IEEE
Refine
Year of publication
Institute
- Fachbereich Elektrotechnik und Informationstechnik (44)
- Fachbereich Medizintechnik und Technomathematik (19)
- MASKOR Institut für Mobile Autonome Systeme und Kognitive Robotik (11)
- ECSM European Center for Sustainable Mobility (8)
- Fachbereich Maschinenbau und Mechatronik (7)
- INB - Institut für Nano- und Biotechnologien (7)
- Fachbereich Energietechnik (5)
- Fachbereich Chemie und Biotechnologie (2)
- Fachbereich Luft- und Raumfahrttechnik (2)
- Freshman Institute (1)
Has Fulltext
- no (78)
Document Type
- Article (45)
- Conference Proceeding (32)
- Contribution to a Periodical (1)
Keywords
- Engineering education (2)
- Hot S-parameter (2)
- Training (2)
- 10BASE-T1L (1)
- 3-D printing (1)
- Accuracy (1)
- Agent-based modeling (1)
- Analytical models (1)
- Asset Administration Shell (1)
- Automated driving (1)
Automated driving is now possible in diverse road and traffic conditions. However, there are still situations that automated vehicles cannot handle safely and efficiently. In this case, a Transition of Control (ToC) is necessary so that the driver takes control of the driving. Executing a ToC requires the driver to get full situation awareness of the driving environment. If the driver fails to get back the control in a limited time, a Minimum Risk Maneuver (MRM) is executed to bring the vehicle into a safe state (e.g., decelerating to full stop). The execution of ToCs requires some time and can cause traffic disruption and safety risks that increase if several vehicles execute ToCs/MRMs at similar times and in the same area. This study proposes to use novel C-ITS traffic management measures where the infrastructure exploits V2X communications to assist Connected and Automated Vehicles (CAVs) in the execution of ToCs. The infrastructure can suggest a spatial distribution of ToCs, and inform vehicles of the locations where they could execute a safe stop in case of MRM. This paper reports the first field operational tests that validate the feasibility and quantify the benefits of the proposed infrastructure-assisted ToC and MRM management. The paper also presents the CAV and roadside infrastructure prototypes implemented and used in the trials. The conducted field trials demonstrate that infrastructure-assisted traffic management solutions can reduce safety risks and traffic disruptions.
The work in modern open-pit and underground mines requires the transportation of large amounts of resources between fixed points. The navigation to these fixed points is a repetitive task that can be automated. The challenge in automating the navigation of vehicles commonly used in mines is the systemic properties of such vehicles. Many mining vehicles, such as the one we have used in the research for this paper, use steering systems with an articulated joint bending the vehicle’s drive axis to change its course and a hydraulic drive system to actuate axial drive components or the movements of tippers if available. To address the difficulties of controlling such a vehicle, we present a model-predictive approach for controlling the vehicle. While the control optimisation based on a parallel error minimisation of the predicted state has already been established in the past, we provide insight into the design and implementation of an MPC for an articulated mining vehicle and show the results of real-world experiments in an open-pit mine environment.
This paper addresses the pixel based recognition of 3D objects with bidirectional associative memories. Computational power and memory requirements for this approach are identified and compared to the performance of current computer architectures by benchmarking different processors. It is shown, that the performance of special purpose hardware, like neurocomputers, is between one and two orders of magnitude higher than the performance of mainstream hardware. On the other hand, the calculation of small neural networks is performed more efficiently on mainstream processors. Based on these results a novel concept is developed, which is tailored for the efficient calculation of bidirectional associative memories. The computational efficiency is further enhanced by the application of algorithms and storage techniques which are matched to characteristics of the application at hand.
This paper addresses the pixel based classification of three dimensional objects from arbitrary views. To perform this task a coding strategy, inspired by the biological model of human vision, for pixel data is described. The coding strategy ensures that the input data is invariant against shift, scale and rotation of the object in the input domain. The image data is used as input to a class of self organizing neural networks, the Kohonen-maps or self-organizing feature maps (SOFM). To verify this approach two test sets have been generated: the first set, consisting of artificially generated images, is used to examine the classification properties of the SOFMs; the second test set examines the clustering capabilities of the SOFM when real world image data is applied to the network after it has been preprocessed to be invariant against shift, scale and rotation. It is shown that the clustering capability of the SOFM is strongly dependant on the invariance coding of the images.
