检索结果-内蒙古大学图书馆

FedKDC: Consensus-Driven Knowledge Distillation for Personalized Federated Learning in EEG-Based Emotion Recognition

引用

IEEE Journal of Biomedical and Health informatics 2025年 PP卷 PP页

作者： Qiu, Xihang Qiu, Wanyong Zhang, Ye Qian, Kun Li, Chun Hu, Bin Schuller, Bjorn W. Yamamoto, Yoshiharu Beijing100081 China Beijing Institute of Technology School of Automation School of Medical Technology Beijing100081 China Shenzhen MSU-BIT University MSU-BIT-SMBU Joint Research Center of Applied Mathematics Shenzhen518172 China Beijing Institute of Technology School of Mathematics and Statistics Beijing100081 China Imperial College London GLAM – the Group on Language Audio & Music LondonSW7 2AZ United Kingdom Technische Universität München CHI Chair of Health Informatics Munich80333 Germany Germany Educational Physiology Laboratory Graduate School of Education University of Tokyo Tokyo1138654 Japan

Federated learning (FL) has gained prominence in electroencephalogram (EEG)-based emotion recognition because of its ability to enable secure collaborative training without centralized data. However, traditional FL faces challenges due to model and data heterogeneity in smart healthcare settings. For example, medical institutions have varying computational resources, which creates a need for personalized local models. Moreover, EEG data from medical institutions typically face data heterogeneity issues stemming from limitations in participant availability, ethical constraints, and cultural differences among subjects, which can slow model convergence and degrade model performance. To address these challenges, we propose FedKDC, a novel FL framework that incorporates clustered knowledge distillation (CKD). This method introduces a consensus-based distributed learning mechanism to facilitate the clustering process. It then enhances the convergence speed through intraclass distillation and reduces the negative impact of heterogeneity through interclass distillation. Additionally, we introduce a DriftGuard mechanism to mitigate client drift, along with an entropy reducer to decrease the entropy of aggregated knowledge. The framework is validated on the SEED, SEED-IV, SEED-FRA, and SEED-GER datasets, demonstrating its effectiveness in scenarios where both the data and the models are heterogeneous. Experimental results show that FedKDC outperforms other FL frameworks in emotion recognition, achieving a maximum average accuracy of 85.2%, and in convergence efficiency, with faster and more stable convergence. © 2013 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Energy informatics - From research to implementation: A review of the track "Energy" at the INFORMATIK 2020 50

Energy informatics - From research to implementation: A revi...

引用

50. Jahrestagung der Gesellschaft fur Informatik, INFORMATIK 2020 - 50th Annual Conference of the German informatics Society, INFORMATIK 2020

作者： Nieße, Astrid Hagenmeyer, Veit Fk. II – Department für Informatik Carl von Ossietzky Universität Oldenburg Abteilung Digitalisierte Energiesysteme Oldenburg26111 Germany Institute for Automation and Applied Informatics Eggenstein-Leopoldshafen76344 Germany

来源：评论

学校读者我要写书评

暂无评论

The vulnerability of securing IoT production lines and their network components in the Industry 4.0 concept

引用

IFAC-PapersOnLine 2020年第2期53卷 11237-11242页

作者： Tibor Horak Zuzana Cervenanska Ladislav Huraj Pavel Vazan Jan Janosik Pavol Tanuska Institute of Applied Informatics Automation and Mechatronics Faculty of Material Science and Technology in Trnava Slovak University of Technology in Bratislava Trnava Slovak Republic Department of Applied Informatics University of SS. Cyril and Methodius Trnava Slovakia

IoT systems are an integral part of every modern industrial enterprise Industry 4.0. IoT is the term for modern remote devices controlled via the Internet. Internet of Things is the name of technologies that allow cheap wireless connection and communication of various sensors and devices to automate, accelerate and streamline processes. In the interconnected world of Industry 4.0, there are many potential resources existing for infiltration. Cybercriminals could take control of manufacturing industries, manipulate machines, or could do an industrial espionage. This type of attack is called Denial of Service. In the second case, the attack preserves the attacker’s anonymity through an IP address by using a potentially innocent third party (a reflector) that is indirectly involved in the attack. Through this attack, the attacker forwards the flow of attacking data to the target victim. The attacker sends the packets with a fake spoof source IP address set to the victim’s IP address to the reflector, thus indirectly overloading the target with the packets, or it will intrude into a network device through a faulty WPS implementation. The simulation model of the production line and the IoT security system Fibaro were used to investigate these attacks. The article demonstrates the possibility of attacks on network devices and the misuse of IoT devices in order to compromise production machines which use DRDoS and Brute-force attacks.

