检索结果-内蒙古大学图书馆

22nd International conference on control, Automation and systems (ICCAS)

作者： Jang, Hee-Deok Park, Jae-Hyeon Nam, Hyunwoo Chang, Dong Eui Korea Adavanced Inst Sci & Technol Sch Elect Engn Daejeon 34141 South Korea Agcy Def Dev Chem Bio Technol Ctr Daejeon 34188 South Korea

ISBN: (纸本)9788993215243

Many studies propose gas concentration estimators using machine learning algorithms owing to their high performance. Recently, estimation models using deep neural network have been studied due to their higher performance than conventional machine learning algorithms. The performance of deep neural network can be increased by hyperparameter optimization. In this paper, we propose two deep neural networks for gas concentration estimation and analyze how hyperparameter optimization affects the performance of the proposed deep neural networks. We optimize the hyperparameters of the proposed neural networks and compare the performance with conventional machine learning models. We train the proposed neural networks and evaluate the performance of the models with an open dataset. We confirm that the optimized neural network models show the high performance in gas concentration estimation, and that models using unoptimized parameters may show worse performance than conventional machine learning model.

关键词： gas concentration estimation deep neural network hyperparameter optimization multilayer perceptron convolutional neural network

来源：评论

学校读者我要写书评

暂无评论

Detecting Cyber Attacks in Industrial control systems Using Spatio-Temporal Autoencoder

Detecting Cyber Attacks in Industrial Control Systems Using ...

引用

International Joint conference on Neural networks (IJCNN)

作者： Lan, Bin Yu, Shunzheng Sun Yat Sen Univ Sch Comp Guangzhou Peoples R China

ISBN: (纸本)9781665488679

Industrial control systems (ICSs) are widely used in various industries. These systems have become prime targets for cyber and physical attacks. The attacks, which have an impact on the physical processes of an ICS, often lead to system misbehaviors through data contamination. Recent research has shown that network-based detection methods cannot monitor the physical level activities well enough to mitigate hybrid cyber attacks and cannot entirely protect ICSs. To protect ICSs from such threats, we propose a Spatio-Temporal Autoencoder (STA) with a Dynamic Thresholding Mechanism. The STA learns the normal physical behaviors of the system by capturing deep spatio-temporal dependencies to form a unified representation of the system state. The unified representation is decoded to reconstruct the input features. Then, the dynamic threshold is used to detect and locate the anomalies. We validate the STA using data set from a real water treatment plant testbed, SWaT. Evaluation results indicate the superior performance of the STA compared with six state-of-the-art methods, achieving an average improvement of 5.8% in the F-Score.

关键词： anomaly detection industrial control systems spatio-temporal dependencies deep learning dynamic thresholding

来源：评论

学校读者我要写书评

暂无评论

Mutual Learning in Optimization - Part ii

Mutual Learning in Optimization - Part II

引用

American control conference (ACC)

作者： Narendra, Kumpati Zheng, Lihao Mukhopadhyay, Snehasis Yale Univ Ctr Syst Sci New Haven CT 06521 USA IUPUI Comp & Informat Sci Indianapolis IN 46202 USA

ISBN: (纸本)9798350382662;9798350382655

This is part ii of a multi-part paper on "Mutual Learning in Optimization". In part I of the paper [1] two agents attempting jointly to determine the optimum of a function f[x(1), x(2),..., x(n)] of n variables, use the same or different optimization techniques (well known in the literature). The agents are assumed to communicate with each other at random instants of time, and convey to the other all the information that they possess which may prove useful. It was shown through simple examples depending upon the initial conditions, the number of steps taken, and the methods used by the agents that totally different decisions may be made by the participants at every stage. At the same time, in all cases, the agents improve their performance and achieve the optimum asymptotically. In this paper the problem of determining the maximum of a function f[x(1), x(2),..., x(n)] of n variables is considered. The number of agents is assumed to be N, and all three cases N > n + 1, N = n + 1 and N < n + 1 are considered. For simplicity, the case when f(center dot) is a quadratic function is considered first and analytic solutions are derived. Following this, non-quadratic functions are considered, and the modifications in the methods are discussed. Simulation results are given for second and fourth order systems.

关键词： Optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

control Enhancement of Traction Electric Drives Using Neural network Predictive controller

Control Enhancement of Traction Electric Drives Using Neural...

引用

2024 International conference on control, Automation and Diagnosis, ICCAD 2024

作者： Cocogne, Romain Bilavarn, Sebastien Mokadem, Mostafa El Douzane, Khaled Silicon Mobility Sophia Antipolis France Université Côte d'Azur Cnrs Leat Sophia Antipolis France

ISBN: (纸本)9798350361025

This paper investigates the use of a hybrid Recurrent Neural network to reproduce the behavior of a nonlinear long-horizon Model Predictive controller (MPC) used in traction motor drive systems. The goal is to assess the operational validity and control performances of such neural network based predictive controller (further referred to as Full Neural network MPC), and to compare against Field Oriented control (FOC), the current industry standard. Based on simulation results using a Model in the Loop (MiL) environment, it is shown that the proposed FNN-MPC can properly learn the behavior and characteristics of the nonlinear long-horizon MPC while being more computationally implementable, with still better control performance and quality than FOC. © 2024 IEEE.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Design of Industrial Automation network Based on TSN and 5G Slice

Design of Industrial Automation Network Based on TSN and 5G ...