Aim of the AXON2 project (Adaptive Expert System for Object Recogniton using Neuml Networks) is the development of an object recognition system (ORS) capable of recognizing isolated 3d objects from arbitrary views. Commonly, classification is based on a single feature extracted from the original image. Here we present an architecture adapted from the Mixtures of Eaqerts algorithm which uses multiple neuml networks to integmte different features. During tmining each neural network specializes in a subset of objects or object views appropriate to the properties of the corresponding feature space. In recognition mode the system dynamically chooses the most relevant features and combines them with maximum eficiency. The remaining less relevant features arz not computed and do therefore not decelerate the-recognition process. Thus, the algorithm is well suited for ml-time applications.
A capacitive electrolyte-insulator-semiconductor (EISCAP) biosensor modified with Tobacco mosaic virus (TMV) particles for the detection of acetoin is presented. The enzyme acetoin reductase (AR) was immobilized on the surface of the EISCAP using TMV particles as nanoscaffolds. The study focused on the optimization of the TMV-assisted AR immobilization on the Ta 2 O 5 -gate EISCAP surface. The TMV-assisted acetoin EISCAPs were electrochemically characterized by means of leakage-current, capacitance-voltage, and constant-capacitance measurements. The TMV-modified transducer surface was studied via scanning electron microscopy.
An interdisciplinary view on humane interfaces for digital shadows in the internet of production
(2022)
Digital shadows play a central role for the next generation industrial internet, also known as Internet of Production (IoP). However, prior research has not considered systematically how human actors interact with digital shadows, shaping their potential for success. To address this research gap, we assembled an interdisciplinary team of authors from diverse areas of human-centered research to propose and discuss design and research recommendations for the implementation of industrial user interfaces for digital shadows, as they are currently conceptualized for the IoP. Based on the four use cases of decision support systems, knowledge sharing in global production networks, human-robot collaboration, and monitoring employee workload, we derive recommendations for interface design and enhancing workers’ capabilities. This analysis is extended by introducing requirements from the higher-level perspectives of governance and organization.
The fourth industrial revolution presents a multitude of challenges for industries, one of which being the increased flexibility required of manufacturing lines as a result of increased consumer demand for individualised products. One solution to tackle this challenge is the digital twin, more specifically the standardised model of a digital twin also known as the asset administration shell. The standardisation of an industry wide communications tool is a critical step in enabling inter-company operations. This paper discusses the current state of asset administration shells, the frameworks used to host them and their problems that need to be addressed. To tackle these issues, we propose an event-based server capable of drastically reducing response times between assets and asset administration shells and a multi-agent system used for the orchestration and deployment of the shells in the field.
Image reconstruction analysis for positron emission tomography with heterostructured scintillators
(2022)
The concept of structure engineering has been proposed for exploring the next generation of radiation detectors with improved performance. A TOF-PET geometry with heterostructured scintillators with a pixel size of 3.0×3.1×15 mm3 was simulated using Monte Carlo. The heterostructures consisted of alternating layers of BGO as a dense material with high stopping power and plastic (EJ232) as a fast light emitter. The detector time resolution was calculated as a function of the deposited and shared energy in both materials on an event-by-event basis. While sensitivity was reduced to 32% for 100 μm thick plastic layers and 52% for 50 μm, the CTR distribution improved to 204±49 ps and 220±41 ps respectively, compared to 276 ps that we considered for bulk BGO. The complex distribution of timing resolutions was accounted for in the reconstruction. We divided the events into three groups based on their CTR and modeled them with different Gaussian TOF kernels. On a NEMA IQ phantom, the heterostructures had better contrast recovery in early iterations. On the other hand, BGO achieved a better contrast to noise ratio (CNR) after the 15th iteration due to the higher sensitivity. The developed simulation and reconstruction methods constitute new tools for evaluating different detector designs with complex time responses.
