检索结果-内蒙古大学图书馆

SOPA-GA-CNN: Synchronous optimisation of parameters and architectures by genetic algorithms with convolutional neural network blocks for securing Industrial Internet-of-Things

引用

IET Cyber-systems and Robotics 2023年第1期5卷 51-66页

作者： Jia-Cheng Huang Guo-Qiang Zeng Guang-Gang Geng Jian Weng Kang-Di Lu College of Cyber Security and the National Joint Engineering Research Center of Network Security Detection and Protection Technology Jinan UniversityGuangzhouChina National-Local Joint Engineering Research Center of Digitalize Electrical Design Technology Wenzhou UniversityWenzhouChina National Laboratory of Industrial Control Technology Institute of Cyber-Systems and ControlZhejiang UniversityHangzhouChina

In recent years,deep learning has been applied to a variety of scenarios in Industrial Internet of Things(IIoT),including enhancing the security of ***,the existing deep learning methods utilised in IIoT security are manually designed by heavily relying on the experience of the *** authors have made the first contribution concerning the joint optimisation of neural architecture search and hyper-parameters optimisation for securing IIoT.A novel automated deep learning method called synchronous optimisation of parameters and architectures by GA with CNN blocks(SOPA-GA-CNN)is proposed to synchronously optimise the hyperparameters and block-based architectures in convolutional neural networks(CNNs)by genetic algorithms(GA)for the intrusion detection issue of *** efficient hybrid encoding strategy and the corresponding GA-based evolutionary operations are designed to characterise and evolve both the hyperparameters,including batch size,learning rate,weight optimiser and weight regularisation,and the architectures,such as the block-based network topology and the parameters of each CNN *** experimental results on five intrusion detection datasets in IIoT,including secure water treatment,water distribution,Gas Pipeline,Botnet in Internet of Things and Power System Attack Dataset,have demonstrated the superiority of the proposed SOPA-GA-CNN to the state-of-the-art manually designed models and neuron-evolutionary methods in terms of accuracy,precision,recall,F1-score,and the number of parameters of the deep learning models.

关键词： cyber-systems deep learning deep neural network Internet of things swarm intelligence

来源：评论

学校读者我要写书评

暂无评论

A modified CNN-IDS model for enhancing the efficacy of intrusion detection system

引用

Measurement: Sensors 2024年 35卷

作者： Abed, Ruqaya Abdulhasan Hamza, Ekhlas Kadhum Humaidi, Amjad J. Control and Systems Engineering Department University of Technology Iraq

As the security of computer networks in enterprises worldwide is dependent on the proper functioning of intrusion detection systems (IDSs) and intrusion prevention systems (IPSs), this effectiveness of both of them is of utmost priority. Leveraging diverse techniques, these network security systems are created to keep the reliability, the availability, and the integrity of the organizational networks safe. One plus point of using ML in intrusion detection system (IDS) is that it has successfully weeded out all the IDS attacks with a high degree of accuracy. In contrast, such systems may be believed to operate to their least competent levels when supersized data spaces have to be dealt with. In the process to solve this, application of feature selection techniques will play the crucial role to ignore non-relevant features which do not impact the issue of classification much. One more thing to keep in mind is that the ML-based IDSs often have problems with high false alarms and percentage accuracy because of the imbalanced training sets. The undertaking of this paper involves a through the analysis of the UNSW-NB15 intrusion detection data set as upon which our models will be tested and trained. We utilize two feature selection approaches: the PCA method, which is denoted as PCA, and the SVD method, called SVD. Furthermore, we categorize the datasets using these methods— Ridge Regression (RR), Stochastic Gradient Descent, and Convolutional Neural network (CNN)– on the transformed feature space. What is the most widely used for, is that it deals with both, binary and multiclass classification. The result measure that PCA and SVD are succeeded in getting better performance of IDS than others with enhancing the accuracy of classification models. More specifically, the RR classifier's precise was outstanding for the binary classification problem experiencing a rise in the accuracy from 98.13 % to 99.85 %. This shows the critical role of feature selection approaches and is

