检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Najafirad, Mohammad Javad Welikala, Shirantha Department of Electrical and Computer Engineering School of Engineering and Science Stevens Institute of Technology HobokenNJ07030 United States

This paper presents a novel dissipativity-based distributed droop-free control and communication topology co-design approach for voltage regulation and current sharing in DC microgrids (DC MGs) with generic "ZIP" (constant impedance (Z), current (I) and power (P)) loads. While ZIP loads accurately capture the varied nature of the consumer loads, its constant power load (CPL) component is particularly challenging (and destabilizing) due to its non-linear form. Moreover, ensuring simultaneous voltage regulation and current sharing and co-designing controllers and topology are also challenging when designing control solutions for DC MGs. To address these three challenges, we model the DC MG as a networked system comprised of distributed generators (DGs), ZIP loads, and lines interconnected according to a static interconnection matrix. Next, we equip each DG with a local controller and a distributed global controller (over an arbitrary topology) to derive the error dynamic model of the DC MG as a networked "error" system, including disturbance inputs and performance outputs. Subsequently, to co-design the controllers and the topology ensuring robust (dissipative) voltage regulation and current sharing performance, we use the dissipativity and sector boundedness properties of the involved subsystems and formulate Linear Matrix Inequality (LMI) problems to be solved locally and globally. To support the feasibility of the global LMI problem, we identify and embed several crucial necessary conditions in the corresponding local LMI problems, thus providing a one-shot approach (as opposed to iterative schemes) to solve the LMI problems. Overall, the proposed approach in this paper provides a unified framework for designing DC MGs. The effectiveness of the proposed solution was verified by simulating an islanded DC MG under different scenarios, demonstrating superior performance compared to traditional control approaches. Copyright © 2025, The Authors. All rights reserved.

关键词： Networked control systems

来源：评论

学校读者我要写书评

暂无评论

Crowd-Counting of Images in Rainy Weather Based on GAN-U-Net Model and Multi-column CNN Counting Approaches 3

Crowd-Counting of Images in Rainy Weather Based on GAN-U-Net...

引用

3rd IEEE International Conference on Electronic engineering, ICEEM 2023

作者： Saad, Rasha A. Abd El-Samie, Fathi E. Abdel-Azim, Mohammed El-Hoseny, Heba M. Obour Higher Institute for Engineering and Technology Department of Electronics and Electrical Communications Egypt Menoufia University Faculty of Electronic Engineering Department of Electronics and Electrical Communications Egypt Mansoura University Faculty of Engineering Department of Electronics and Electrical Communications Egypt The Higher Future Institute for Specialized Technological Studies Department of Computer Science Egypt

ISBN: (纸本)9798350323511

A count of crowds tries to determine the total number of persons in crowded places in order to avoid disruption of the safety system and maintain crowd safety. Accurate estimation of the size of crowds is a challenging process because of numerous challenges like occlusion, complicated backgrounds, varying sizes, uneven distribution, alteration of lighting, and adaptable weather. In this study, we propose a solution to two crowd counting challenges: poor weather such as rain and image scaling. We investigate crowd counting in the rainy Shanghai-Tec images. We adopt the GAN-U-Net de-raining model and the SAS-Net [1] and SGA-Net [2] crowd-counting models in our study. © 2023 IEEE.

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

Dissipativity-Based Distributed Control and Communication Topology Co-Design for Voltage Regulation and Current Sharing in DC Microgrids

arXiv

引用

arXiv 2025年

This paper presents a novel dissipativity-based distributed droop-free control approach for voltage regulation and current sharing in DC microgrids (MGs) comprised of an interconnected set of distributed generators (DGs), loads, and power lines. First, we describe the closed-loop DC MG as a networked system where the DGs and lines (i.e., subsystems) are interconnected via a static interconnection matrix. This interconnection matrix demonstrates how the inputs, outputs, and disturbances of DGs and lines are connected in a DC MG. Each DG has a local controller and a distributed global controller. To design the controllers, we use the dissipativity properties of the subsystems and formulate a linear matrix inequality (LMI) problem. To support the feasibility of this problem, we identify a set of necessary local and global conditions that we then enforce in a specifically developed LMI-based local controller design process. In contrast to existing DC MG control solutions, our approach proposes a unified framework for co-designing the distributed controller and communication topology. As the co-design process is LMI-based, it can be efficiently implemented and evaluated using existing convex optimization tools. The effectiveness of the proposed solution was verified by simulating an islanded DC MG in a MATLAB/Simulink environment under different scenarios, such as load changes and topological constraint changes, and then comparing the performance with a recent droop control algorithm. Copyright © 2025, The Authors. All rights reserved.

