3633 research outputs found
Sort by
A digital dithering phase shift modulator for enhanced resolution
No full textThis article proposes a digital dithering phase shift modulator to control power electronic converters. The presented modulator allows a higher resolution for digitally generated phase-shift signals without the need for fine time steps using very high-frequency or high-resolution timers. To achieve this, two coarse, counter-based phase-shift signals that differ in their phase shift values by one bit are used, and a total phase-shift output signal is derived by periodically switching between these two. The result is a dithered phase-shift control signal with an increased switching frequency or resolution. A prototype circuit for the control signal generation for a dual-active half-bridge converter is presented. Using only a 48 MHz digital clock signal, a control signal with a switching frequency of 188 kHz and a resolution of 12 bits is achieved. The modulation concept is described, and its application is verified using simulations and experimental tests
Optimal dynamic operation of electrolyzers considering energy dispatch intervals due to short term power allocation
No full textThe production of green hydrogen is of paramount importance for the global transition to a carbon free energy supply. Electrolyzers that run on renewable energy are used to produce green hydrogen. The installed renewable power capacity is expected to vastly increase over the coming years. This greatly increased capacity in combination with the volatile nature of such power sources will require a short term energy dispatch. In this scenario an electrolyzer must be able to operate in dynamic conditions. This paper presents an optimal control solution to this future real world application considering the unique operational and economical characteristics of an electrolyzer as well as the dynamic mode of operation due to future short term energy dispatch. The necessity of such a control strategy is demonstrated by simulation results
An analysis of blockchain versus relational databases for digitalising information flows in global supply chains using the analytic network process
No full textGlobal trade is plagued by slow and inefficient manual processes associated with physical documents. Firms are constantly looking for new ways to improve transparency and increase the resilience of their supply chains. This can be solved by the digitalisation of supply chains and the automation of document- and information-sharing processes. Blockchain is touted as a solution to these issues due to its unique combination of features, such as immutability, decentralisation and transparency. A lack of business cases that quantify the costs and benefits causes uncertainty regarding the truth of these claims. This paper explores how the costs and benefits of a blockchain-based solution for digitalising and automating documentation flows in cross-border supply chains compare to a conventional centralised relational database solution. The research described in this paper uses primary data collected through semi-structured interviews with industry experts, as well as secondary data from literature. Two models based on existing services were developed and the costs and benefits compared and then analysed using the Architecture Trade-off Analysis Method (ATAM) and the Analytic Network Process (ANP). Findings from the analysis show that a consortium blockchain solution like TradeLens is the favourable solution for digitalising and automating information flows in cross-border supply chains
5G-Campusnetze: Kommunikationstechnologie für mobile Roboter?
Get PDFMobile Roboter sind entscheidend für die automatisierte Intralogistik der Industrie 4.0. Eine sichere drahtlose Anbindung an Flottenmanager oder Steuerungssysteme ist essenziell. Private 5G-Campusnetzwerke mit lizenzierten Frequenzen gelten als vielversprechende Lösung. Aus diesem Grund beleuchtet der Beitrag die Grundlagen der 5G-Technologie für mobile Roboter sowie die aktuelle Leistungsfähigkeit von privaten 5G-Campusnetzwerken anhand erhobener Messungen.Mobile robots enable the automated intralogistics of Industry 4.0. Prerequisite is a secure wireless connection to fleet managers or control systems . Private 5G campus networks with licensed frequencies are a promising solution. For this reason, this article examines the basics of 5G technology for mobile robots and the current performance of private 5G campus networks based on measurement
An enhanced ACBC three-stage amplifier using complementary indirect Miller compensation
No full textThe design of operational amplifiers faces the trade-off between power and speed. Increasing power-efficiency according to a certain speed requires a multi-stage design with an optimal compensation network that should be stable under all possible operating conditions. From this point of view, this paper proposes a three-stage operational amplifier with an enhanced AC boosting compensation (ACBC) using complementary indirect Miller compensation. The proposed amplifier is implemented in GlobalFoundries 22FDX (22 nm FD-SOI technology). In the worst case of the post-layout simulation, the proposed design provides a power efficiency IFOMs of 231k MHz.pF/mW, which is 2x larger compared to similar designs with ACBC as the standard compensation technique while driving a 500 pF load. According to this IFOMs, the design achieves an open-loop gain of 103.5 dB, a gain-bandwidth product of 8.5 MHz with a phase margin of 41.4 while the amplifier operates with a 10% reduced supply voltage of 0.72 V in ss-corner under 125. A total current of only 18.7 including the bias-network, is consumed
How to conduct successful business process automation projects? An analysis of key factors in the context of robotic process automation
Get PDFIn recent years, the robotic process automation (RPA) technology, a software-based method to automate routine tasks in business processes, has gained significant interest and adoption. However, many implementation projects fail and current literature lacks a synthesis and comprehensive overview of factors that challenge the implementation of RPA, have an impact on success or failure of projects, or, play an enabling role in an RPA project. Hence, the purpose of this research is to identify key factors that should be considered by organizations when conducting an RPA project.
