检索结果-内蒙古大学图书馆

Dynamic surface control with a nonlinear disturbance observer for multi-degree of freedom underactuated mechanical systems

引用

INTERNATIONAL JOURNAL OF ROBUST AND NONLINEAR control 2022年第14期32卷 7809-7827页

作者： Ding, Feng Huang, Jian Xu, WenXia Yang, Chenguang Sun, Chong Ai, Yong South Cent Minzu Univ Sch Comp Sci Wuhan Hubei Peoples R China Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Key Lab Image Proc & Intelligent Control Luoyu Rd 1037 Wuhan Hubei Peoples R China Wuhan Inst Technol Sch Comp Sci & Engn Wuhan Hubei Peoples R China Univ West England Bristol Robot Lab Bristol Avon England

Underactuated systems are extensively utilized in practice while attracting a huge deal of attention in theoretical studies. There are few robust control strategies for general underactuated systems because of the variety of their dynamic models. A dynamic surface control strategy with a nonlinear disturbance observer is proposed in this study, to stabilize multi-degree of freedom underactuated systems. In such systems the number of underactuated degrees of freedom is not higher compared to the actuated ones. A disturbance observer is utilized to dispose of the uncertain disturbance and cross terms in dynamic model which may cause failure to the controller. Then, a dynamic surface control strategy is presented which is not sensitive to the diversity of dynamic models. The stability of whole system is proven by Lyapunov-based method. The control law is successfully applied to nonlinear underactuated systems in benchmark cascade form such as two-translational oscillator with rotational actuator, crane, wheeled inverted pendulum. The effectiveness of proposed controllers are illustrated by MATlab simulation results. Finally, comparative studies are presented to verify the superiority of the proposed method.

关键词： dynamic surface control nonlinear disturbance observer underactuated systems uniformly ultimately bounded

来源：评论

学校读者我要写书评

暂无评论

Universal adversarial perturbations for CNN classifiers in EEG-based BCIs

引用

JOURNAL OF NEURAL ENGINEERING 2021年第4期18卷

作者： Liu, Zihan Meng, Lubin Zhang, Xiao Fang, Weili Wu, Dongrui Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Minist Educ Image Proc & Intelligent Control Key Lab Wuhan 430074 Peoples R China Natl Univ Singapore Sch Design & Environm Singapore 117566 Singapore Zhejiang Lab Hangzhou 311121 Peoples R China

Objective. Multiple convolutional neural network (CNN) classifiers have been proposed for electroencephalogram (EEG) based brain-computer interfaces (BCIs). However, CNN models have been found vulnerable to universal adversarial perturbations (UAPs), which are small and example-independent, yet powerful enough to degrade the performance of a CNN model, when added to a benign example. Approach. This paper proposes a novel total loss minimization (TLM) approach to generate UAPs for EEG-based BCIs. Main results. Experimental results demonstrated the effectiveness of TLM on three popular CNN classifiers for both target and non-target attacks. We also verified the transferability of UAPs in EEG-based BCI systems. Significance. To our knowledge, this is the first study on UAPs of CNN classifiers in EEG-based BCIs. UAPs are easy to construct, and can attack BCIs in real-time, exposing a potentially critical security concern of BCIs.

关键词： brain-computer interface convolutional neural network electroencephalogram universal adversarial perturbation

来源：评论

学校读者我要写书评

暂无评论

Remaining useful life prediction of lithium battery based on capacity regeneration point detection

引用

ENERGY 2021年 234卷 121233-121233页

作者： Ma, Qiuhui Zheng, Ying Yang, Weidong Zhang, Yong Zhang, Hong Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Key Lab Image Informat Proc & Intelligent Control Wuhan 430074 Peoples R China Wuhan Univ Sci & Technol Sch Informat Sci & Engn Wuhan 430081 Peoples R China

Lithium batteries have been widely used in various electronic devices, and the accurate prediction of its remaining useful life (RUL) can prevent the occurrence of sudden equipment failure. Battery capacity is a commonly used indicator to represent the health status of lithium batteries. However, the capacity regeneration is usually unavoidable due to the impact of battery "rest time" between two cycles, which leads to inaccurate prediction of the RUL. To solve this problem, this paper combines the particle filter (PF) and Mann-Whitney U test (PF-U) to detect the capacity regeneration point (CRP). In this light, the autoregressive (AR) model and PF algorithm are adopted for RUL prediction. The predicted capacity through AR model is used to update the degradation model parameters of PF algorithm, and the validation of our approach is verified through the lithium battery dataset of NASA. In comparison, our proposed method exhibits the highest precision and provides a platform to detect the points with capacity regeneration, and further reduce the RUL prediction error. (c) 2021 Elsevier Ltd. All rights reserved.