关键词： Industry 4.0 IoT devices DRDoS attacks WPS Production line Simulation

来源：评论

学校读者我要写书评

暂无评论

Distributed AC Optimal Power Flow: A Scalable Solution for Large-Scale Problems

arXiv

引用

arXiv 2025年

作者： Dai, Xinliang Jiang, Yuning Guo, Yi Jones, Colin N. Diehl, Moritz Hagenmeyer, Veit Institute for Automation and Applied Informatics The Karlsruhe Institute of Technology Germany Automatic Control Laboratory EPFL Switzerland Urban Energy Systems Laboratory Swiss Federal Laboratory for Materials Science and Technology Switzerland Department of Mathematics University of Freiburg Germany

This paper introduces a novel distributed optimization framework for large-scale AC Optimal Power Flow (OPF) problems, offering both theoretical convergence guarantees and rapid convergence in practice. By integrating smoothing techniques and the Schur complement, the proposed approach addresses the scalability challenges and reduces communication overhead in distributed AC OPF. Additionally, optimal network decomposition enables efficient parallel processing under the single program multiple data (SPMD) paradigm. Extensive simulations on large-scale benchmarks across various operating scenarios indicate that the proposed framework is 2 to 5 times faster than the state-of-the-art centralized solver IPOPT on modest hardware. This paves the way for more scalable and efficient distributed optimization in future power system applications. Copyright © 2025, The Authors. All rights reserved.

关键词： Parallel processing systems

来源：评论

学校读者我要写书评

暂无评论

Corrigendum to “ALDI++: Automatic and parameter-less discord and outlier detection for building energy load profiles” [Energy Build. 265 (2022) 112096]

引用

Energy and Buildings 2023年 286卷

作者： Matias Quintana Till Stoeckmann June Young Park Marian Turowski Veit Hagenmeyer Clayton Miller College of Design and Engineering National University of Singapore Singapore Institute for Automation and Applied Informatics Karlsruhe Institute of Technology Germany Department of Civil Engineering University of Texas at Arlington USA

来源：评论

学校读者我要写书评

暂无评论

Label assistant: A workflow for assisted data annotation in image segmentation tasks

arXiv

引用

arXiv 2021年

作者： Schilling, Marcel P. Rettenberger, Luca Münke, Friedrich Cui, Haijun Popova, Anna A. Levkin, Pavel A. Mikut, Ralf Reischl, Markus Institute for Automation and Applied Informatics Karlsruhe Institute of Technology Hermann-von-Helmholtz-Platz 1 Eggenstein-Leopoldshafen76344 Germany Institute of Biological and Chemical Systems Karlsruhe Institute of Technology Hermann-von-Helmholtz-Platz 1 Eggenstein-Leopoldshafen76344 Germany

Recent research in the field of computer vision strongly focuses on deep learning architectures to tackle image processing problems. Deep neural networks are often considered in complex image processing scenarios since traditional computer vision approaches are expensive to develop or reach their limits due to complex relations. However, a common criticism is the need for large annotated datasets to determine robust parameters. Annotating images by human experts is time-consuming, burdensome, and expensive. Thus, support is needed to simplify annotation, increase user efficiency, and annotation quality. In this paper, we propose a generic workflow to assist the annotation process and discuss methods on an abstract level. Thereby, we review the possibilities of focusing on promising samples, image pre-processing, pre-labeling, label inspection, or post-processing of annotations. In addition, we present an implementation of the proposal by means of a developed flexible and extendable software prototype nested in hybrid touchscreen/laptop device. © 2021, CC BY.

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

An Industry 4.0-Compliant Digital Product Passport Approach for Realising Dairy Product Traceability

An Industry 4.0-Compliant Digital Product Passport Approach ...

引用

Annual Conference of Industrial Electronics Society

作者： Martín Alejandro Bär Armando Walter Colombo José Luis Torres Erica Fernandez Mariela Rico Maria Laura Caliusco Institute for Industrial Informatics Automation and Robotics University of Applied Sciences Emden/Leer Emden Germany UTN - Fac. Reg. Santa Fe CIESE Research Center Santa Fe Argentina Industrial Engineering UTN - Fac. Reg. Rafaela - Santa Fe Santa Fe Argentina UTN - Fac. Reg. Santa Fe CIDISI Research Center Santa Fe Argentina

This paper provides an approach to create the digital thread associated to the value stream along a dairy process value chain. Identifying the different stakeholders within the dairy process, each of them is concretely positioned as provider of assets to be digitalized and networked using the Asset Administration Shell technology and this within an Eco-system formally specified by the 3D-Reference Architecture Model for Industry 4.0 (RAMI 4.0). The different sub-models of the Asset Administration Shells along the digital thread are then an essential part of the so called Digital Passport, which will be associated to different dairy products (milk powders, butter, whey, cheese, etc.). The Digital Passport is the core tool allowing all stakeholders for performing traceability of products and processes. It provides digitalized data and information, to be exposed and/or consumed as services, developed and shared in real-time by all the actors of the dairy supply chain, responding to strict dairy process business requirements.

关键词：

来源：评论

学校读者我要写书评

暂无评论

ED4GAP: Efficient Detection for GOOSE-Based Poisoning Attacks on IEC 61850 Substations

ED4GAP: Efficient Detection for GOOSE-Based Poisoning Attack...

