100kW Energy Transfer Multiplexer Power Converter Prototype Development Project

The Impact of Modular Assembly on Supply Chain Efficiency

This paper presents a modular low cost approach for building ultra-linear amplifiers for base station telecommunications applications with the use of a low cost stripline serial and parallel combining techniques. The operation of the amplifiers provide "feedforward performance level" with very low complexity assembly and test costs.

The feasibility of developing a modular system approach for the construction of containment mats has been established. The investigative results obtained from a plant used for the modular study has determined that (1) modularization results in a 10-week reduction in the construction schedule critical path over that for the conventional approach, and (2) modularization is cost effective and could result in a savings of approximately $10,700,000. Accordingly, it is recommended that the modular system approach be adopted as standard practice for the construction of containment mats.

The construction sector is notorious for the dichotomy between (a) its intensive use of information in its decision-making processes and (b) its limited access to, and insufficient use of, the pertinent information that is potentially available. In the context of the potential availability of valid information on the Web, we have developed a question-and-answer system, which enables construction practitioners to seek for information by posing questions in English or French, instead of entering a list of relevant keywords. Based on a modular systems approach using natural language, relevant answers to questions are selected and presented in a convivial way, thus improving and speeding up the classical Web search procedure. Our system consists of two main components: An intelligent robot that traverses a Web space and decides whether a page is construction oriented or not, and a question-answer (Q-A) component (comprising three modules) that uses a domain-specific thesaurus to extract meaningful parts of a question and to detect, process, and present paragraph passages extracted from relevant Web pages. The robot is trained on positive and unlabeled examples using a machine learning approach, while the Q-A component uses natural language processing techniques. In our experiments, we show that our automatically collected database consists of approximately 16% of noise, while the performance of the Q-A component is 65.45% in mean reciprocal rank.

Course and instructor evaluations (CIE) are essential tools for assessing educational quality in higher education. However, traditional CIE systems often suffer from inconsistencies between structured responses and open-ended feedback, leading to unreliable insights and increased administrative workload. This study suggests a modular system to address these challenges, leveraging sentiment analysis and inconsistency detection to enhance the reliability and efficiency of CIE processes.BackgroundImproving the reliability of CIE data is crucial for informed decision-making in higher education. Existing methods fail to address discrepancies between numerical scores and textual feedback, resulting in misleading evaluations. This study proposes a system to identify and exclude inconsistent data, providing more reliable insights.MethodsUsing the Design Science Research Methodology (DSRM), a system architecture was developed with five modules: data collection, preprocessing, sentiment analysis, inconsistency detection, and reporting. A dataset of 13,651 anonymized Turkish CIE records was used to train and evaluate machine learning algorithms, including support vector machines, naive Bayes, random forest, decision trees, K-nearest neighbors, and OpenAI’s GPT-4 Turbo Preview model. Sentiment analysis results from open-ended responses were compared with structured responses to identify inconsistencies.ResultsThe GPT-4 Turbo Preview model outperformed traditional algorithms, achieving 85% accuracy, 88% precision, and 95% recall. Analysis of a prototype system applied to 431 CIEs identified a 37% inconsistency rate. By excluding inconsistent data, the system generated reliable reports with actionable insights for course and instructor performance. The purpose of this study is to design and evaluate a new system using the Design Science Research (DSR) approach to enhance the accuracy and reliability of course evaluation processes employed in higher education institutions. The modular system effectively addresses inconsistencies in CIE processes, offering a scalable and adaptable solution for higher education institutions. By integrating advanced machine learning techniques, the system enhances the accuracy and reliability of evaluation reports, supporting data-driven decision-making. Future work will focus on refining sentiment analysis for neutral comments and broadening the system’s applicability to diverse educational contexts. This innovative approach represents a significant advancement in leveraging technology to improve educational quality.

