A comprehensive mathematical model and simulation-based analysis of single and three-phase interleaved synchronous buck converters specifically designed for electric vehicle (EV) battery charging applications is prese...
详细信息
The daily detection of highspeed electric multiple units (EMU) body is very important for China railway maintenance system. This paper proposes a new method based on machine vision to detect bolts and switches on EMU ...
详细信息
This paper explores the development and application of low resource Natural Language Processing (NLP) modules, addressing the challenges of processing underrepresented languages and domains with limited linguistic res...
详细信息
ISBN:
(数字)9798331519056
ISBN:
(纸本)9798331519063
This paper explores the development and application of low resource Natural Language Processing (NLP) modules, addressing the challenges of processing underrepresented languages and domains with limited linguistic resources. It discusses key methodologies such as transfer learning, unsupervised and semi-supervised learning, and data augmentation techniques that enable effective NLP in resource-constrained environments. The paper presents case studies in machine translation, named entity recognition, and sentiment analysis, demonstrating the practical impact of these approaches. Additionally, it outlines persistent challenges in the field and proposes future research directions, emphasizing the importance of enhancing data acces-sibility, model robustness, computational efficiency, and ethical considerations in advancing low resource NLP.
The relentless drive for advanced technologies, fueled by the demands of AI and safety-critical applications, has intensified the focus on transistor aginga pivotal concern that undermines both transistor reliability ...
详细信息
ISBN:
(数字)9798331509422
ISBN:
(纸本)9798331509439
The relentless drive for advanced technologies, fueled by the demands of AI and safety-critical applications, has intensified the focus on transistor aginga pivotal concern that undermines both transistor reliability and overall circuit performance. This issue is further exacerbated by advancements in packaging and 3D integration, where elevated operating temperatures accelerate aging mechanisms. As technology nodes scales below 3 nm, transistor self-heating emerges as a fundamental challenge, driven by the thermal constraints of 3D confined structures. Traditional physics-based simulation tools, such as Technology CAD (TCAD), struggle to meet the growing computational demands of these intricate designs, with escalating simulation times that impede comprehensive design exploration and optimization. Here, we present a novel framework leveraging machine learning (ML) to accelerate transistor and circuit reliability analysis. These ML-driven methodologies achieve accurate predictions of self-heating and aging effects, enabling rapid identification of aging-prone transistors while drastically reducing computational overhead. Furthermore, by obviating the need to share proprietary, physics-based models from semiconductor foundries, these techniques preserve data confidentiality, addressing critical industry concerns. Such approach not only enhances the scalability of reliability assessments but also offers a transformative pathway for tackling the multifaceted challenges of next-generation semiconductor technologies.
This research introduces a novel method for traffic sign identification utilizing a variant 2-D convolutional neural network (CNN) architecture with a shift-invariant operation. Addressing challenges such as variation...
详细信息
ISBN:
(数字)9798350359299
ISBN:
(纸本)9798350359305
This research introduces a novel method for traffic sign identification utilizing a variant 2-D convolutional neural network (CNN) architecture with a shift-invariant operation. Addressing challenges such as variations in size, illumination, occlusions, and perspective, the proposed approach efficiently learns and identifies traffic signs regardless of their location in the input image. The CNN architecture employs common convolutional layers followed by separate shift-invariant convolutional layers for each type of traffic sign, enabling the network to extract both shared and individual traits effectively. Tests conducted on benchmark datasets including the LISA Traffic Sign Dataset (LISA-TSD) and German Traffic Sign Recognition Benchmark (GTSRB) demonstrate superior recognition accuracy compared to state-of-the-art methodologies. The hierarchical design of the proposed model enhances recognisability, making it suitable for intelligent transportation systems. Overall, the study presents a promising variant 2-D CNN model for traffic sign identification with significant potential for real-world applications.
The detection of Alzheimer's disease is a critical task in medical diagnostics due to its rapid progression and profound impact on cognitive function. Deep learning (DL) offers unprecedented capabilities in analyz...
