检索结果-内蒙古大学图书馆

Islanding Detection in Microgrid Compromising Missing Values Using NI Sensors

IEEE SYSTEMS JOURNAL 2022年第2期16卷 1786-1795页

作者： Kumar, Dhruba Natl Inst Technol Patna Dept Elect Engn Patna 800005 Bihar India

Islanding detection has become an antiquated topic in microgrid, but during the shabby network condition, the conventional islanding detection methodologies become impractical. This article develops a new islanding detection technique considering interrupted communication during islanding detection. This article has experimented a new predictive model (PM)-based method for islanding detection in the laboratory-based multiple distributed generation (DG) system (system with more than one DG). Microphasor measurement units ($\mu $PMUs) are incorporated for measuring phasors from the real-time system. The measurements are analyzed by the artificial neural network (ANN) for initiating PM, which specifies the occurrence of islanding in terms of probability. The following two subalgorithms: evaluation and decoding are simultaneously working to estimate probability of the islanding during sound network condition. The posterior probability evaluation comes into action when communication is interrupted. The accuracy of the algorithm is tested under UL 1741 loading conditions. The performance of the algorithm is analyzed under the simultaneous manifestation of communication interruption and several islanding or nonislanding condition. Nondetection zone has been evaluated for different scenarios of missing frames. The algorithm has also been compared with existing methodologies.

关键词： neural networks Microgrids Decoding Signal processing algorithms Probabilistic logic Prediction algorithms Phasor measurement units Islanding detection microgrid missing data predictive model (PM) statistical method

来源：评论

学校读者我要写书评

暂无评论

DeepTM: Efficient Tensor Management in Heterogeneous Memory for DNN Training

引用

IEEE TRANSACTIONS ON PARALLEL AND distributed SYSTEMS 2024年第11期35卷 1920-1935页

作者： Zhou, Haoran Rang, Wei Chen, Hongyang Zhou, Xiaobo Cheng, Dazhao Wuhan Univ Sch Comp Sci Wuhan 430072 Hubei Peoples R China Univ Macau Lab Internet Things Smart City Taipa 999078 Macau Peoples R China Shandong Normal Univ Sch Informat Sci & Engn Jinan 250098 Shandong Peoples R China Zhejiang Lab Hangzhou 311500 Zhejiang Peoples R China Univ Macau Dept Comp & Informat Sci Taipa 999078 Macau Peoples R China

Deep neural networks (DNNs) have gained widespread adoption in diverse fields, including image classification, object detection, and natural language processing. However, training large-scale DNN models often encounters significant memory bottlenecks, which ask for efficient management of extensive tensors. Heterogeneous memory system, which combines persistent memory (PM) modules with traditional DRAM, offers an economically viable solution to address tensor management challenges during DNN training. However, existing memory management methods on heterogeneous memory systems often lead to low PM access efficiency, low bandwidth utilization, and incomplete analysis of model characteristics. To overcome these hurdles, we introduce an efficient tensor management approach, DeepTM, tailored for heterogeneous memory to alleviate memory bottlenecks during DNN training. DeepTM employs page-level tensor aggregation to enhance PM read and write performance and executes contiguous page migration to increase memory bandwidth. Through an analysis of tensor access patterns and model characteristics, we quantify the overall performance and transform the performance optimization problem into the framework of Integer Linear Programming. Additionally, we achieve tensor heat recognition by dynamically adjusting the weights of four key tensor characteristics and develop a global optimization strategy using Deep Reinforcement Learning. To validate the efficacy of our approach, we implement and evaluate DeepTM, utilizing the TensorFlow framework running on a PM-based heterogeneous memory system. The experimental results demonstrate that DeepTM achieves performance improvements of up to 36% and 49% compared to the current state-of-the-art memory management strategies AutoTM and Sentinel, respectively. Furthermore, our solution reduces the overhead by 18 times and achieves up to 29% cost reduction compared to AutoTM.

关键词： Tensors Memory management Training Bandwidth Random access memory Instruction sets Artificial neural networks Deep neural network training heterogeneous memory memory management performance optimization

来源：评论

学校读者我要写书评

暂无评论

Evaluating a Deep Learning Model for Cyberattack Detection Based on network Traffic 5th

Evaluating a Deep Learning Model for Cyberattack Detection B...