Real world environments are so dynamic and unpredictable that a goal-oriented autonomous system performing a set of tasks repeatedly never experiences the same situation even though the task routines are the same. Hence, manually designed solutions to execute such tasks are likely to fail due to such variations. Developmental approaches seek to solve this problem by implementing local learning mechanisms to the systems that can unfold capabilities to achieve a set of tasks through interactions with the environment. However, gathering all the information available in the environment for local learning mechanisms to process is hardly possible due to limited resources of the system. Thus, an information acquisition mechanism is necessary to find task-relevant information sources and applying a strategy to update the knowledge of the system about these sources efficiently in time. A modular systems approach may provide a useful structured and formalized basis for that. In such systems different modules may request access to the constrained system resources to acquire information they are tuned for. We propose a reward-based learning framework that achieves an efficient strategy for distributing the constrained system resources among modules to keep relevant environmental information up to date for higher level task learning and executing mechanisms in the system. We apply the proposed framework to a visual attention problem in a system using the iCub humanoid in simulation.

Dividing a System on Chip (SoC) into multiple smaller chiplets and embedding them into a single package has gained significant traction in recent years through more widespread adoption in digital circuitry. Despite benefits of reduced interference, protection against environmental influences and overcoming the “Interconnection Gap”, its usage in integrated analog circuits fails to match pace with digital designs. Adoption is likely impeded by an over-reliance on single-chip packages as they prioritize high-frequency performance over integration densities. We therefore demonstrate the first modular system approach with an embedded Wafer-Level Ball grid array (eWLB) package that does not compromise on either integration density nor performance. Restricted only by the maximum package dimensions, the number of channels can be adjusted application-specifically. This allows the system to scale down for low-cost, low-power applications while conversely facilitating massive MIMO. Exemplarily, a 4 × 4 radar System in Package (SiP) with five 130 nm B11HFC SiGe chiplets in a small form factor of 7.8 mm × 8.8 mm was manufactured for this work. It contains a central VCO that feeds four transceivers of identical design that can be configured as receivers or transceivers through the package’s layout. The configuration is solely package-based, enabling chip designs to be reused and thus drastically reducing development time. It also permits homogeneous or heterogeneous substitution of the chiplets based on available fabrication facilities and economic considerations. With its 15.6 dBi comb-line antennas, target detection within 76–77 GHz has been verified up to 36 m in range and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\pm$</tex-math></inline-formula>30° in azimuth. Adverse to single-chip solutions, this novel chiplet approach splits up temperature hotspots into smaller, localized areas of elevated temperature. While advantageous for the dissipation of heat, it imposes additional challenges thermomechanically as well as electromagnetically. The compromise between performance and reliability is therefore addressed with a detailed examination of the solderball placement and package-to-PCB interfaces.

Fog computing aims to provide the solution of bandwidth, network latency and energy consumption problems of cloud computing. Likewise, management of data generated by healthcare IoT devices is one of the significant applications of fog computing. Huge amount of data is being generated by healthcare IoT devices and such types of data is required to be managed efficiently, with low latency, without failure, and with minimum energy consumption and low cost. Failures of task or node can cause more latency, maximum energy consumption and high cost. Thus, a failure free, cost efficient, and energy aware management and scheduling scheme for data generated by healthcare IoT devices not only improves the performance of the system but also saves the precious lives of patients because of due to minimum latency and provision of fault tolerance. Therefore, to address all such challenges with regard to data management and fault tolerance, we have presented a Fault Tolerant Data management (FTDM) scheme for healthcare IoT in fog computing. In FTDM, the data generated by healthcare IoT devices is efficiently organized and managed through well-defined components and steps. A two way fault-tolerant mechanism i.e., task-based fault-tolerance and node-based fault-tolerance, is provided in FTDM through which failure of tasks and nodes are managed. The paper considers energy consumption, execution cost, network usage, latency, and execution time as performance evaluation parameters. The simulation results show significantly improvements which are performed using iFogSim. Further, the simulation results show that the proposed FTDM strategy reduces energy consumption 3.97%, execution cost 5.09%, network usage 25.88%, latency 44.15% and execution time 48.89% as compared with existing Greedy Knapsack Scheduling (GKS) strategy. Moreover, it is worthwhile to mention that sometimes the patients are required to be treated remotely due to non-availability of facilities or due to some infectious diseases such as COVID-19. Thus, in such circumstances, the proposed strategy is significantly efficient.

A recently proposed active input current shaper (AICS) is used to turn a conventional buck converter into a single stage AC-to-DC power converter that complies with IEC 1000-3-2 regulations with high efficiency, fast transient response and low cost. This topology is used in this paper to design an online battery charger with no galvanic isolation. This solution is a very simple one and only implies slight changes to the conventional topology. Besides, efficiency is high and the size increase is small, all of which are obtained at a very low cost.