Upcoming gasoline engines should run with a larger number of fuels beginning from petrol over methanol up to gas by a wide range of compression ratios and a homogeneous charge. In this article, the microwave (MW) spark plug, based on a high-speed frequency hopping system, is introduced as a solution, which can support a nitrogen compression ratio up to 1:39 in a chamber and more. First, an overview of the high-speed frequency hopping MW ignition and operation system as well as the large number of applications are presented. Both gives an understanding of this new base technology for MW plasma generation. Focus of the theoretical part is the explanation of the internal construction of the spark plug, on the achievable of the high voltage generation as well as the high efficiency to hold the plasma. In detail, the development process starting with circuit simulations and ending with the numerical multiphysics field simulations is described. The concept is evaluated with a reference prototype covering the frequency range between 2.40 and 2.48 GHz and working over a large power range from 20 to 200 W. A larger number of different measurements starting by vector hot-S11 measurements and ending by combined working scenarios out of hot temperature, high pressure and charge motion are winding up the article. The limits for the successful pressure tests were given by the pressure chamber. Pressures ranged from 1 to 39 bar and charge motion up to 25 m/s as well as temperatures from 30◦ to 125◦.
The recent amendment to the Ethernet physical layer known as the IEEE 802.3cg specification, allows to connect devices up to a distance of one kilometer and delivers a maximum of 60 watts of power over a twisted pair of wires. This new standard, also known as 10BASE-TIL, promises to overcome the limits of current physical layers used for field devices and bring them a step closer to Ethernet-based applications. The main advantage of 10BASE- TIL is that it can deliver power and data over the same line over a long distance, where traditional solutions (e.g., CAN, IO-Link, HART) fall short and cannot match its 10 Mbps bandwidth. Due to its recentness, IOBASE- TIL is still not integrated into field devices and it has been less than two years since silicon manufacturers released the first Ethernet-PHY chips. In this paper, we present a design proposal on how field devices could be integrated into a IOBASE-TIL smart switch that allows plug-and-play connectivity for sensors and actuators and is compliant with the Industry 4.0 vision. Instead of presenting a new field-level protocol for this work, we have decided to adopt the IO-Link specification which already includes a plug-and-play approach with features such as diagnosis and device configuration. The main objective of this work is to explore how field devices could be integrated into 10BASE-TIL Ethernet, its adaption with a well-known protocol, and its integration with Industry 4.0 technologies.
For typical cases of non-isolated lightning protection systems (LPS) the impulse currents are investigated which may flow through a human body directly touching a structural part of the LPS. Based on a basic LPS model with conventional down-conductors especially the cases of external and internal steel columns and metal façades are considered and compared. Numerical simulations of the line quantities voltages and currents in the time domain are performed with an equivalent circuit of the entire LPS.
As a result it can be stated that by increasing the number of conventional down-conductors and external steel columns the threat for a human being can indeed be reduced, but not down to an acceptable limit. In case of internal steel columns used as natural down-conductors the threat can be reduced sufficiently, depending on the low-resistive connection of the steel columns to the lightning equipotential bonding or the earth termination system, resp. If a metal façade is used the threat for human beings touching is usually very low, if the façade is sufficiently interconnected and multiply connected to the lightning equipotential bonding or the earth termination system, resp.
One central challenge for self-driving cars is a proper path-planning. Once a trajectory has been found, the next challenge is to accurately and safely follow the precalculated path. The model-predictive controller (MPC) is a common approach for the lateral control of autonomous vehicles. The MPC uses a vehicle dynamics model to predict the future states of the vehicle for a given prediction horizon. However, in order to achieve real-time path control, the computational load is usually large, which leads to short prediction horizons. To deal with the computational load, the control algorithm can be parallelized on the graphics processing unit (GPU). In contrast to the widely used stochastic methods, in this paper we propose a deterministic approach based on grid search. Our approach focuses on systematically discovering the search area with different levels of granularity. To achieve this, we split the optimization algorithm into multiple iterations. The best sequence of each iteration is then used as an initial solution to the next iteration. The granularity increases, resulting in smooth and predictable steering angle sequences. We present a novel GPU-based algorithm and show its accuracy and realtime abilities with a number of real-world experiments.
The fourth industrial revolution introduces disruptive technologies to production environments. One of these technologies are multi-agent systems (MASs), where agents virtualize machines. However, the agent's actual performances in production environments can hardly be estimated as most research has been focusing on isolated projects and specific scenarios. We address this gap by implementing a highly connected and configurable reference model with quantifiable key performance indicators (KPIs) for production scheduling and routing in single-piece workflows. Furthermore, we propose an algorithm to optimize the search of extrema in highly connected distributed systems. The benefits, limits, and drawbacks of MASs and their performances are evaluated extensively by event-based simulations against the introduced model, which acts as a benchmark. Even though the performance of the proposed MAS is, on average, slightly lower than the reference system, the increased flexibility allows it to find new solutions and deliver improved factory-planning outcomes. Our MAS shows an emerging behavior by using flexible production techniques to correct errors and compensate for bottlenecks. This increased flexibility offers substantial improvement potential. The general model in this paper allows the transfer of the results to estimate real systems or other models.