引用

5th International Conference on Cognitive Computing and Information processing, CCIP 2023

作者： Ha, Muon Tran, Duc–Manh Telecommunications University Nha Trang Viet Nam St. Petersburg State Electrotechnical University St. Petersburg Russia

ISBN: (纸本)9783031607240

In the current digital era, distributed Denial of Service (DDoS) attacks can be recognized as a prevalent and perilous network threat, posing substantial risks to both organizations and individuals. Consequently, the timely identification and notification of DDoS attacks play a vital role in their mitigation and reduction of ensuing harm. Long Short-Term Memory (LSTM) networks, a prevalent form of deep neural network, find extensive application in tasks like natural language processing and forecasting time series data. This study suggests the use of LSTM networks to train and categorize network data features, determining their association with cyberattacks. The effectiveness of our approach in detecting and signaling cyberattacks, especially DDoS attacks, is confirmed through experiments on real-world datasets. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Denial-of-service attack

来源：评论

学校读者我要写书评

暂无评论

A Cost-Efficient Failure-Tolerant Scheme for distributed DNN Training 41

A Cost-Efficient Failure-Tolerant Scheme for Distributed DNN...

引用

41st IEEE International Conference on Computer Design (ICCD)

作者： Chen, Menglei Hua, Yu Bai, Rong Huang, Jianming Huazhong Univ Sci & Technol WNLO Wuhan Hubei Peoples R China

ISBN: (纸本)9798350342918

distributed deep neural network (DNN) training is important to support artificial intelligence (AI) applications, such as image classification, natural language processing, and autonomous driving. Unfortunately, the distributed property makes the DNN training vulnerable to system failures. Check-pointing is generally used to support failure tolerance, which however suffers from high runtime overheads. In order to enable high-performance and low-latency checkpointing, we propose a lightweight checkpointing system for distributed DNN training, called LightCheck. To reduce the checkpointing overheads, we leverage fine-grained asynchronous checkpointing by pipelining checkpointing in a layer-wise way. To further decrease the checkpointing latency, we leverage the software-hardware codesign methodology by coalescing new hardware devices into our checkpointing system via a persistent memory (PM) manager. Experimental results on six representative real-world DNN models demonstrate that LightCheck offers more than 10x higher checkpointing frequency with lower runtime overheads than stateof-the-art checkpointing schemes. We have released the opensource codes for public use in https://***/LighT-chenml/ ***.

关键词： Computer systems Checkpointing systems Failure tolerance Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

FCTNet: A CNN-Transformer Hybrid for Single Remote Sensing Image Super-Resolution 22

FCTNet: A CNN-Transformer Hybrid for Single Remote Sensing I...

引用

22nd IEEE International Symposium on Parallel and distributed processing with Applications, ISPA 2024

作者： Shi, Ning Zhou, Hui Ye, Chunyang Yao, Biyuan Wu, Yilin Li, Bo Hainan University Haikou China

ISBN: (纸本)9798331509712

Convolutional neural network (CNN) and Transformer architectures have been extensively applied in the domain of remote sensing image super-resolution. However, to achieve optimal performance, many existing methods are designed with a large number of parameters, thereby increasing the complexity of the model and hindering practical deployment. To address this issue, we propose an innovative model that synergizes CNN and Transformer architectures while employing a minimal number of their respective modules. This approach significantly reduces the parameter count while maintaining superior performance. Firstly, by constructing additional shallow feature representations as input, we enhance the feature extraction capabilities for individual images. Secondly, we utilize residual connections between various modules to integrate multi-scale, high-dimensional feature information, thus ensuring efficient transmission. Finally, the image reconstruction module is employed to restore the high-resolution image. Experimental results show that FCTNet significantly outperforms existing methods while maintaining a substantially lower parameter count, as demonstrated through evaluations on two public datasets. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

CPADP: Novel Processor for Efficient Threat Detection in CPS using RISC and DC-GCNN

CPADP: Novel Processor for Efficient Threat Detection in CPS...