Circuit-based optical networks have the potential to provide end-to-end dedicated channels with flexible data rates between edge node pairs. However, the data-rate granularity of spectral-slicing in the optical domain based on wavelength-selective switch (WSS) is limited by the optical filter frequency selectivity. Digital subcarrier multiplexing (DSCM) in the electrical domain is able to provide much finer data rate granularity and flexibility. A downside of DSCM is its increasingly higher speed of digital electronics required to accommodate traffic carried by a wavelength. In this paper we demonstrate a digital-analog hybrid subcarrier multiplexing (SCM) technique that can reduce the speed requirement of digital electronics such as ADC and DAC, while providing wideband capability, high spectral efficiency, controllable data-rate granularity and operational flexibility. Hybrid SCM is a modular approach in which multiple digitally generated subcarriers are aggregated through RF oscillators and IQ mixers for frequency up- and down-conversions. While digital electronic processing provides high level flexibility for software controllable and spectrally efficient modulation, mature RF multiplexing and demultiplexing can inexpensively fill up the relatively wide bandwidth on a wavelength. Thanks to the rapid development of wireless systems, high quality RF oscillators and mixers are readily available with low power consumption and low cost. This paper presents experimental results which demonstrate feasibility and application perspectives of a hybrid SCM system.

The fundamental innovative design of managed optical communications array (MOCA) features some distinct advantages over other lasercomm approaches: (1) MOCA relies on proven, Commercial off-the-shelf (COTS) technologies, as opposed to more exotic optical phase array or beam-forming approaches. MOCA has already been demonstrated (2) the MOCA technology array supports a low-profile conformal terminals for low-drag; (3) MOCA eliminates failure-prone gimbals (4) the modular approach allows development dollars to be focused at the shared subaperture level, from which unique platform terminals can be developed at lower cost; and (5) use of commoditized components and elimination of gimbals reduces size, weight, and cost.

Tomato Care 5.0 is an advanced precision agriculture framework that integrates Edge Artificial Intelligence (Edge-AI) and robotics to optimize crop health management and resource management in smart tomato farms. The system’s goal is to receive input from AI-oriented processing of image data at the Edge, to perform real-time color-aware diagnostics of leaf conditions. Real-time, color-aware diagnostics in agriculture allows for timely responses for managing optimal nutrient element concentrations in the crops, and managing on-plant diseases, which improve crops overall health. In addition to Leaf Condition diagnostics, the Tomato Care framework incorporates robotics and autonomous mobility to perform targeted spraying of fertilizers or pesticides, reducing chemical use and environmental impact. The combination of color aware diagnostics and autonomous actuation will provide a way to manage accurate timing and localize the interventions, which will collectively improve plant health and efficiency of resource use, as well as the overall quality of yield. Tomato Care 5.0 is a scalable and affordable solution for managing processes for contemporary management of agricultural crops, especially in regions with limited access to centralized computing. Tomato Care 5.0 operates with sensor fusion, intelligent decision-making, and changing surroundings and evolve as the plant grows. This system has a modular approach that enables connectivity with many IoT devices so farmers can continuously monitor the temperature, soil moisture, and humidity. This improves the contextual accuracy of detecting disease and when to spray. Since Tomato Care 5.0 operates at the edge, there is little latency, so there is no dependence on the cloud, and it enhances data privacy. This solution assists farmers in providing actionable items while practicing environmentally-friendly farming through maximizing sustainability and lowering costs through practices that provide data-driven logical decisions. Tomato Care 5.0 represents an advancement in an Agri tech solution within the merger of AI, robotics, and precision agriculture and sets new standards for the new paradigm of smart farming through autonomous crop management. Tomato Care 5.0 represents a new Edge-AI and robotics-based framework for precision agriculture with the specificity of detecting leaf disease through color sensitive and real time vision and autonomous application of resources for proper and timely interventions, including but not limited to automated spraying of fertilizer, irrigation activated by soil moisture and sowing seeds. Tomato Care 5.0 leverage’s AI based on-device processing, autonomous mobility, and multi-modal sensor fusion to enable automated & real-time applications; no cloud processing and no human action - only the right things done at the right time. This fundamentally differs from the traditional means for precision agriculture since it minimizes latency on time, improves efficacy on purpose and mode of application, and supports sustainable crop management and production and particularly tomato crop cultivation.