关键词： network intrusion

来源：评论

学校读者我要写书评

暂无评论

Data-driven consensus control of fully distributed event-triggered multi-agent systems

引用

Science China(Information Sciences) 2023年第5期66卷 147-161页

作者： Yifei LI Xin WANG Jian SUN Gang WANG Jie CHEN School of Automation Beijing Institute of Technology The State Key Lab of Autonomous Intelligent Unmanned Systems Beijing Institute of Technology Beijing Institute of Technology Chongqing Innovation Center Department of Control Science and Engineering Tongji University

This study investigates the consensus control issue in discrete-time linear multi-agent systems(MASs) using data-driven control under undirected communication networks. To alleviate the communication burden, an adaptive event-triggered control strategy involving only local information is proposed and a model-based stability condition is derived that guarantees the asymptotic consensus of MASs. Furthermore,a data-based consensus condition for unknown MASs is established by combining a data-based system representation with the model-based stability condition, using only pre-collected noisy input-state data instead of the accurate system information a priori. Specifically, both model-based and data-driven event-triggered controllers can be utilized without requiring any global information. The validity and correctness of the controllers and associated theoretical results are demonstrated via numerical simulations.

关键词： distributed control event-triggered control data-driven control discrete-time MASs

来源：评论

学校读者我要写书评

暂无评论

Flexibility Enhancement Planning of Active Distribution networks Considering Optimal Dispatch of Demand Response 7

Flexibility Enhancement Planning of Active Distribution Netw...

引用

7th IEEE International Conference on Energy Internet, ICEI 2023

作者： Fan, Ruizhe Hu, Changbin Luo, Shanna Li, Xuecheng Lu, Heng Feng, Jun School of Electrical and Control Engineering North China University of Technology Beijing China University of Duisburg-Essen Department of Automatic Control and Complex Systems Duisburg Germany

ISBN: (纸本)9798350386684

To enhance the operational flexibility of active distribution network (ADN) with high proportion renewable energy, this paper proposes a flexibility resource planning model considering the optimal dispatch of demand response. Firstly, the flexibility in supply side and demand side are analyzed. Then, a two-layer planning model of flexibility resources is established, which includes the planning layer and operating layer. The planning layer introduces the system comprehensive security index to evaluate the system security, aiming at the optimal annual comprehensive cost of the system. The operation layer introduces the flexibility insufficiency rate as the system flexibility evaluation index, and the network loss and renewable energy curtailment are counted as economic penalties, which is aiming at the optimal annual operation cost of the system. The proposed planning model is solved with particle swarm optimization algorithm. Case studies on the IEEE 33-node distribution system show that the proposed model can effectively reduce the network loss and the renewable generation curtailment. © 2023 IEEE.

关键词： Electric power distribution

来源：评论

学校读者我要写书评

暂无评论

Enhancing Data Labeling Through Integration of SS-DBSCAN Clustering and Human-in-the-Loop 24

Enhancing Data Labeling Through Integration of SS-DBSCAN Clu...

引用

6th Asia Conference on Machine Learning and Computing, ACMLC 2024

作者： Monko, Gloriana J. Kimura, Masaomi Department of Functional Control Systems Shibaura Institute of Technology Tokyo Japan School of Engineering Shibaura Institute of Technology Koto City Tokyo Japan