In positron emission tomography improving time, energy and spatial detector resolutions and using Compton kinematics introduces the possibility to reconstruct a radioactivity distribution image from scatter coincidences, thereby enhancing image quality. The number of single scattered coincidences alone is in the same order of magnitude as true coincidences. In this work, a compact Compton camera module based on monolithic scintillation material is investigated as a detector ring module. The detector interactions are simulated with Monte Carlo package GATE. The scattering angle inside the tissue is derived from the energy of the scattered photon, which results in a set of possible scattering trajectories or broken line of response. The Compton kinematics collimation reduces the number of solutions. Additionally, the time of flight information helps localize the position of the annihilation. One of the questions of this investigation is related to how the energy, spatial and temporal resolutions help confine the possible annihilation volume. A comparison of currently technically feasible detector resolutions (under laboratory conditions) demonstrates the influence on this annihilation volume and shows that energy and coincidence time resolution have a significant impact. An enhancement of the latter from 400 ps to 100 ps leads to a smaller annihilation volume of around 50%, while a change of the energy resolution in the absorber layer from 12% to 4.5% results in a reduction of 60%. The inclusion of single tissue-scattered data has the potential to increase the sensitivity of a scanner by a factor of 2 to 3 times. The concept can be further optimized and extended for multiple scatter coincidences and subsequently validated by a reconstruction algorithm.
Project work and inter disciplinarity are integral parts of today's engineering work. It is therefore important to incorporate these aspects into the curriculum of academic studies of engineering. At the faculty of Electrical Engineering and Information Technology an interdisciplinary project is part of the bachelor program to address these topics. Since the summer term 2020 most courses changed to online mode during the Covid-19 crisis including the interdisciplinary projects. This online mode introduces additional challenges to the execution of the projects, both for the students as well as for the lecture. The challenges, but also the risks and chances of this kind of project courses are subject of this paper, based on five different interdisciplinary projects
During the Covid-19 pandemic, vocational colleges, universities of applied science and technical universities often had to cancel laboratory sessions requiring students’ attendance. These above of all are of decisive importance in order to give learners an understanding of theory through practical work.This paper is a contribution to the implementation of distance learning for laboratory work applicable for several upper secondary educational facilities. Its aim is to provide a paradigm for hybrid teaching to analyze and control a non-linear system depicted by a tank model. For this reason, we redesign a full series of laboratory sessions on the basis of various challenges. Thus, it is suitable to serve different reference levels of the European Qualifications Framework (EQF).We present problem-based learning through online platforms to compensate the lack of a laboratory learning environment. With a task deduced from their future profession, we give students the opportunity to develop own solutions in self-defined time intervals. A requirements specification provides the framework conditions in terms of time and content for students having to deal with the challenges of the project in a self-organized manner with regard to inhomogeneous previous knowledge. If the concept of Complete Action is introduced in classes before, they will automatically apply it while executing the project.The goal is to combine students’ scientific understanding with a procedural knowledge. We suggest a series of remote laboratory sessions that combine a problem formulation from the subject area of Measurement, Control and Automation Technology with a project assignment that is common in industry by providing extracts from a requirements specification.
This article introduces a new maritime search and rescue system based on S-band illumination harmonic radar (HR). Passive and active tags have been developed and tested attached to life jackets and a rescue boat. This system was able to detect and range the active tags up to a range of 5800 m in tests on the Baltic Sea with an antenna input power of only 100 W. All electronic GHz components of the system, excluding the S-band power amplifier, were custom developed for this purpose. Special attention is given to the performance and conceptual differences between passive and active tags used in the system and integration with a maritime X-band navigation radar is demonstrated.
Industry 4.0 imposes many challenges for manufacturing companies and their employees. Innovative and effective training strategies are required to cope with fast-changing production environments and new manufacturing technologies. Virtual Reality (VR) offers new ways of on-the-job, on-demand, and off-premise training. A novel concept and evaluation system combining Gamification and VR practice for flexible assembly tasks is proposed in this paper and compared to existing works. It is based on directed acyclic graphs and a leveling system. The concept enables a learning speed which is adjustable to the users’ pace and dynamics, while the evaluation system facilitates adaptive work sequences and allows employee-specific task fulfillment. The concept was implemented and analyzed in the Industry 4.0 model factory at FH Aachen for mechanical assembly jobs.