引用

2024 International Conference on distributed Systems, Computer networks and Cybersecurity, ICDSCNC 2024

作者： Kumar, K. Sravan Suresh, G. Penchalaiah, Usthulamuri Sundeep Kumar, K. Prasanthi, E. Geethanjali Institute of Science and Technology Department of Ece S. P. S. R. Nellore India Geethanjali Institute of Science and Technology Dept. of Ece S. P. S. R. Nellore India Geethanjali Institute of Science and Technology Dept. of Cse S. P. S. R. Nellore India

ISBN: (纸本)9798350375442

In the realm of cybersecurity for cyber-physical systems (CPS), the need for robust and efficient intrusion detection mechanisms has become increasingly critical. The existing CPS had limited processing capabilities of traditional detection systems in CPS, and incompatibility of existing IT- centric security solutions with CPS environments. The challenges in real-time threat detection and response due to high data volumes. Vulnerabilities arising from interconnected nature of cyber-physical systems. This paper proposes a novel Cyber-Physical Attack Detection Processor (CPADP) based on Reduced Instruction Set Computing (RISC) architecture. The CPADP integrates advanced techniques for real-time threat detection in CPS environments, leveraging a comprehensive approach spanning data collection, preprocessing, Independent Component Analysis (ICA)-based feature extraction, and a Dual Channel Graph Convolutional neural network (DC-GCNN) for attack classification. The architecture aims to address the challenges posed by CPS environments, where traditional IT- centric security measures often fall short. By harnessing the capabilities of RISC architecture, the CPADP promises enhanced processing efficiency and scalability, crucial for real-time threat mitigation in diverse CPS applications. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Lower Bounds and Nearly Optimal Algorithms in distributed Learning with Communication Compression 36

Lower Bounds and Nearly Optimal Algorithms in Distributed Le...

引用

36th Conference on neural Information processing Systems (NeurIPS)

作者： Huang, Xinmeng Chen, Yiming Yin, Wotao Yuan, Kun Univ Penn Philadelphia PA USA Alibaba Grp DAMO Acad Hangzhou Peoples R China MetaCarbon New York NY USA Peking Univ Beijing Peoples R China

ISBN: (纸本)9781713871088

Recent advances in distributed optimization and learning have shown that communication compression is one of the most effective means of reducing communication. While there have been many results for convergence rates with compressed communication, a lower bound is still missing. Analyses of algorithms with communication compression have identified two abstract properties that guarantee convergence: the unbiased property or the contractive property. They can be applied either unidirectionally (compressing messages from worker to server) or bidirectionally. In the smooth and non-convex stochastic regime, this paper establishes a lower bound for distributed algorithms whether using unbiased or contractive compressors in unidirection or bidirection. To close the gap between this lower bound and the best existing upper bound, we further propose an algorithm, NEOLITHIC, that almost reaches our lower bound (except for a logarithm factor) under mild conditions. Our results also show that using contractive compressors in bidirection can yield iterative methods that converge as fast as those using unbiased compressors unidirectionally. We report experimental results that validate our findings.

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

MPLNet: Industrial Anomaly Detection with Memory Bank and Prompt Learning 22

MPLNet: Industrial Anomaly Detection with Memory Bank and Pr...

引用

22nd IEEE International Symposium on Parallel and distributed processing with Applications, ISPA 2024

作者： Zhou, Hao Guo, Xuanru Ji, Mingcheng Jiang, Jielin Xiang, Haolong Xue, Shengjun University of Science and Technology Beijing Beijin China Nanjing University of Information Science and Technology Nanjing China Jiangsu Province Engineering Research Center of Advanced Computing and Intelligent Services China Silicon Lake College Suzhou China

ISBN: (纸本)9798331509712

Anomaly detection is a critical aspect of industrial production processes. Most of the self-supervised training system utilise Convolutional neural networks (CNNs) for feature extraction. However, these methods often exhibit poor detection performance on small structural anomalies due to the limited receptive field of CNN. The prompt learning method has been demonstrated to be an effective means of enhancing the global information extraction abilities of the model. However, this approach necessitates the input of a significant number of manual prompts. To address these issues, we propose a Memory And Prompt Learning Based network (MPLNet) for anomaly detection. MPLNet obtains the differences information between normal and detected images by comparing their features and then uses these differences to generate prompts. By automatically generating prompts, it reduces the workload of manually entering prompts in prompt learning. At the same time, it effectively enhances the model's ability to extract and use global information. Extensive experiments have shown that the proposed MPLNet achieves state-of-the-art anomaly detection performance on the widely used and challenging MVTec AD dataset and MVTec AD-3D dataset. Index Terms - Anomaly detection, U-Net, Transformer, Prompt, artificial intelligence System © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

EdgeShield: Hybrid Real-Time Attack Detection in IOTA Tangle via Edge Devices 15

EdgeShield: Hybrid Real-Time Attack Detection in IOTA Tangle...