关键词： Linear matrix inequalities

来源：评论

学校读者我要写书评

暂无评论

Steganographic Encryption and Decryption Using Duality Preserving GANs 22

Steganographic Encryption and Decryption Using Duality Prese...

引用

22nd IEEE International Conference on Industrial Informatics, INDIN 2024

作者： Shah, Aaryak Kamal, Alok Kumar Neeraj Dept. of Information Technology ABV-IIITM Gwalior Gwalior India Dept. of Electrical/Electronics ABV-IIITM Gwalior Gwalior India Dept. of Computer Science and Engineering Mahindra University Hyderabad India

ISBN: (纸本)9798331527471

Image steganography has been a significant research area in the security sphere. The core idea revolves around embedding secret messages in existing images (covers) such that the resultant image displays no noticeable detectable artifacts of the secret while also being statistically similar to the original cover. Various steganographic techniques have been developed and tested. However, these traditional techniques often exploit some specific property of the image and are often easy to detect with the right tools. In recent times, generative adversarial networks, or GANs for short, have shown massive promise in their capability to modify and synthesize data. They are now ubiquitous in image synthesis and image translation tasks. With various recent advancements in this technology allowing these models to generate, in many cases, human-imperceptible data, GAN-based steganographic systems have become a significant focus of security research. This report analyzes and compares some existing models and attempts to provide a new solution that performs better on multiple well-defined metrics. A DualGAN-based architecture is proposed that trains both the encoder and decoder processes simultaneously while maintaining the underlying dual relationship of these tasks. The proposed scheme reports better results on the selected metrics (SBR, DER, PSNR) at three values of embedding capacity set at 0.3, 0.6, and 0.9. © 2024 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Learning good state and action representations for Markov decision process via tensor decomposition

The Journal of Machine Learning Research

引用

The Journal of Machine Learning Research 2023年第1期24卷 5157-5209页

作者： Chengzhuo Ni Yaqi Duan Munther Dahleh Mengdi Wang Anru R. Zhang Department of Electrical & Computer Engineering Princeton University Laboratory for Information & Decision Systems Massachusetts Institute of Technology Electrical Engineering & Computer Science Department Massachusetts Institute of Technology Department of Biostatistics & Bioinformatics Duke University

The transition kernel of a continuous-state-action Markov decision process (MDP) admits a natural tensor structure. This paper proposes a tensor-inspired unsupervised learning method to identify meaningful low-dimensional state and action representations from empirical trajectories. The method exploits the MDP's tensor structure by kernelization, importance sampling and low-Tucker-rank approximation. This method can be further used to cluster states and actions respectively and find the best discrete MDP abstraction. We provide sharp statistical error bounds for tensor concentration and the preservation of diffusion distance after embedding. We further prove that the learned state/action abstractions provide accurate approximations to latent block structures if they exist, enabling function approximation in downstream tasks such as policy evaluation.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Semantic-Consistent Conditional Generative Adversarial Model for Zero-Shot Learning in Bearing Fault Diagnosis 13

A Semantic-Consistent Conditional Generative Adversarial Mod...