关键词： Capacity regeneration RUL Lithium battery Particle filter Mann-Whitney U test Autoregressive model

来源：评论

学校读者我要写书评

暂无评论

Optimizing Parameter Design with Frequency and Clock Drift Constraints in Microgrids⋆

Optimizing Parameter Design with Frequency and Clock Drift C...

引用

IEEE Conference on Decision and control

作者： Chang Yu Xiaoqing Lu School of Information Science and Engineering Wuhan University of Science and Technology Wuhan China Key Laboratory of Image Processing and Intelligent Control Huazhong University of Science and Technology Ministry of Education Wuhan China School of Electrical Engineering and Automation Wuhan University Wuhan China

ISBN: (数字)9798350316339

ISBN: (纸本)9798350316346

Microgrids subjected to secondary cooperative control encounter significant challenges, including operational constraints and clock drifts, adversely affecting their stability and efficiency. This paper provides conditions that assure optimal microgrid performance in both transient and steady-state scenarios, focusing on the effects of clock drifts and fluctuations in load. Furthermore, we introduce a novel approach for designing secondary control parameters, specifically engineered to minimize steady-state discrepancies attributable to clock drifts while ensuring adherence to standards for transient operations. Comprehensive experimental validations corroborate the effectiveness of our proposed solutions.

关键词： Fluctuations Stability criteria Focusing Microgrids Steady-state Transient analysis Standards Frequency control Clocks Load modeling

来源：评论

学校读者我要写书评

暂无评论

Hip Joint Trajectory Generation Based on Human Limb Motion Synergy 7

Hip Joint Trajectory Generation Based on Human Limb Motion S...

引用

7th IEEE International Conference on Advanced Robotics and Mechatronics

作者： Li, Yuge Tao, Chunjing Wang, Enkai He, Xinrun Wang, Jing Huang, Jian Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Key Lab Image Proc & Intelligent Control Wuhan 430074 Peoples R China Beihang Univ Beijing Adv Innovat Ctr Biomed Engn Sch Engn Med Beijing 100083 Peoples R China

ISBN: (数字)9781665483063

ISBN: (纸本)9781665483063

Lower limb exoskeleton robot can help hemiplegic patients with rehabilitation training and enhance human movement ability. Lower limb trajectory generation is an indispensable part of lower limb exoskeleton robot control, which can better realize human-machine collaboration and significantly improve the quality of life of patients. One of the main challenges in this field is to predict the wearer's hip joint trajectory in advance. In this study, we analyzed the synergy law between limbs in the process of human walking, and studied how to use the shoulder motion data to predict the hip joint trajectory based on the synergy law. We recruited four subjects, and measured their three-dimensional coordinate motion trajectories data of 8 markers at the left and right shoulders, elbows, hip joints, and knee joints by the optical motion capture system OptiTrack. The non-linear mapping model between the shoulder and the hip joint is established using the least square (LS) method. An error correction model is also proposed based on long short term memory (LSTM) to reduce the hip joint trajectory prediction error. Based on the proposed synergy law of upper and lower limbs, the hip joint trajectory can be predicted in advance only by using the shoulder motion information. The predicted hip joint trajectory error can reach as low as 2.1 degrees.