We report on a new manufacturing paradigm to build scalable microbolometers at the lowest cost possible. A VGA microbolometer focal plane array has been designed and fabricated using flat panel display production tools, using the Large Area MEMS Platform (LAMP). Complex MEMs sensors have been integrated with Low Temperature Polysilicon (LTPS) transistors on a Gen 3.5 glass substrate in a display production fab to realize an order of magnitude gain in the number of devices per substrate. In this modular approach, a custom readout integrated circuit (ROIC) is bonded to the glass focal plane with the Chip on Flex (COF) technique, a high volume and low-cost staple of the display industry. We discuss the potential implications of the pricing disruption to important commercial applications.

Ceres has developed a novel and highly differentiated SteelCell™ solid oxide fuel cell technology based upon the use of thick-film ceramics deposited upon a ferritic stainless-steel substrate, using doped ceria as the predominant oxygen-ion conducting ceramic within the cell. The use of ceria allows operation of the SteelCell at lower temperatures (500 - 620 °C) than conventional SOFCs, and the use of a metal support allows much greater mechanical robustness and lower production cost than is typically the case with a planar ceramic SOFC, whilst maintaining the high volumetric power density typical of planar SOFCs.Ceres Power are now qualifying their SteelCell™ technology for use in steam electrolysis operation with encouraging results. Tests suggest that functional electrochemical and physical processes of the SteelCell are equally efficient in SOFC and SOEC mode meaning there is no major difference in the area specific resistance and degradation within the nominal operating space.As a next step, the development of a 1MW-class Solid Oxide Electrolysis (SOEL) system is proposed that offers a path to green hydrogen production costs of <$2/kgH2 with equally compelling capital equipment costs enabled by the SteelCell technology. One of the clear advantages of solid oxide electrolysis systems is the ability to thermally integrate with industrial processes such as steel, ammonia and e-fuels production. Integration with existing low pressure “waste” steam feeds available at industrial plants combined with available high grade heat streams may enable system efficiencies of greater than 90% to be achieved. A study of predicted system efficiencies and associated techno-economics of green hydrogen production under various industrial use cases is presented.The first-of-a-kind 1MW-class Solid Oxide Electrolysis system under consideration will be packaged within a 40ft ISO shipping container and is expected to produce approximately 600 to 800kg of green hydrogen per day. The system will be made up of a number of Electrolysis Cell Modules (ECM). Each ECM will contain an array of Ceres’ standard Gen1B stacks along with hot balance of plant components such as heat exchangers and heaters. This modular approach allows each module to be constructed and tested prior to installation at the demonstration site. It also allows for a high degree of fault tolerance and redundancy that is particularly valuable for prototype systems. Furthermore, it also enables a distributed supply chain to be developed for various elements up and down the system structure.Finally, a roadmap to the first commercial SOEL systems is presented setting out predicted system CapEx costs, hydrogen production costs and possible arrangements for systems in the 100+ MW range.

Resistive memories have limited lifetime caused by limited write endurance and highly non-uniform write access patterns. Two main techniques to mitigate endurance-related memory failures are 1) wear-leveling, to evenly distribute the writes across the entire memory, and 2) fault tolerance, to correct memory cell failures. However, one of the main open challenges in extending the lifetime of existing resistive memories is to make both techniques work together seamlessly and efficiently. To address this challenge, we propose WoLFRaM, a new mechanism that combines both wear-leveling and fault tolerance techniques at low cost by using a programmable resistive address decoder (PRAD). The key idea of WoLFRaM is to use PRAD for implementing 1) a new efficient wear-leveling mechanism that remaps write accesses to random physical locations on the fly, and 2) a new effiCient fault tolerance mechanism that recovers from faults by remapping failed memory blocks to available physical locations. Our evaluations show that, for a Phase Change Memory (PCM) based system with cell endurance of 108 writes, WoLFRaM increases the memory lifetime by 68% compared to a baseline that implements the best state-of-the-art wear-leveling and fault correction mechanisms. WoLFRaM's average / worst-case performance and energy overheads are 0.51% /3.8% and 0.47% /2.1% respectively.