引用

13th IEEE Data Driven Control and Learning Systems Conference, DDCLS 2024

作者： Chen, Jizheng Yang, Rui School of Advanced Technology Xi'an Jiaotong-Liverpool University Suzhou215123 China School of Electrical Engineering Electronics and Computer Science University of Liverpool LiverpoolL69 3BX United Kingdom

ISBN: (纸本)9798350361674

The scarcity of labeled data remains one of the most significant challenges in industrial fault diagnosis, where the industrial processes lack readily sampled in many fault types. This issue is often called the zero-shot learning problem in industrial fault diagnosis. To address this challenge, this paper proposes a novel fault diagnosis model for rotating machinery, leveraging Generative Adversarial Networks (GANs) and semantic consistent loss. Our approach aims to generate a rich set of samples for fault types that are difficult to collect, thereby enabling the training of an accurate classification model. The proposed model introduces a one-dimensional convolutional neural network to enhance the quality of samples and the accuracy of the model. A semantic consistent discriminator is proposed to ensure that the distribution of generated samples closely aligns with the distribution of actual samples. The performance of the model is validated using the CWRU bearing fault dataset, and the results demonstrate its efficacy in addressing the zero-shot fault diagnosis challenge. © 2024 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Classification of Sleep-Related Disorders Using Machine Learning Techniques 25

Classification of Sleep-Related Disorders Using Machine Lear...

引用

25th International Arab Conference on Information technology, ACIT 2024

作者： Nasir, Ola Alazaidah, Raed Al-Safarini, Maram Y. Alamleh, Amneh Zaid, Ali Abu Mansur, Hasan Elrashidi, Ali Alzoubi, Haneen Bani-Bakr, Alaa Zarqa University Faculty of Information Technology Department of Computer Science Zarqa Jordan Zarqa University Faculty of Information Technology Department of Data Science and Ai Zarqa Jordan Zarqa University Faculty of Information Technology Department of Cyber Security Zarqa Jordan American University of Madaba Faculty of Business Department of Accounting Madaba Jordan University of Business and Technology Department of Electrical Engineering Jeddah21432 Saudi Arabia Private University Faculty of It Applied Science Department of Computer Science Amman Jordan

ISBN: (纸本)9798331540012

Several people suffer from irregular sleep patterns, a problem that negatively affects their daily lives. It presents a multitude of challenges, including daytime distress, disturbances in sleep-wake patterns, feelings of unease, and a variety of other complications. Diabetes, obesity, and psychological disorders are among the diseases that may be responsible for this sleep disorder. Therefore, the primary aim of this research is to employ the robust capabilities of machine learning in forecasting sleep disorders. This research classifies three diseases: diabetes, obesity, and psychological disorders, using thirty-two different classifiers belonging to five learning strategies. The research compares and evaluates their predictive performance on four datasets. The comparative analysis includes four evaluation metrics: Accuracy, Precision, ROC Area, and MCC. The research results indicate that the best classifier is SGD, followed by LMT, while NBM is the weakest classifier. In addition, the research results revealed that the best strategy is Functions, followed by Rules, with Bayes being the least effective strategy. While conducting the experiment and analyzing the data, it is observed that the best feature selection method that yielded the most favorable outcome is the CorrelationAttributeEval. Furthermore, it has been found that the classification algorithm known as SGD achieved the highest Accuracy and Precision in categorizing the data. This indicates that the combination of CorrelationAttributeEval as the feature selection approach and SGD as the classification algorithm proved to be the most successful and efficient strategy for this research © 2024 IEEE.

关键词： Sleep research

来源：评论

学校读者我要写书评

暂无评论

R-IDPS: Real Time SDN-Based IDPS System for IoT Security

引用

computers, Materials & Continua 2022年第11期73卷 3099-3118页

作者： Noman Mazhar Rosli Saleh Reza Zaba Muhammad Zeeshan M.Muzaffar Hameed Nauman Khan Faculty of Computer Science and Information Technology University of MalayaKuala Lumpur50603Malaysia Centre for Research in Industry 4.0 University of MalayaKuala Lumpur50603Malaysia MIMOS Berhad National Applied R&D CentreKuala Lumpur57000Malaysia School of Electrical Engineering and Computer Science National University of Sciences and TechnologyIslamabad44000Pakistan