引用

IEEE International Conference on Smart Grid Communications (SmartGridComm)

作者： Atul Bohara Jordi Ros-Giralt Ghada Elbez Alfonso Valdes Klara Nahrstedt William H. Sanders Information Trust Institute University of Illinois at Urbana-Champaign Reservoir Labs Institute of Automation and Applied Informatics Karlsruhe Institute of Technology

ISBN: (数字)9781728161273

ISBN: (纸本)9781728163598

Devices in IEC 61850 substations use the generic object-oriented substation events (GOOSE) protocol to exchange protection-related events. Because of its lack of authentication and encryption, GOOSE is vulnerable to man-in-the-middle attacks. An adversary with access to the substation network can inject carefully crafted messages to impact the grid's availability. One of the most common such attacks, GOOSE-based poisoning, modifies the StNum and SqNum fields in the protocol data unit to take over GOOSE publications. We present ED4GAP, a network- level system for efficient detection of the poisoning attacks. We define a finite state machine model for network communication concerning the attacks. Guided by the model, ED4GAP analyzes network traffic out-of-band and detects attacks in real-time. We implement a prototype of the system and evaluate its detection accuracy. We provide a systematic approach to assessing bottlenecks, improving performance, and demonstrating that ED4GAP has low overhead and meets GOOSE's timing constraints.

关键词： Substations Protocols Relays IEC Standards Security Real-time systems Ethernet

来源：评论

学校读者我要写书评

暂无评论

Local versus global stability in dynamical systems with consecutive Hopf bifurcations

引用

Physical Review Research 2023年第3期5卷 033139-033139页

作者： Philipp C. Böttcher Benjamin Schäfer Stefan Kettemann Carsten Agert Dirk Witthaut Institute of Energy and Climate Research Systems Analysis and Technology Evaluation (IEK-STE) Forschungszentrum Jülich 52428 Jülich Germany Institute of Energy and Climate Research Energy Systems Engineering (IEK-10) Forschungszentrum Jülich 52428 Jülich Germany Institute for Automation and Applied Informatics (IAI) Karlsruhe Institute of Technology (KIT) Hermann-von-Helmholtz-Platz 1 76344 Eggenstein-Leopoldshafen Germany School of Computer Science & Engineering Constructor University Campus Ring 1 28759 Bremen Germany DLR Institute of Networked Energy Systems Carl-von-Ossietsky Straße 15 26129 Oldenburg Germany Institute for Theoretical Physics University of Cologne Zülpicher Straße 77 50937 Cologne Germany

Quantifying the stability of an equilibrium is central in the theory of dynamical systems as well as in engineering and control. A comprehensive picture must include the response to both small and large perturbations, leading to the concepts of local (linear) and global stability. Here, we show how systems displaying Hopf bifurcations show contrarian results for these two aspects of stability: Global stability is large close to the point where the system loses its stability altogether. We demonstrate this effect for an elementary model system, an anharmonic oscillator, and a realistic model of power system dynamics with delayed control. Detailed investigations of the bifurcation explain the seeming paradox in terms of the location of the attractors relative to the equilibrium.

关键词： Bifurcations Coupled oscillators Synchronization Dynamical systems Nonlinear time-delay systems Bifurcation analysis

来源：评论

学校读者我要写书评

暂无评论

Design and Implementation of Adaptive NMPC in industrial applications: case study

Design and Implementation of Adaptive NMPC in industrial app...

引用

International Carpathian Control Conference (ICCC)

作者： Stepan Ozana Filip Krupa Jakub Nemcik Gergely Takacs Jakub Mozaryn Petr Beremlijski Abhaya Singh Aleksandra Kawala-Sterniuk Department of Cybernetics and Biomedical Engineering VSB-Technical University of Ostrava Ostrava Czech Republic Institute of Automation Measurement and Applied Informatics Slovak University of Technology in Bratislava Bratislava Slovak Republic Institute of Automatic Control and Robotics Warsaw University of Technology Warsaw Poland Department of Applied Mathematics VSB-Technical University of Ostrava Ostrava Czech Republic Department of engineering University of Sannio Sannio Italy Department of Electrical Power and Renewable Energy Opole University of Technology Opole Poland

This paper deals with the design and implementation of adaptive nonlinear predictive control (NMPC). The resulting control system is deployed on a softPLC embedded target in real-time via the REXYGEN control system. The paper describes functional prototype using softPLC hosting the NMPC algorithm applied for a nonlinear system consisting of three coupled tanks. The NMPC algorithm is solved via the receding horizon control method (RHC), formulated as a nonlinear programming problem (NLP) and numerically computed using the SOLNP solver. The adaptivity is provided by online identification via moving horizon estimation (MHE), keeping the model updated to avoid a model-plant mismatch. The methodology described in this paper can be easily transferred to a wider group of industrial nonlinear processes. Such created control unit can run in a standalone mode or can be connected to some exiting control system providing a necessary computational power and thus upgrading the quality of control significantly.

关键词：

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：