The paper adopts a qualitative methodology based on data collected in a systematic literature review (SLR) and interviews with 10 RPA experts. Using inductive coding, an integrated framework of key factors is developed.FindingsThe results suggest that the key factors for a successful RPA introduction can be divided into human, organizational and technical factors. Important aspects include for example project management techniques, capabilities and skills of employees, as well as data security considerations.
The paper contributes to knowledge by synthesizing previously dispersed knowledge into an integrated framework, as well as by complementing previous results with new qualitative, empirical data. Additionally, the RPA-specific factors are put into the perspective of persistent problems in information systems development
Grading student behavior
No full textNumerous countries mandate comportment grades rating students’ social and work behavior in the classroom from teachers, yet their impact on student outcomes remains unclear. We exploit the staggered introduction of comportment grading across German federal states to estimate its causal effect on students’ school-to-work transitions, non-cognitive skills, and reading skills. Analyzing two different household surveys and student assessment data, point estimates of causal effects are close to zero for all outcomes. However, while confidence intervals for school-to-work transitions and non-cognitive skills allow us to reject meaningful effect sizes, those for reading skills are wider and need to be interpreted more cautiously. We use additional data sources to investigate potential explanations
Is it necessary to set up a personality for the development of AI robots with a qualitative identity?
No full textMoonshot Project Goal 3 aims to develop an AI robot that grows alongside people’s lives by 2050. In order for robots to grow together with our lives, it is necessary for them to have personalities. We will discuss what characteristics an AI robot should have both internally and externally. Currently, chatbots using LLM such as ChatGPT are being developed, but problems have arisen such as chatbots encouraging users to commit suicide. We will also discuss the problems caused by chatbots having personalities. In addition to personality, qualitative identity is also important for robots to stay close to people for a lifetime. Today’s robots have fixed personalities and cannot be changed. Therefore, personality cannot be inherited. Furthermore, in the case of robots, their individuality is limited to their appearance and physical functions. We will discuss how robots can continue to be passed down from generation to generation despite these differences
Continuous remaining useful life prediction by self-guided attention convolutional neural network and memory consciousness adjustment
No full textTo accurately predict the remaining useful life (RUL) of rotating machinery while continuously providing the task data, a novel continuous RUL prediction methodology was proposed. The methodology comprises a self-guided attention convolutional neural network (SGACNN) and memory consciousness adjustment (MCA) mechanism. First, a multihead focal channel-wise self-attention (MFCWSA) mechanism was implemented to effectively capture the degradation information across all the channels and achieve the attentional focus. Next, the SGACNN was constructed using the MFCWSA, squeeze-and-excitation mechanism, and convolutional block attention module. A new network gradient direction was synthesized by leveraging the gradients from both the previous task and the current task. Further, a weight constraint loss term based on the gradient magnitude was designed to constrain the learning process of important parameters. With the new network gradient direction and weight constraint loss, a novel MCA mechanism was proposed and integrated into the SGACNN for implementing the continuous RUL prediction tasks. Finally, various RUL prediction experiments on the life-cycle bearing and gear data sets were carried out, and its outcomes were compared to those of the advanced methods of the same kind. The comparative results validated the superiority of the proposed methodology
Enhancing power skiving tool longevity: the synergy of AI and robotics in manufacturing automation
Get PDFIn gear manufacturing, the longevity and cost-effectiveness of power skiving tools are essential. This study presents an innovative approach that combines artificial intelligence and robotics in manufacturing automation to prevent tool breakage to improve the remaining useful life (RUL). Using a robotic cell, the system captures six images per tooth from different angles. An unsupervised generative deep learning model approach is used because it is more suitable for industrial application as it can be trained with a small number of defect-free images. It is used in a first step as a classifier and, in a second step, to segment tool wear. This approach promises economic benefits by reducing manual inspection and, through automated tool inspection, detecting wear earlier to prevent tool breakage