The advent of the latest technologies like the Internet of things(IoT)transforms the world from a manual to an automated way of ***,IoT sector open numerous security *** traditional networks,intrusion detection and prevention systems(IDPS)have been the key player in the market to ensure *** challenges to the conventional IDPS are implementation cost,computing power,processing delay,and ***,online machine learning model training has been an *** these challenges still question the IoT network *** has been a lot of research for IoT based detection systems to secure the IoT devices such as centralized and distributed architecture-based detection *** centralized system has issues like a single point of failure and load balancing while distributed system design has scalability and heterogeneity *** this study,we design and develop an agent-based hybrid prevention system based on software-defined networking(SDN)*** system uses lite weight agents with the ability to scaleup for bigger networks and is feasible for heterogeneous IoT *** baseline profile for the IoT devices has been developed by analyzing network flows from all the IoT *** profile helps in extracting IoT device *** features help in the development of our dataset that we use for anomaly *** anomaly detection,support vector machine has been used to detect internet control message protocol(ICMP)flood and transmission control protocol synchronize(TCP SYN)flood *** proposed system based on machine learning model is fully capable of online and offline *** than detection accuracy,the system can fully mitigate the attacks using the software-defined technology SDN *** major goal of the research is to analyze the accuracy of the hybrid agent-based intrusion detection systems as compared to conventional centralized only solutions,especially under the flood attac

关键词： Machine learning Internet of things software defined networking distributed denial of service attacks

来源：评论

学校读者我要写书评

暂无评论

Quickest Detection of Adversarial Attacks Against Correlated Equilibria 39

Quickest Detection of Adversarial Attacks Against Correlated...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Kazari, Kiarash Kanellopoulos, Aris Dán, György Division of Network and Systems Engineering School of Electrical Engineering and Computer Science KTH Royal Institute of Technology Stockholm Sweden

ISBN: (纸本)157735897X

We consider correlated equilibria in strategic games in an adversarial environment, where an adversary can compromise the public signal used by the players for choosing their strategies, while players aim at detecting a potential attack as soon as possible to avoid loss of utility. We model the interaction between the adversary and the players as a zero-sum game and we derive the maxmin strategies for both the defender and the attacker using the framework of quickest change detection. We define a class of adversarial strategies that achieve the optimal trade-off between attack impact and attack detectability and show that a generalized CUSUM scheme is asymptotically optimal for the detection of the attacks. Our numerical results on the Sioux-Falls benchmark traffic routing game show that the proposed detection scheme can effectively limit the utility loss by a potential adversary. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

AdaFedProx: A Heterogeneity-Aware Federated Deep Reinforcement Learning for Medical Image Classification

引用

IEEE Transactions on Consumer Electronics 2024年

作者： Sahoo, Pranab Tripathi, Ashutosh Saha, Sriparna Mondal, Samrat Singh, Jyoti Prakash Sharma, Bhisham Computer Science and Engineering Department Indian Institute of Technology Patna India Electrical and Electronics Engineering Department The Rajiv Gandhi Institute of Petroleum Technology India Indian Institute of Technology Computer Science and Engineering Department Patna India National Institute of Technology Computer Science and Engineering Department Patna India Chitkara University Centre of Research Impact and Outcome Punjab Rajpura140401 India

In the realm of smart healthcare, vast amounts of valuable patient data are generated worldwide. However, healthcare providers face challenges in data sharing due to privacy concerns. Federated learning (FL) offers a privacy-preserving solution by enabling collaborative model training without direct access to patient data. This decentralized approach utilizes data from diverse sources, resulting in a globally learned model with satisfactory performance across individual sites. However, federated training encounters challenges such as "system heterogeneity"in client specifications and "data heterogeneity"in imbalanced data distribution. Standard FL methods become unstable and require extensive hyperparameter tuning for optimal performance when facing heterogeneous clients, posing challenges in real-world applications. This research introduces a novel heterogeneity-aware federated learning approach named AdaFedProx to address the performance degradation resulting from heterogeneity. AdaFedProx utilizes a reinforcement learning-driven strategy, introducing a tailored proximal term into the objective function to incorporate partial client information in heterogeneous settings. Dynamically determining the proximal term based on individual client states, including critical characteristics like data distribution, system specification, and performance feedback, empowers AdaFedProx to make regularized decisions throughout the federated training. AdaFedProx consistently outperforms four state-of-the-art methods on three medical datasets across diverse heterogeneity settings (0%, 50%, and 75% straggler). The code can be found at https://***/Pranabiitp/AdaFedProx. © 1975-2011 IEEE.

关键词： Deep reinforcement learning

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：