引用

2024 International conference on Power, Electrical Engineering, Electronics and control, PEEEC 2024

作者： Jin, Bo Chen, Jialin Zhou, Zheng Yu, Zheng Zhan, Peng Hou, Dai State Grid Hubei Information & Telecommunication Company Limited Hubei Wuhan430000 China

ISBN: (纸本)9798350378917

In this paper, a new industrial automation network design scheme combining time-sensitive network (TSN) and 5G slicing technology is proposed. This scheme aims to solve the shortcomings of traditional network technology in real-time, reliability and flexibility, so as to meet the needs of modern industrial production. By constructing a network architecture consisting of data acquisition layer, data transmission layer, data processing and control layer, application and service layer and management and security layer, the design can ensure the accuracy and real-time performance of data acquisition, and meet different business requirements through 5 G slicing technology, such as real-time control, monitoring data upload, remote maintenance and so on. In addition, the scheme also adopts intelligent traffic scheduling algorithm and end-to-end security measures to improve the efficiency and security of data transmission. The experimental results show that the design scheme is superior to the traditional network scheme in real-time, reliability and security, especially in the high load and complex network environment, showing higher stability and efficiency. Therefore, this scheme provides an efficient, stable and safe network communication solution for the industrial automation field. © 2024 IEEE.

关键词： 5G mobile communication systems

来源：评论

学校读者我要写书评

暂无评论

A Novel Hybrid control Strategy for Grid-Connected PV-Battery systems under Adverse Environmental Conditions 13

A Novel Hybrid Control Strategy for Grid-Connected PV-Batter...

引用

13th IEEE International conference on Communication systems and network Technologies, CSNT 2024

作者： Pandey, Pawan Kumar Gyan Ganga Institute of Technology and Science Dept. of Electrical Engineering Jabalpur India

ISBN: (纸本)9798350305463

This work presents a new controller for gridconnected PV/Battery systems that combines a bidirectional battery controller with a voltage source converter (VSC) to solve problems caused by power fluctuations on the DC-link. The hybrid VSC controller is developed to maintain a continuous power supply on the DC-link. It does this by responding to changes in environmental factors and load fluctuations in the PV-generated output, which in turn triggers dynamic adjustments to the inverter pulse signals. To increase system stability, the bidirectional battery controller orchestrates the charge and discharge operations in collaboration with the hybrid VSC, optimizing power delivery to the grid. Through a detailed performance study of the proposed controller, faults on both the DC and AC sides of the system are analyzed and efficiently addressed. The controller proved important in enhancing power quality and decreasing harmonic distribution. Furthermore, the hybrid controller's performance is assessed under several one-day environmental situations. Ultimately, the results of simulation done in MATLAB/Simulink corroborate the usefulness of the suggested controller in providing optimal performance in the described grid-connected PV/Battery system. © 2024 IEEE.

关键词： controllers

来源：评论

学校读者我要写书评

暂无评论

SDN Enabled L2 Switch Implementation and Its performance Evaluation Through P4 Programming 1

引用

20th International conference on Computing and Information Technology (IC2IT2)

作者： Pudasaini, Sakar Dawadi, Babu R. Ghimire, Roshani Pau, Giovanni Sapkota, Binod Tribuvan Univ Inst Engn Pulchowk Campus Kathmandu Nepal Tribhuvan Univ Adv Coll Engn & Management Kathmandu Nepal Kore Univ Enna Fac Engn & Architecture Enna Italy

ISBN: (数字)9783031585616

ISBN: (纸本)9783031585609;9783031585616

The concept of SDN (Software-Defined networking) originally came with the purpose of disaggregating control plane and data plane. However, current scope of SDN has surpassed the original goal. It sets out to introduce programmability in the control plane for better visibility and manageability of networking devices. The idea has been recently extended to program the data plane. In the present scenario P4 (Programming Protocol-Independent Packet Processors) stands as the de-facto language to program the data plane. In this study, we implemented an L2 (Layer 2) switch and evaluated it performance using P4 programming language. The performance of the switch was evaluated in terms of throughput (Mbps) and latency (ms). The throughput was found to be nominal as the switch was implemented in software and the latency for first packet was also high since the switch's table was empty. However, once the table was filled, subsequent packets suffered very low latency.

关键词： SDN Programmable Data Plane P4 Data Plane control Plane network Programmability

来源：评论

学校读者我要写书评

暂无评论

TinyLid: a RISC-V accelerated Neural network For LiDAR Contaminant Classification in Autonomous Vehicle 21

TinyLid: a RISC-V accelerated Neural Network For LiDAR Conta...