ISBN: (纸本)9798400710018

This study introduces a novel method for integrating Stratified Sampling for Density-Based Spatial Clustering of Applications with Noise (SS-DBSCAN) clustering with the human-in-the-loop approach to semi-supervised data labeling. We enhance the feature representation of text data using attention mechanisms, leading to the formation of more contextually relevant clusters. The clusters are then processed using SS-DBSCAN, where a custom similarity function aligns them closely with expert-labeled examples, facilitating efficient label propagation. Subsequently, human experts reevaluate the clustering outputs, refining the process by reassessing the feature similarities to the expert-labeled classes. This iterative process ensures high fidelity in label assignment and enhances the overall reliability of the semi-supervised learning approach. Applied to MIMIC III and Cancer Doc Classification datasets, this method not only aligns the data labels more closely with expert evaluations but also leverages human judgment to confirm the accuracy and reliability of the re-clustering process. Based on the BERT transformers model, the fine-tuned prediction model achieved 100% accuracy in predicting the correct label and evaluated the re-clustering technique's effectiveness, demonstrating 90.3% and 97.8% correctness in the re-clustered labels in MIMIC III and Cancer Doc Classification datasets, respectively. © 2024 Copyright held by the owner/author(s).

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Onboard Near-Real Time System Identification From Open and Closed-Loop Flight Data

Onboard Near-Real Time System Identification From Open and C...

引用

AIAA Science and Technology Forum and Exposition, AIAA SciTech Forum 2025

作者： McQuinn, Cassie-Kay Valasek, John Vehicle Systems & Control Laboratory Aerospace Engineering Department Texas A&M University College StationTX77843-3141 United States

ISBN: (数字)9781624107238

ISBN: (纸本)9781624107238

This paper develops and demonstrates an approach for identifying linear time invariant state-space dynamical models of small Unmanned Air systems (UAS) with and without active flight controllers, in near-real time. Models are generated onboard the vehicle from both openloop and closed-loop data using low cost commercial-off-the-shelf components. A controller, external to the primary flight controller, is developed and implemented to generate and record controller inputs for the closed-loop system using an extension of the Developmental Flight Test Instrumentation 2 (DFTI2). The Observer/Kalman filter Identification (OKID) algorithm generates linear, time invariant state-space realizations of longitudinal, lateral/directional, and combined longitudinal and lateral/directional models. The models are generated independent of actuator dynamics by mapping control effector deflection angles to measured servo potentiometer readings. Orthogonal Schroeder sine sweep excitations are used to reduce potential control coupling while also exciting multiple frequencies. Models identified from flight data are presented and evaluated in the paper. Results presented in the paper demonstrate that this system can generate models suitable for describing the dynamics of the vehicle, operating both with and without a controller implemented. © 2025, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved.

关键词： Invariance

来源：评论

学校读者我要写书评

暂无评论

Performance Improvement control Strategy of Microgrid Multi-Inverter Parallel System 7

Performance Improvement Control Strategy of Microgrid Multi-...

引用

7th IEEE Conference on Energy Internet and Energy System Integration, EI2 2023

作者： Lu, Heng Hu, Changbin Luo, Shanna Li, Xuecheng Huang, Haiyang Deng, Jufu University of Duisburg-Essen Department of Automatic Control and Complex Systems Duisburg Germany North China University of Technology College of Electrical and Control Engineering Beijing China

ISBN: (纸本)9798350345094

This paper develops a performance improvement control strategy based on residual generator to improve the control performance of the multi-inverter parallel system in the presence of load disturbances and line parameter perturbation. This study establishes the state space models of the electrical network and inverters to analyze the performance degradation issues. The model-match technique and multi-agent consistency algorithm are used to design the dynamic and steady-state performance improvement controllers, respectively. Experimental results validate the effectiveness of the proposed control method. © 2023 IEEE.

关键词： Electric loads

来源：评论

学校读者我要写书评

暂无评论

Long-Range Vehicle Detection of LiDAR using Adaboost Classification

Long-Range Vehicle Detection of LiDAR using Adaboost Classif...

