检索结果-内蒙古大学图书馆

2023 China automation Congress, CAC 2023

作者： Hu, Yumei Li, Xing Zhang, Yang Zhu, Yuxuan Sun, Yuechao College of Computer Science and Technology Dongguan University of Technology Dongguan China School of Electrical Engineering & Intelligentization Dongguan University of Technology Dongguan China Northeastern University State Key Laboratory of Synthetical Automation for Process Industries Shenyang China School of Mechanical Engineering of Dgut Dongguan University of Technology Dongguan China Northeastern University College of Information Science and Engineering Northeastern University Shenyang China

ISBN: (纸本)9798350303759

Cardiovascular disease is a sudden and extremely disabling and deadly disease that hinders the development of public health care in China and poses a serious threat to the health of users. Therefore, this paper proposes a deep neural network-assisted diagnosis of cardiovascular disease based on TAdam algorithm optimization under a distributed medical monitoring network to enable rapid identification of cardiovascular patients, timely detection and reporting of the condition. This paper addresses the problems that deep neural network models are susceptible to the influence of their structure and hyperparameters and poor generalization convergence, firstly, the mainstream optimization algorithms based on deep learning are sorted out, and finally the T Adam algorithm is used to optimize the model. By comparing the common gradient optimization algorithms SGD (Stochastic Gradient Descent), Adagrad (Adaptive Gradient Algorithm), RMSprop (Root Mean Square propagation), Adam (Adaptive Moment Estimation) and the improved TAd am optimization algorithm trained and tested on the same dataset, the experimental results show that the TAdam algorithm exhibits the fastest convergence speed and the best generalization performance on the cardiovascular disease dataset. The method proposed in this paper enables rapid diagnosis of the user's disease risk and feedback to the relevant personnel to maximise user safety. © 2023 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Performance-Oriented Event-Triggered Merging-Following Control of Multi-Lane Vehicles

Performance-Oriented Event-Triggered Merging-Following Contr...

引用

Chinese Control and Decision Conference, CCDC

作者： Yan-Xi Liu Chen-Liang Zhang Ge Guo College of Information Science and Engineering Northeastern University Shenyang China State Key Laboratory of Synthetical Automation for Process Industries Northeastern University Shenyang China School of Control Engineering Northeastern University at Qinhuangdao Qinhuangdao China

ISBN: (数字)9798350387780

ISBN: (纸本)9798350387797

This paper focuses on the merging-following control of vehicles standing on different lanes in presence of performance requirements and limited communication resources. A constraint of tracking error is developed to prescribe the performance, and then addressed via error mapping transformation, leading to a prescribed performance control (PPC) algorithm that can make the both distance and angular errors between consecutive vehicles evolve within the prescribed region. Then an event-triggered control protocol driven by the performance function is incorporated to reduce the signal transmission frequency between the controller and actuator with a performance-communication balance. Based on the Lyapunov stability proof, it is verified that all signals in the vehicular system remain uniformly ultimately bounded. A numerical simulation is presented to illustrate the effectiveness of the proposed scheme.

关键词： Actuators Protocols Numerical simulation Steady-state Planning Transient analysis Frequency control

来源：评论

学校读者我要写书评

暂无评论

Bearing fault diagnosis algorithm based on multi-source data and auxiliary classifier

Bearing fault diagnosis algorithm based on multi-source data...

引用

Chinese Control and Decision Conference, CCDC

作者： Yulin Jin Xiaochuan Luo Lei Zhang College of Information Science and Engineering Northeastern University Shenyang China State Key Laboratory of Synthetical Automation for Process Industries Northeastern University Shenyang China

In recent years, with the development of sensor technology and deep learning, intelligent bearing diagnosis algorithms using multi-source data have achieved great success. However, when faced with the input of single sensor data, the deep learning model trained with multi-source data cannot work, which affects the flexibility of model application. Meanwhile, the method of only adding a loss function in the final classification layer is insufficient to restrict each branch model, and will also affect the performance of the model. Based on this, this paper proposes a parallel convolutional neural network with auxiliary classifier. The auxiliary classifiers added in each branch can not only give fault diagnosis results when only a single data is used, but also improve the performance of the final multi-data fusion by optimizing the output features of each branch, and broaden the application range of the diagnosis model. The proposed method is tested on the bearing fault dataset, the experimental results show that, after adding multiple losses, the structural form of the auxiliary classifier can effectively improve the fault diagnosis accuracy and robustness of the parallel convolutional neural network. Meanwhile, the auxiliary classifiers of each branch can also obtain good classification performance when facing single sensor data, which better improves the flexibility of the model.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Proxy-Supervised Cross-Spectral Stereo Matching

Proxy-Supervised Cross-Spectral Stereo Matching

引用

IEEE International Conference on Cyber Technology in automation, Control, and Intelligent Systems

