setup. It is described how to model and consider the components' absorbencies for lifetime estimation and how beneficial they are during implant operation. © 2018 Liane Koker et al.

关键词： Humidity sensors

An industry 4.0 framework for tooling production using metal additive manufacturing-based first-time-right smart manufacturing system 53

学校读者我要写书评

暂无评论

An industry 4.0 framework for tooling production using metal...

53rd CIRP Conference on Manufacturing Systems, CMS 2020

作者： Moshiri, Mandaná Charles, Amal Elkaseer, Ahmed Scholz, Steffen Mohanty, Sankhya Tosello, Guido Department of Mechanical Engineering Technical University of Denmark Building 427 Produktionstorvet 5 Kgs. LyngbyDK-2800 Denmark Institute for Automation and Applied Informatics Karlsruhe Institute of Technology Eggenstein-Leopoldshafen76344 Germany Faculty of Engineering Port Said University Port Fuad42526 Egypt Karlsruhe Nano Micro Facility Hermann-von-Helmholtz-Platz 1 Eggenstein-Leopoldshafen76344 Germany

This paper presents a concept for an integrated process chain for tooling production based on metal additive manufacturing. The proposed approach aims at taking advantage of a fully digitized production line, describing the main steps for the synergetic integration of the manufacturing assets. The production line entails digital infrastructure that collects and elaborate data from various monitoring sensors to execute corrective actions and continuously optimize the process. This line will bring breakthrough benefits, like flexibility and full traceability. Also challenges, as change management in the industry, skills gap, the requirement of new business models and product re-design have been addressed. © 2020 The Authors.

关键词： Industry 4.0

Modeling and simulation of sector-coupled networks: A gas-power benchmark

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Fokken, E. Mühlpfordt, T. Faulwasser, T. Göttlich, S. Kolb, O. Department of Mathematics University of Mannheim Institute for Automation and Applied Informatics Karlsruhe Institute of Technology Institute for Energy Systems Energy Efficiency and Energy Economics TU Dortmund University

In this contribution, we aim at presenting a gas-to-power benchmark problem that can be used for the simulation of electricity and gas networks in a time-dependent environment. Based on realistic data from the IEEE database and the GasLib suite, we describe the full set up of the underlying equations and motivate the choice of parameters. The simulation results demonstrate the applicability of the proposed approach and also allow for a clear visualization of gas-power conversion. Copyright © 2020, The Authors. All rights reserved.

关键词：

Internet of things solution for non-invasive vital data acquisition: a step towards smart healthcare system 6th

学校读者我要写书评

暂无评论

Internet of things solution for non-invasive vital data acqu...

6th KES International Conference on Sustainable Design and Manufacturing, KES-SDM 2019

作者： Salem, Mahmoud El-Maddah, Islam Youssef, Khaled Mohamed, Hoda Faculty of Engineering Ain Shams University Cairo Egypt Institute for Automation and Applied Informatics Karlsruhe Institute of Technology Karlsruhe Germany Faculty of Navigation Science and Space Technology BSU Beni-Suef Egypt

ISBN: (纸本)9789811392702

The advent and subsequent explosive growth of hi-tech medical industry is one of the main features of the modern era. It is driven by the legitimate plea to apply smart healthcare environment for a broad range of working systems, entailing hospitals, schools, workshops, etc. In particular, smart healthcare offers a unique opportunity to optimise the interaction between the human and the medical system (equipment/tools and doctors) for diagnosis, monitoring and treatment measures, which certainly will decrease infections and disease’s transmissions. However, developing a reliable Internet of Things solution, to facilitate a seamless and flawless data transmission among smart assets, is an indispensable task to fully implement. Accordingly, the motivation for this paper is to develop an IoT technique for a non-invasive healthcare system to enable vast collecting and analysis of human vital data via a computer vision technique in order to reduce infections. In particular, concurrent IoT communication was designed for a mesh of three nodes, entailing a sensing node for heartbeat measurement, a vital data visualising node and a broker node. This was developed to guarantee a real-time visualisation of the vital data of the human/patient and it was enriched by an offline visualisation and online cloud visualisation via the broker node. Furthermore, an embedded software architecture was developed that offers a reliable solution to minimise communication latency and provides real-time response to acquired information. Also, the software design is applicable to easily interface with newly added nodes in the future to increase the scale of IoT mesh. © 2019, Springer Nature Singapore Pte Ltd.

关键词： Data acquisition

A New Communication Concept for Efficient Configuration of Energy Systems Integration Co-Simulation

学校读者我要写书评

暂无评论

A New Communication Concept for Efficient Configuration of E...

International Workshop/Symposium on Distributed Simulation and Real-Time Applications (DS-RT)

作者： Anselm Erdmann Hüseyin K. Çakmak Uwe Kühnapfel Veit Hagenmeyer Institute for Automation and Applied Informatics Karlsruhe Institute of Technology Karlsruhe Germany

By combining several existing simulators, co-simulation approaches are designed for efficient simulator development. Especially for energy systems integration simulations involving different physical domains, co-simulation is the required approach. However, co-simulation setup is complex, time-consuming, and additionally, error-prone for large simulations, especially if visualizations are included. Therefore, for an intensive information exchange between simulators, the present paper introduces a new network-based communication concept including the co-simulation setup procedure. The proposed setup concept enables an efficient configuration for the coupling of distributed simulation modules. The evaluation of the implemented concept on an exemplary power grid simulation with integration of power-to-gas and gas-to-power components and power grid visualization shows the ability for efficient handling of a high amount of exchanged simulation data.

关键词：