详细信息
ISBN:
(数字)9798350371567
ISBN:
(纸本)9798350371574
The detection of Alzheimer's disease is a critical task in medical diagnostics due to its rapid progression and profound impact on cognitive function. Deep learning (DL) offers unprecedented capabilities in analyzing medical imaging data, particularly in neuroimaging like MRI scans. However, class imbalances within the dataset persist, making it difficult to discern intricate patterns indicative of Alzheimer's pathology. This project aims to overcome this challenge by integrating advanced DL techniques with data augmentation methodologies, specifically SMOTE. This strategic integration aims to overcome the limitations of class imbalances within the initial MRI dataset, enhancing the precision and reliability of the classification model. This interdisciplinary exploration aims to redefine Alzheimer's detection capabilities, enhancing diagnoses and offering the potential for broader applications in understanding and managing complex neurodegenerative disorders.
In transportation simulation, the prediction accuracy of travel times on road segments can have substantial impacts on the simulation outcomes. The travel times are impacted, among other things, by traffic signals. Mo...
详细信息
In transportation simulation, the prediction accuracy of travel times on road segments can have substantial impacts on the simulation outcomes. The travel times are impacted, among other things, by traffic signals. modeling traffic signals is not straightforward in large scale simulations, especially when data on their characteristics is not available. Due to this, in most of the applications using the agent-based transport simulation software MATSim, traffic signals are not explicitly modeled. This paper addresses this issue and proposes a method based on Webster's formula to compute delays experienced by motorists at traffic-signaled intersections without having access to the actual traffic signal data. Within this study, results regarding the imputation of road flow capacities and estimating the impacts of congestion and crossroads on travel times will be presented and investigated. Evidence shows that Webster's approach can substantially improve the quality of the travel time estimates for the regional model of Zurich, Switzerland. (C) 2021 The Authors. Published by Elsevier B.V.
The proceedings contain 50 papers. The topics discussed include: ASCENT+ European infrastructure for nanoelectronics: a deep dive to All-GaN IC technology for power electronics;modeling and simulation of charge trappi...
ISBN:
(纸本)9788363578190
The proceedings contain 50 papers. The topics discussed include: ASCENT+ European infrastructure for nanoelectronics: a deep dive to All-GaN IC technology for power electronics;modeling and simulation of charge trapping in 1/f noise, RTN and BTI: from devices to circuits;modeling passive devices for CMOS RF circuits;a model-oriented methodology for the automatic parameter extraction of TFT model;compact device modeling and simulation with Qucs/Qucs-S/Xyce modular libraries;simulation and modelingmethodologies: enabler for neuromorphic computing applications;and variability-aware characterization of current mirrors based on organic thin-film transistors on flexible substrates.
Maximum power point tracking (MPPT) strategies are applied in PV applications for maximizing power and continuously evaluating the whole power point of photovoltaic (PV) modules under various atmospheric conditions. F...
详细信息
ISBN:
(数字)9798350381887
ISBN:
(纸本)9798350381894
Maximum power point tracking (MPPT) strategies are applied in PV applications for maximizing power and continuously evaluating the whole power point of photovoltaic (PV) modules under various atmospheric conditions. For it to be possible to match the source impedance to the load impedance for MPPT, numerous DC to DC converters (such as the Buck-boost, Boost, and Buck) are inserted between PV arrays and loads in the PV system, and their efficacy and assessments will be discussed in the remainder of this paper. All these DC-DC converters and MPPT tracking (P & O) algorithm are modelled using MATLAB/Simulink and results shows that the maximum power is enhanced with Buck-Boost converter than other dc-dc converters.
This contribution focuses on the modeling and performance analysis of the manufacturing process for transport platforms at the Alimak Group facility in La Muela, Spain. The objective is to leverage formal tools to gai...
This contribution focuses on the modeling and performance analysis of the manufacturing process for transport platforms at the Alimak Group facility in La Muela, Spain. The objective is to leverage formal tools to gain insights into this complex manufacturing process and explore potential applications such as control and optimization. This work presents a preliminary study, where a Generalized Stochastic Petri net model of the manufacturing process is proposed to be used for simulation, performance analysis, and optimization of the system.
暂无评论