Control engineering theory is hard to grasp for undergraduates during the first semesters, as it deals with the dynamical behavior of systems also in combination with control strategies on an abstract level. Therefore, operational amplifier (OpAmp) processes are reasonable and very effective systems to connect mathematical description with actual system’s behavior. In this paper, we present an experiment for a laboratory session in which an embedded system, driven by a LabVIEW human machine interface (HMI) via USB, controls the analog circuits.With this setup we want to show the possibility of firstly, analyzing a first order process and secondly, designing a P-and PI-controller. Thereby, the theory of control engineering is always applied to the empirical results in order to break down the abstract level for the students.
In collaborative research projects, both researchers and practitioners work together solving business-critical challenges. These projects often deal with ETL processes, in which humans extract information from non-machine-readable documents by hand. AI-based machine learning models can help to solve this problem.
Since machine learning approaches are not deterministic, their quality of output may decrease over time. This fact leads to an overall quality loss of the application which embeds machine learning models. Hence, the software qualities in development and production may differ.
Machine learning models are black boxes. That makes practitioners skeptical and increases the inhibition threshold for early productive use of research prototypes. Continuous monitoring of software quality in production offers an early response capability on quality loss and encourages the use of machine learning approaches. Furthermore, experts have to ensure that they integrate possible new inputs into the model training as quickly as possible.
In this paper, we introduce an architecture pattern with a reference implementation that extends the concept of Metrics Driven Research Collaboration with an automated software quality monitoring in productive use and a possibility to auto-generate new test data coming from processed documents in production.
Through automated monitoring of the software quality and auto-generated test data, this approach ensures that the software quality meets and keeps requested thresholds in productive use, even during further continuous deployment and changing input data.
Safety of subjects during radiofrequency exposure in ultra-high-field magnetic resonance imaging
(2020)
Magnetic resonance imaging (MRI) is one of the most important medical imaging techniques. Since the introduction of MRI in the mid-1980s, there has been a continuous trend toward higher static magnetic fields to obtain i.a. a higher signal-to-noise ratio. The step toward ultra-high-field (UHF) MRI at 7 Tesla and higher, however, creates several challenges regarding the homogeneity of the spin excitation RF transmit field and the RF exposure of the subject. In UHF MRI systems, the wavelength of the RF field is in the range of the diameter of the human body, which can result in inhomogeneous spin excitation and local SAR hotspots. To optimize the homogeneity in a region of interest, UHF MRI systems use parallel transmit systems with multiple transmit antennas and time-dependent modulation of the RF signal in the individual transmit channels. Furthermore, SAR increases with increasing field strength, while the SAR limits remain unchanged. Two different approaches to generate the RF transmit field in UHF systems using antenna arrays close and remote to the body are investigated in this letter. Achievable imaging performance is evaluated compared to typical clinical RF transmit systems at lower field strength. The evaluation has been performed under consideration of RF exposure based on local SAR and tissue temperature. Furthermore, results for thermal dose as an alternative RF exposure metric are presented.
In this paper we present SMART-FACTORY, a setup for a research and teaching facility in industrial robotics that is based on the RoboCup Logistics League. It is driven by the need for developing and applying solutions for digital production. Digitization receives constantly increasing attention in many areas, especially in industry. The common theme is to make things smart by using intelligent computer technology. Especially in the last decade there have been many attempts to improve existing processes in factories, for example, in production logistics, also with deploying cyber-physical systems. An initiative that explores challenges and opportunities for robots in such a setting is the RoboCup Logistics League. Since its foundation in 2012 it is an international effort for research and education in an intra-warehouse logistics scenario. During seven years of competition a lot of knowledge and experience regarding autonomous robots was gained. This knowledge and experience shall provide the basis for further research in challenges of future production. The focus of our SMART-FACTORY is to create a stimulating environment for research on logistics robotics, for teaching activities in computer science and electrical engineering programmes as well as for industrial users to study and explore the feasibility of future technologies. Building on a very successful history in the RoboCup Logistics League we aim to provide stakeholders with a dedicated facility oriented at their individual needs.