引用

15th International Conference on Computing Communication and networking Technologies, ICCCNT 2024

作者： Arunodhai, V. Susan, Leena P. Kannimoola, Jinesh M. Amrita Vishwa Vidyapeetham Amrita School of Computing Department of Computer Science and Applications Amritapuri India

ISBN: (纸本)9798350370249

The Internet of Things (IoT) revolutionizes technology interaction by enabling seamless connectivity and automation, but it introduces significant security challenges. A distributed ledger technology (DLT) called IOTA Tangle, a utilizing a Directed Acyclic Graph (DAG), enables secure and feeless transactions among IoT devices. Despite its advantages, the IOTA Tangle faces threats such as replay attacks and double spending, which highlight vulnerabilities in distributed ledger technologies. To address these security concerns, we initially developed EdgeGuard, a rule-based system for spam detection within the IOTA ecosystem. However, its reliance on predetermined rulesets limited its adaptability. We have now enhanced our approach with EdgeShield, incorporating a neural network for superior threat *** enhances adaptability over EdgeGuard in threat detection for the IOTA Tangle by integrating machine learning. EdgeShield analyzes network packets to identify threats, leveraging machine learning to adapt to evolving risks. Trained using the MQTTSet dataset, it detects network-based threats such as denial-of-service (DoS) attacks and packet flooding. Tested with simulated IoT data, EdgeShield demonstrates effectiveness across various domains, particularly in healthcare IoT. Its decentralized processing on edge devices like the Raspberry Pi 4 enables immediate network traffic analysis, reducing reliance on centralized servers and improving system resilience. Additionally, real-time alerts empower administrators to proactively secure IoT deployments and maintain transaction integrity within the IOTA Tangle. This paper traces our research evolution from EdgeGuard to EdgeShield, highlighting advancements and demonstrating the enhanced security of IoT transactions on the IOTA Tangle. © 2024 IEEE.

关键词： and Decentralised Security Measures distributed Ledger Edge Computing Edge Devices Internet of Things (IoT) IOTA Tangle Machine Learning network Packet Analysis Real-Time Security

来源：评论

学校读者我要写书评

暂无评论

Predicting Temperature of Molten Steel in LF-Refining Process Using IF-ZCA-DNN Model

引用

METALLURGICAL AND MATERIALS TRANSACTIONS B-PROCESS METALLURGY AND MATERIALS processing SCIENCE 2023年第3期54卷 1181-1194页

作者： Xin, Zi-Cheng Zhang, Jiang-Shan Zhang, Jun-Guo Zheng, Jin Jin, Yu Liu, Qing Univ Sci & Technol Beijing State Key Lab Adv Met Beijing Peoples R China Hebei Iron & Steel Co Ltd Tangshan Branch Tangshan Hebei Peoples R China

Controlling the temperature of molten steel in ladle furnace (LF)-refining process is one of the main tasks to ensure that the steelmaking-continuous casting process runs smoothly. In this work, a hybrid model based on metallurgical mechanism, isolation forest (IF), zero-phase component analysis whitening (ZCA whitening), and a deep neural network (DNN) was established to predict the temperature of molten steel in LF-refining process. The metallurgical mechanism, Pearson correlation coefficient, ZCA whitening, IF, and t-distributed stochastic neighbor embedding (t-SNE) were used to obtain the main factors affecting the temperature, analyze the correlation between two random variables, eliminate the correlation among the input variables, reduce the abnormal data of the original datasets, and visualize high-dimensional data, respectively. The single-machine-learning (ML) models, ZCA-ML models, and IF-ZCA-DNN model were comparatively examined by evaluating the coefficient of determination (R-2), root-mean-square error (RMSE), mean absolute error (MAE), and hit ratio. The optimal structure of IF-ZCA-DNN model had 4 hidden layers, 45 hidden layer neurons, a learning rate of 0.03, regularization coefficient of 2 x 10(-4), batch size of 128, leaky-rectified linear unit activation function, and an optimization algorithm of mini-batch stochastic gradient descent with momentum. The R-2, RMSE, and MAE of the IF-ZCA-DNN model were 0.916, 2.827, and 2.048, respectively. Meanwhile, the prediction hit ratio for the temperature of IF-ZCA-DNN model in the error ranges of [- 3, 3], [- 5, 5], and [- 10, 10] were 77.9, 92.3, and 99.6 pct, respectively. This study will be beneficial to realize precise control of temperature of molten steel in LF-refining process.

关键词： neural network models

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：