作者： Jiayu Zhang Tong Jia Shiyi Guo Wenhao Li Yunchun Zhou Tao Yang College of Information Science and Engineering Northeastern University Shenvang China college of Information Science and Engineering Northeastern University Shenyang China State Key Laboratory of Synthetical Automation for Process Industries Northeastern University Shenyang China

ISBN: (数字)9798331506056

ISBN: (纸本)9798331506063

Cross-spectral stereo matching directly retrieves disparity from different spectral images and is widely used in the fields of image alignment and 3D perception. The lack of high-quality labeled data is currently one of the main challenges for cross-spectral stereo matching. To solve this problem, we propose a new paradigm for cross-spectral stereo matching. Firstly, high-quality dense pseudo-labels are generated using the monocular depth estimation method and conventional stereo matching method. Then the pseudo-labels are used as proxy supervision to train the stereo-matching networks. To further refine the geometric distribution of disparity results, we propose a stereo normalized gradient loss that constrains the network to learn a better global geometric distribution. Finally, we conduct extensive experiments on the Pittsburgh dataset, and the results show that our method significantly outperforms existing methods in terms of accuracy and global geometric distribution, and our accuracy is improved by about 11.8% compared to the state-of-the-art (SOTA) method. The code is available at: https://***/jiayuzhang128/pcs-stereo

关键词： Training Accuracy Three-dimensional displays Codes automation Control systems Intelligent systems

来源：评论

学校读者我要写书评

暂无评论

Finite time stabilization for multi-path cyber physical system with time-varying parameters

Finite time stabilization for multi-path cyber physical syst...

引用

Conference on Vehicle Control and Intelligence (CVCI)

作者： Jianying Di Qinglong Zhang Cheng Tan Ge Guo College of Information Science and Engineering Northeastern University Shenyang China School of Engineering Qufu Normal University Rizhao Shandong China The State Key Laboratory of Synthetical Automation for Process Industries Northeastern University Shenyang China

ISBN: (数字)9798331504892

ISBN: (纸本)9798331504908

This paper investigates the finite time stabilization for multi-path cyber physical systems with random denial-of-service (DoS) attacks and time-varying parameters. Unlike previous research, we use auxiliary systems to establish sufficient conditions for finite time stability in closed-loop multi-path CPSs with various delays and parameters, based on a detailed analysis of the state transition matrix. We also present a condition derived from the Lyapunov function and propose an LMI-based method for designing a non-fragile controller. Finally, we apply these theoretical contributions to a stock investment model to demonstrate their validity.

关键词： Sufficient conditions Denial-of-service attack Control systems Stability analysis Delays Time-varying systems Lyapunov methods Investment

来源：评论

学校读者我要写书评

暂无评论

Distributed Fault-Tolerant Consensus Control for Linear Multi-Agent System with K-Step Hierarchical Protocol

Distributed Fault-Tolerant Consensus Control for Linear Mult...

引用

International Conference on Guidance, Navigation and Control, ICGNC 2022

作者： Wang, Yuanwen Dong, Jiuxiang College of Information Science and Engineering Northeastern University Shenyang110819 China Liaoning Province Key Laboratory of Safe Operation Technique for Autonomous Unmanned Systems Northeastern University Shenyang110819 China State Key Laboratory of Synthetical Automation for Process Industries Northeastern University Shenyang110819 China Key Laboratory of Vibration and Control of Aero-Propulsion Systems Ministry of Education Northeastern University Shenyang110819 China

ISBN: (纸本)9789811966125

The distributed fault-tolerant consensus control using k-step hierarchical protocol for linear multi-agent system (MAS) is studied in this note. Unlike most existing fault-tolerant methods for compensating actuator fault, a hierarchical protocol consisting of a virtual layer and a actual layer that considers both fault compensation and introducing fault information into the trajectory planning process which means that the motion of the MAS does not exceed the fault-tolerant compensation range of the faulty agent is proposed. Trajectory planning is responsible for the virtual layer consisting of virtual agent system and a co-controller that introduces fault information. The actual layer, consisting of the actual agent system, actuator fault observer, and tracking trajectory controller, traces the virtual layer for consensus. It is shown that the integrity of the MAS is guaranteed. Finally, the validity of the developed protocols is illustrated by simulation examples. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Multi agent systems

来源：评论

学校读者我要写书评

暂无评论

Resilient control under denial-of-service and uncertainty: An adaptive dynamic programming approach

arXiv

引用

arXiv 2024年

作者： Gao, Weinan Jiang, Zhong-Ping Chai, Tianyou State Key Laboratory of Synthetical Automation for Process Industries Northeastern University Shenyang110819 China Department of Electrical and Computer Engineering Tandon School of Engineering New York University BrooklynNY11201 United States

In this paper, a new framework for the resilient control of continuous-time linear systems under denial-of-service (DoS) attacks and system uncertainty is presented. Integrating techniques from reinforcement learning and output regulation theory, it is shown that resilient optimal controllers can be learned directly from real-time state and input data collected from the systems subjected to attacks. Sufficient conditions are given under which the closed-loop system remains stable given any upper bound of DoS attack duration. Simulation results are used to demonstrate the efficacy of the proposed learning-based framework for resilient control under DoS attacks and model uncertainty. © 2024, CC BY.