The increasing digitalization brings new opportunities but also puts new challenges to modern industrial systems. Software agents are one of the key technologies towards self-optimizing factories and are currently used to address the needs of cyber-physical production systems (CPPS). However their interplay in industrial settings needs to be understood better.This paper focusses on securing a cloud infrastructure for multi-agent systems for industrial sites. An industrial site contains multiple production processes that need to communicate with each other and each physical resource is abstracted with a software agent. This volatile architecture needs to be managed and protected from manipulation. The proposed infrastructure presents a security concept for TCP/IP communication between agents, machines, and external networks. It is based on open-source software and tested on a three-node edge cloud controlling a model-plant.
To meet the challenges of manufacturing smart products, the manufacturing plants have been radically changed to become smart factories underpinned by industry 4.0 technologies. The transformation is assisted by employment of machine learning techniques that can deal with modeling both big or limited data. This manuscript reviews these concepts and present a case study that demonstrates the use of a novel intelligent hybrid algorithms for Industry 4.0 applications with limited data. In particular, an intelligent algorithm is proposed for robust data modeling of nonlinear systems based on input-output data. In our approach, a novel hybrid data-driven combining the Group-Method of Data-Handling and Singular-Value Decomposition is adapted to find an offline deterministic model combined with Pareto multi-objective optimization to overcome the overfitting issue. An Unscented-Kalman-Filter is also incorporated to update the coefficient of the deterministic model and increase its robustness against data uncertainties. The effectiveness of the proposed method is examined on a set of real industrial measurements.
To meet the challenges of manufacturing smart products, the manufacturing plants have been radically changed to become smart factories underpinned by industry 4.0 technologies. The transformation is assisted by employment of machine learning techniques that can deal with modeling both big or limited data. This manuscript reviews these concepts and present a case study that demonstrates the use of a novel intelligent hybrid algorithms for Industry 4.0 applications with limited data. In particular, an intelligent algorithm is proposed for robust data modeling of nonlinear systems based on input-output data. In our approach, a novel hybrid data-driven combining the Group-Method of Data-Handling and Singular-Value Decomposition is adapted to find an offline deterministic model combined with Pareto multi-objective optimization to overcome the overfitting issue. An Unscented-Kalman-Filter is also incorporated to update the coefficient of the deterministic model and increase its robustness against data uncertainties. The effectiveness of the proposed method is examined on a set of real industrial measurements.
Design and Development of a Hot S-Parameter Measurement System for Plasma and Magnetron Applications
(2020)
This paper presents the design, development and calibration procedures of a novel hot S-parameter measurement system for plasma and magnetron applications with power level up to 6 kW. Based on a vector network analyzer, a power amplifier and two directional couplers, the input matching hotS 11 and transmission hotS 21 of the device under test are measured at 2.45 GHz center frequency and 300MHz bandwidth, while the device is driven by the magnetron. This measurement system opens a new horizon to develop many new industrial applications such as microwave plasma jets, dryer systems, dryers and so forth. Furthermore, the developing, controlling and monitoring a 2kW 2.45GHz plasma jet and a dryer system using the measurement system are presented and explained.
In this paper we report on an architecture for a self-driving car that is based on ROS2. Self-driving cars have to take decisions based on their sensory input in real-time, providing high reliability with a strong demand in functional safety. In principle, self-driving cars are robots. However, typical robot software, in general, and the previous version of the Robot Operating System (ROS), in particular, does not always meet these requirements. With the successor ROS2 the situation has changed and it might be considered as a solution for automated and autonomous driving. Existing robotic software based on ROS was not ready for safety critical applications like self-driving cars. We propose an architecture for using ROS2 for a self-driving car that enables safe and reliable real-time behaviour, but keeping the advantages of ROS such as a distributed architecture and standardised message types. First experiments with an automated real passenger car at lower and higher speed-levels show that our approach seems feasible for autonomous driving under the necessary real-time conditions.
With the many achievements of Machine Learning in the past years, it is likely that the sub-area of Deep Learning will continue to deliver major technological breakthroughs [1]. In order to achieve best results, it is important to know the various different Deep Learning frameworks and their respective properties. This paper provides a comparative overview of some of the most popular frameworks. First, the comparison methods and criteria are introduced and described with a focus on computer vision applications: Features and Uses are examined by evaluating papers and articles, Adoption and Popularity is determined by analyzing a data science study. Then, the frameworks TensorFlow, Keras, PyTorch and Caffe are compared based on the previously described criteria to highlight properties and differences. Advantages and disadvantages are compared, enabling researchers and developers to choose a framework according to their specific needs.