引用

2025 IEEE/SICE International Symposium on System Integration, SII 2025

作者： Lim, Na-Young Lee, Jeon-Hyeok Park, Tae-Hyoung Chungbuk National University Department of Intelligent Systems & Robotics Cheongju Korea Republic of Chungbuk National University Department of Control & Robot Engineering Cheongju Korea Republic of

ISBN: (纸本)9798331531614

In recent developments, autonomous racing has garnered attention as it aims to overcome the limitations of standard autonomous driving systems. Achieving safe racing conditions necessitates both fast and long-range perception. However, current 3D LiDAR object detection methods face challenges with high computational costs and limited detection ranges. These issues make them unsuitable for racing scenarios. To address these challenges, this paper proposes a clustering-based long-range vehicle detection method that relies solely on LiDAR. First, the method removes ground points and foreground points and clusters the remaining points. Subsequently, these clusters are classified as vehicles using AdaBoost, generating 2D bounding boxes in the range view. Experimental results demonstrate superior performance, achieving a computational efficiency of 53 Hz and a long-range detection accuracy of over 80%, compared to voxel-based and range-based methods. This approach offers a viable solution for autonomous racing environments. © 2025 IEEE.

关键词： Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

Comparative Study of Parametric Disturbances Effect on Cart-Inverted Pendulum System Stabilization 1

引用

International Conference on Modern Electronics Devices and Communication systems, MEDCOM 2021

作者： Verma, Neelam Valluru, Sudarshan K. Department of Electrical Engineering Center for Control of Dynamical Systems and Computation Delhi Technological University Delhi India

ISBN: (数字)9789811963834

ISBN: (纸本)9789811963827

The main motivation behind the present work was to validate the impact of pendulum mass, cart mass, and length of pendulum on stabilization and swing-up of cart-inverted pendulum. Inverted pendulum system is a classic benchmark example of underactuated electro-mechanical systems, where the degree of freedom is less than the number of inputs available. PID controller was designed for a set of parameter values of the system. MATLAB simulation was carried out for different values of pendulum mass (0.2–0.25 kg), pendulum length (0.3–0.34 m), and cart mass (2.3–2.45 kg). The performance of considered cart- inverted pendulum system is demonstrated with simulation results for a different set of results, and a comparative analysis was made to comment on their effects on swing-up and stabilization. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Stabilization

来源：评论

学校读者我要写书评

暂无评论

Uncertainty and its Effect on Optimal Multidrug control of Hemodynamic Variables

Uncertainty and its Effect on Optimal Multidrug Control of H...

引用

2024 American control Conference, ACC 2024

作者： Popescu, Teodora Birs, Isabela R. Othman, Ghada Ben Yumuk, Erhan Mihai, Marcian Hegedus, Erwin Copot, Dana De Keyser, Robin Ionescu, Clara M. Muresan, Cristina I. Technical University of Cluj-Napoca Department of Automation Cluj-Napoca Romania Ghent University Research Group on Dynamical Systems and Control Department of Electromechanics Systems and Metal Engineering Tech Lane Science Park 125 Gent9052 Belgium Istanbul Technical University Department of Control and Automation Engineering Maslak Istanbul34469 Turkey

ISBN: (纸本)9798350382655

Computer aided control in biomedical applications is gaining more and more popularity due to numerous research studies that have proven the efficiency of automatic control over manual dosing, which is highly susceptible to human errors. Optimal drug dosing is best achieved using automatic control, which triggers important benefits in terms of both costs and patient side-effects. However, mathematical models for patients are highly susceptible to large modeling uncertainty. A predictive control algorithm is designed in this paper for optimal multidrug control of hemodynamic variables. Improved closed loop performance is obtained compared to similar control strategies, for ± 30% modeling uncertainty. The simulation results demonstrate that predictive control is a feasible solution for optimal drug dosing. An analysis of the closed loop performance for significant patient variability shows that controllers tuned using a nominal patient model often fail to achieve desired robustness. To limit the effect of modeling uncertainty, the prediction model should be updated using an online identification tool to extract patient features. © 2024 AACC.

关键词： Predictive control systems

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：