引用

21st ACM International conference on Computing Frontiers (CF)

作者： Jati, Grafika Molan, Martin Barchi, Francesco Bartolini, Andrea Acquaviva, Andrea Univ Bologna DEI Dept Bologna Italy

ISBN: (纸本)9798400705977

LiDAR plays a critical role in autonomous car perception. Hence, the robustness of LiDAR data is imperative. However, malfunctions resulting from sensor cover contaminants are unavoidable and can lead to erroneous data that slowly degrade performance. As such, detecting contamination in LiDAR is essential but remains an open challenge due to varying contaminant types, changing properties over time, and deployment aspects. Automatic classification of the contaminants would enable the automated response (like cleaning the sensor) to ensure the integrity of the data collected by the LiDAR sensor. To minimize the effect on the whole vehicle perception system, the contamination classification has to be performed near the sensor and in a computationally efficient way. To address these challenges, we have conducted a feasibility study of developing an efficient near-sensor machine learning-powered contaminant classification running on the RISC-V architecture. This paper proposes a lightweight 2D CNN network, TinyLid, trained to classify contaminants based on the most comprehensive LiDAR contaminant dataset. The results presented in this paper show that the proposed solution can achieve high classification performance while being computationally efficient and running on hardware with negligible power consumption compared to the LiDAR sensor itself. Specifically, implementing a proposed ML model on a reference RISC-V architecture GAP8 achieves the inference time of 2.575 milliseconds, 6.138 operations/cycle, and uses only 6.8% of 512 KiB L2 memory. The results presented in this work showcase the possibility of increasing the reliability and integrity of the LiDAR-collected sensor data without significant computational or energy consumption impact on the broader system.

关键词： Advanced driver assistance systems (ADAS) autonomous vehicle perception systems LiDAR reliability LiDAR contamination classification autonomous vehicle safety RISC-V GAP8 TinyML

来源：评论

学校读者我要写书评

暂无评论

Surveying Lightweight Neural network Architectures for Enhanced Mobile performance 5th

Surveying Lightweight Neural Network Architectures for Enhan...

引用

5th International conference on Smart Applications and Data Analysis for Smart Cyber Physical systems

作者： Briouya, Hasnae Briouya, Asmae Choukri, Ali Ibn Tofail Univ Comp Sci Lab Fac Sci Kenitra Morocco

ISBN: (纸本)9783031770425;9783031770432

This study explores the creation of lightweight neural network architectures specifically designed for mobile devices. The main focus is on MobileNet, but we also examine other models like ShuffleNet, SqueezeNet, EfficientNet, MnasNet, and NASNet Mobile. The main challenge is to carefully evaluate these models to find the best balance between simplicity, accuracy, and efficiency, considering the diverse needs of mobile applications. We assess important metrics such as simplicity, accuracy, and efficiency within these architectures. The main goal is to provide guidance to practitioners in choosing the most suitable architecture. We offer insights into the trade-offs and advantages of each model through both quantitative and qualitative assessments. We consider factors like computational resources, accuracy requirements, and processing speed. The findings of this research provide valuable insights for practitioners who want to make informed decisions about the best neural network architecture for mobile devices. This guidance is tailored to their specific computational limitations and application requirements, helping them make strategic decisions in this specialized field.

关键词： MobileNet Lightweight neural networks Computational efficiency Semantic segmentation Mobile applications

来源：评论

学校读者我要写书评

暂无评论

control Theory-Inspired Model Design for Single Image De-Raining

引用

IEEE TRANSACTIONS ON CIRCUITS AND systems ii-EXPRESS BRIEFS 2022年第2期69卷 649-653页

作者： Zhou, Man Wang, Rujing Chinese Acad Sci Inst Intelligent Machines Intelligent Agr Engn Lab Anhui Prov Hefei 230031 Peoples R China Chinese Acad Sci Hefei Inst Phys Sci Hefei 230031 Peoples R China Univ Sci & Technol China Dept Automat Hefei 230026 Peoples R China

While existing image de-raining methods have obtained satisfactory performance, lacking theoretical basis prevents them from further improvement. In this brief, we propose an efficient image de-raining network empowered by control theory to address this issue. To the best of our knowledge, it is the first attempt to integrate the control theory into image de-raining model design. Different from previous methods that roughly construct complicated neural networks, our method is theoretical and provides a new perspective for model design. Specifically, by mimicking the signal processing flow of state observable and controllable standard form, we expand them to two network modules, named C-IM and O-IM with every component in the module one-to-one corresponding to each operation involved in state equation. Equipped with C-IM and O-IM, our proposed network could efficiently explore and exploit the features of rain streaks in a recursive fashion. Extensive experiments demonstrate that control theory equipped method is capable of obtaining promising performance and speeding up the model training.

关键词： Mathematical model control theory Training Rain Circuits and systems Convolution Feature extraction control theory single image de-raining neural network

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：