Magnetic nanoparticles (MNPs) are used as therapeutic and diagnostic agents for local delivery of heat and image contrast enhancement in diseased tissue. Besides magnetization, the most important parameter that determines their performance for these applications is their magnetic relaxation, which can be affected when MNPs immobilize and agglomerate inside tissues. In this letter, we investigate different MNP agglomeration states for their magnetic relaxation properties under excitation in alternating fields and relate this to their heating efficiency and imaging properties. With focus on magnetic fluid hyperthermia, two different trends in MNP heating efficiency are measured: an increase by up to 23% for agglomerated MNP in suspension and a decrease by up to 28% for mixed states of agglomerated and immobilized MNP, which indicates that immobilization is the dominant effect. The same comparatively moderate effects are obtained for the signal amplitude in magnetic particle spectroscopy.
Thermal and Optical Study on the Frequency Dependence of an Atmospheric Microwave Argon Plasma Jet
(2019)
This paper describes the development of a capacitively coupled high-pressure lamp with input power between 20 and 43 W at 2.45 GHz, using a coaxial line network. Compared with other electrodeless lamp systems, no cavity has to be used and a reduction in the input power is achieved. Therefore, this lamp is an alternative to the halogen incandescent lamp for domestic lighting. To serve the demands of domestic lighting, the filling of the lamp is optimized over all other resulting requirements, such as high efficacy at low induced powers and fast startups. A workflow to develop RF-driven plasma applications is presented, which makes use of the hot S-parameter technique. Descriptions of the fitting process inside a circuit and FEM simulator are given. Results of the combined ignition and operation network from simulations and measurements are compared. An initial prototype is built and measurements of the lamp's lighting properties are presented along with an investigation of the efficacy optimizations using large signal amplitude modulation. With this lamp, an efficacy of 135 lmW -1 is achieved.
High gradient magnetic separation (HGMS) has been established since the early 1970s. A more recent application of these systems is the use in bioprocesses. To integrate the HGMS in a fermentation process, it is necessary to optimize the separation matrix with regard to the magnetic separation characteristics and permeability of the non-magnetizable components of the fermentation broth. As part of the work presented here, a combined fluidic and magnetic force finite element model simulation was created using the software COMSOL Multiphysics and compared with separation experiments. Finally, as optimal lattice orientation of the separation matrix, a transversal rhombohedral arrangement was defined. The high suitability of the new filter matrix has been verified by separation experiments.
The capacitive electrolyte–insulator–semiconductor (EIS) structure is a typical device based on a field-effect sensor platform. With a simple silicon-based structure, EIS have been useful for several sensing applications, especially with incorporation of nanostructured films to modulate the ionic transport and the flat-band potential. In this paper, we report on ion transport and changes in flat-band potential in EIS sensors made with layer-by-layer films containing poly(amidoamine) (PAMAM) dendrimer and single-walled carbon nanotubes (SWNTs) adsorbed on p-Si/SiO 2 /Ta 2 O 5 chips with an Al ohmic contact. The impedance spectra were fitted using an equivalent circuit model, from which we could determine parameters such as the double-layer capacitance. This capacitance decreased with the number of bilayers owing to space charge accumulated at the electrolyte–insulator interface, up to three PAMAM/SWNTs bilayers, after which it stabilized. The charge-transfer resistance was also minimum for three bilayers, thus indicating that this is the ideal architecture for an optimized EIS performance. The understanding of the influence of nanostructures and the fine control of operation parameters pave the way for optimizing the design and performance of new EIS sensors.