关键词： Denial-of-service attack

来源：评论

学校读者我要写书评

暂无评论

Multi label feature selection for copper flotation process design

Multi label feature selection for copper flotation process d...

引用

Chinese Control and Decision Conference, CCDC

作者： Haipei Dong Fuli Wang Dakuo He College of Information Science and Engineering Northeastern University Shenyang China State Key Laboratory of Synthetical Automation for Process Industries Northeastern University Shenyang China

The copper flotation process design is a multi label learning. Before multi label classification and regression, label-specific features for each label need to be selected from the original features. This paper puts forward a multi label feature selection (MLFS) based on domain knowledge and label correlation. In the proposed MLFS, the domain knowledge function applies the knowledge of copper flotation in feature selection, and the label correlation function introduces the Kullback Leibler divergence between labels in feature selection. The experimental results demonstrate that the proposed MLFS has superiority over benchmark multi label feature selection algorithms in the copper flotation backbone process design.

关键词：

来源：评论

学校读者我要写书评

暂无评论

GenoHoption: Bridging Gene Network Graphs and Single-Cell Foundation Models

arXiv

引用

arXiv 2024年

作者： Cheng, Jiabei Li, Jiachen Yang, Kaiyuan Shen, Hongbin Yuan, Ye Department of Automation School of Electronic Information and Electrical Engineering Shanghai Jiaotong University Shanghai China Key Laboratory of Biopharmaceutical Preparation and Delivery Chinese Academy of Sciences Beijing China State Key Laboratory of Biochemical Engineering Institute of Process Engineering Chinese Academy of Sciences Beijing China

The remarkable success of foundation models has sparked growing interest in their application to single-cell biology. Models like Geneformer and scGPT promise to serve as versatile tools in this specialized field. However, representing a cell as a sequence of genes remains an open question since the order of genes is interchangeable. Injecting the gene network graph offers gene relative positions and compact data representation but poses a dilemma: limited receptive fields without in-layer message passing or parameter explosion with message passing in each layer. To pave the way forward, we propose GenoHoption, a new computational framework for single-cell sequencing data that effortlessly combines the strengths of these foundation models with explicit relationships in gene networks. We also introduce a constraint that lightens the model by focusing on learning the predefined graph structure while ensuring further hops are deducted to expand the receptive field. Empirical studies show that our model improves by an average of 1.27% on cell-type annotation and 3.86% on perturbation prediction. Furthermore, our method significantly decreases computational overhead and exhibits few-shot potential. Overall, GenoHoption can function as an efficient and expressive bridge, connecting existing single-cell foundation models to gene network graphs. Copyright © 2024, The Authors. All rights reserved.

关键词： Network theory (graphs)

来源：评论

学校读者我要写书评

暂无评论

GenoHoption: Bridging Gene Network Graphs and Single-Cell Foundation Models

GenoHoption: Bridging Gene Network Graphs and Single-Cell Fo...

引用

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Jiabei Cheng Jiachen Li Kaiyuan Yang Hongbin Shen Ye Yuan Department of Automation School of Electronic Information and Electrical Engineering Shanghai Jiaotong University Shanghai China Key Laboratory of Biopharmaceutical Preparation and Delivery Chinese Academy of Sciences Beijing China State Key Laboratory of Biochemical Engineering Institute of Process Engineering Chinese Academy of Sciences Beijing China

ISBN: (数字)9798350386226

ISBN: (纸本)9798350386233

The remarkable success of foundation models has sparked growing interest in their application to single-cell biology. Models like Geneformer and scGPT promise to serve as versatile tools in this specialized field. However, representing a cell as a sequence of genes remains an open question since the order of genes is interchangeable. Injecting the gene network graph offers gene relative positions and compact data representation but poses a dilemma: limited receptive fields without in-layer message passing or parameter explosion with message passing in each layer. To pave the way forward, we propose GenoHoption, a new computational framework for single-cell sequencing data that effortlessly combines the strengths of these foundation models with explicit relationships in gene networks. We also introduce a constraint that lightens the model by focusing on learning the predefined graph structure while ensuring further hops are deducted to expand the receptive field. Empirical studies show that our model improves by an average of 1.27% on cell-type annotation and 3.86% on perturbation prediction. Furthermore, our method significantly decreases computational overhead. Overall, GenoHoption can function as an efficient and expressive bridge, connecting existing single-cell foundation models to gene network graphs. 1

关键词： Bridges Sequential analysis Foundation models Biological system modeling Message passing Computational modeling Perturbation methods Predictive models Genetics Periodic structures

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：