This contribution discusses the utilization of RF power in an MRI system with RF mode shimming which enables the superposition of circularly polarized modes of a transmit RF coil array driven by a Butler matrix. Since the required power for the individual modes can vary widely, mode-shimming can result in a significant underutilization of the total available RF power. A variable power combiner (VPC) is proposed to improve the power utilization: it can be realized as a reconfiguration of the MRI transmit system by the inclusion of one additional matrix network which receives the power from all transmit amplifiers at its input ports and provides any desired (combined) power distribution at its output ports by controlling the phase and amplitude of the amplifiers’ input signals. The power distribution at the output ports of the VPC is then fed into the “mode” ports of the coil array Butler matrix in order to superimpose the spatial modes at the highest achievable power utilization. The VPC configuration is compared to the standard configuration of the transmit chain of our MRI system with 8 transmit channels and 16 coils. In realistic scenarios, improved power utilization was achieved from 17% to 60% and from 14% to 55% for an elliptical phantom and a region of interest in the abdomen, respectively, and an increase of the power utilization of 1 dB for a region of interest in the upper leg. In general, it is found that the VPC allows significant improvement in power utilization when the shimming solution demands only a few modes to be energized, while the technique can yield loss in power utilization in cases with many modes required at high power level.
This paper presents the latest prototype of the integrated emitter turn-off thyristor concept, which potentially ranks among thyristor high-power devices like the gate turn-off thyristor and the integrated gate-commutated thyristor (IGCT). Due to modifications of the external driver stage and mechanical press-pack design optimization, this prototype allows for full device characterization. The turn-off capability was increased to 1600 A with an active silicon area of 823mm2 . This leads to a transient peak power of 672.1kW/cm² . Within this paper, measurements and concept assessment are presented and a comparison to state-of-the-art IGCT devices is provided.
The continuously growing amount of renewable sources starts compromising the stability of electrical grids. Contradictory to fossil fuel power plants, energy production of wind and photovoltaic (PV) energy is fluctuating. Although predictions have significantly improved, an outage of multi-MW offshore wind farms poses a challenging problem. One solution could be the integration of storage systems in the grid. After a short overview, this paper focuses on two exemplary battery storage systems, including the required power electronics. The grid integration, as well as the optimal usage of volatile energy reserves, is presented for a 5- kW PV system for home application, as well as for a 100- MW medium-voltage system, intended for wind farm usage. The efficiency and cost of topologies are investigated as a key parameter for large-scale integration of renewable power at medium- and low-voltage.
High-intensity discharge lamps can be driven by radio-frequency signals in the ISM frequency band at 2.45 GHz, using a matching network to transform the impedance of the plasma to the source impedance. To achieve an optimal operating condition, a good characterization of the lamp in terms of radio frequency equivalent circuits under operating conditions is necessary, enabling the design of an efficient matching network. This paper presents the characterization technique for such lamps and presents the design of the required matching network. For the characterization, a high-intensity discharge lamp was driven by a monofrequent large signal at 2.45 GHz, whereas a frequency sweep over 300 MHz was performed across this signal to measure so-called small-signal hot S-parameters using a vector network analyzer. These parameters are then used as an equivalent load in a circuit simulator to design an appropriate matching network. Using the measured data as a black-box model in the simulation results in a quick and efficient method to simulate and design efficient matching networks in spite of the complex plasma behavior. Furthermore, photometric analysis of high-intensity discharge lamps are carried out, comparing microwave operation to conventional operation.
The interest in PET detectors with monolithic block scintillators is growing. In order to obtain high spatial resolutions dedicated positioning algorithms are required. But even an ideal algorithm can only deliver information which is provided by the detector. In this simulation study we investigated the light distribution on one surface of cuboid LSO scintillators of different size. Scintillators with a large aspect ratio (small footprint and large height) showed significant position information only for a minimum interaction depth of the gamma particle. The results allow a quantitative estimate for a useful aspect ratio.
In our case the double-side-method is used to minimize the complexity of a matrix-readout. Here the number of channels is reduced to 2√N̅. It is also possible to benefit from the method in a single pixel readout system. One signal can be used to measure position and energy of the event, the other one can be applied to a fast trigger-circuit at the same time. In a next step we will investigate timing behavior and electrical crosstalk of the circuit.
The readout of gamma detectors is considerably simplified when the event intensity is encoded as a pulse width (Pulse Width Modulation, PWM). Time-to-Digital-Converters (TDC) replace the conventional ADCs and multiple TDCs can be realized easily in one PLD chip (Programmable Logic Device). The output of a PWM stage is only one digital signal per channel which is well suited for transport so that further processing can be performed apart from the detector. This is particularly interesting for large systems with high channel density (e.g. high resolution scanners). In this work we present a circuit with a linear transfer function that requires a minimum of components by performing the PWM already in the preamp stage. This allows a very compact and also cost-efficient implementation of the front-end electronics.