关键词： Training Robot kinematics Exoskeletons Robot control Shoulder Optical variables measurement Predictive models

来源：评论

学校读者我要写书评

暂无评论

Large Language Models are Few-shot Generators: Proposing Hybrid Prompt Algorithm To Generate Webshell Escape Samples

arXiv

引用

arXiv 2024年

作者： Ma, Mingrui Han, Lansheng Zhou, Chunjie School of Cyber Science and Engineering Huazhong University of Science and Technology Hubei Wuhan430074 China The Key Laboratory of Ministry of Education for Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology Hubei Wuhan430074 China

The frequent occurrence of cyber-attacks has made webshell attacks and defense gradually become a research hotspot in the field of network security. However, the lack of publicly available benchmark datasets and the over-reliance on manually defined rules for webshell escape sample generation have slowed down the progress of research related to webshell escape sample generation and artificial intelligence (AI)-based webshell detection. To address the drawbacks of weak webshell sample escape capabilities, the lack of webshell datasets with complex malicious features, and to promote the development of webshell detection, we propose the Hybrid Prompt algorithm for webshell escape sample generation with the help of large language models. As a prompt algorithm specifically developed for webshell sample generation, the Hybrid Prompt algorithm not only combines various prompt ideas including Chain of Thought, Tree of Thought, but also incorporates various components such as webshell hierarchical module and few-shot example to facilitate the LLM in learning and reasoning webshell escape strategies. Experimental results show that the Hybrid Prompt algorithm can work with multiple LLMs with excellent code reasoning ability to generate high-quality webshell samples with high Escape Rate (88.61% with GPT-4 model on VirusTotal detection engine) and Survival Rate (54.98% with GPT-4 model). Copyright © 2024, The Authors. All rights reserved.

关键词： Network security

来源：评论

学校读者我要写书评

暂无评论

Full-Analog Reservoir Computing Circuit Based on Memristor With a Hybrid Wide-Deep Architecture

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS 2024年第2期71卷 501-514页

作者： Chen, Liangyu Wang, Xiaoping Yang, Chao Chen, Zhanfei Zhang, Junming Zeng, Zhigang Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Key Lab Image Proc & Intelligent Control Educ Minist China Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Inst Artificial Intelligence Sch Artificial Intelligence & Automat Wuhan 430074 Peoples R China Guangxi Univ Sci & Technol Sch Comp Sci & Technol Liuzhou 545006 Peoples R China Huazhong Univ Sci & Technol Sch Integrated Circuits Wuhan 430074 Peoples R China

Reservoir computing (RC) contains two significant variants: wide RC and deep RC. The hybrid wide-deep architecture absorbs their strengths with a powerful parallel processing capability while enhancing the memory capacity of reservoirs. However, the fully analog RC circuit combining the two structures has yet to be proposed, mainly due to unmanageable hierarchical signal processing. Here we report a full-analog memristive RC circuit with a hybrid wide-deep architecture comprising an input module, mask module, reservoir module, and readout module. The input module can generate continuous voltages with temporal sequences. The mask module provides parallel mask processes, laying the foundation for implementing a wide RC structure. The reservoir module includes dynamic memristors and postprocessing circuits. Dynamic memristors can produce high-dimensional reservoir states, and postprocessing circuits allow memristive reservoir circuits to be cascaded to achieve a deep RC structure. The readout module mainly consists of a nonvolatile memristor crossbar and an analog integrator, enabling an efficient multiplication-and-accumulation operation. The simulation results in LTspice illustrate that the memory capacity of the proposed circuit is 91.6% higher than that of wide RC. Moreover, it can efficiently perform temporal tasks, obtaining a high accuracy of 98.99% in arrhythmia detection.

关键词： Reservoir computing memristor neuromorphic architecture arrhythmia detection analog circuit design

来源：评论

学校读者我要写书评

暂无评论

Tight Formulations of Unit Commitment with General Start-Up Ramp Rate Constraints

Tight Formulations of Unit Commitment with General Start-Up ...

引用

2021 IEEE Sustainable Power and Energy Conference, iSPEC 2021

作者： Gao, Yijie Zhang, Xuan Yu, Yaowen Huazhong University of Science and Technology Key Laboratory of Image Processing and Intelligent Control School of Artificial Intelligence and Automation Wuhan China Hitachi Abb Power Grid San JoseCA United States

ISBN: (纸本)9781665414395

Unit commitment (UC) is a central market clearing process in wholesale electricity markets. With the increasing size and complexity of market-clearing models, UC problems become more and more complicated. To improve the computation efficiency, this paper tightens UC formulations with the general start-up ramp rate. Through analyzing their parameters over three periods, units are categorized into several categories for constructing corresponding tight formulations, and three of the categories are beyond existing works. By applying a recent systematic formulation tightening approach on top of these categories, new tight constraints of the general start-up ramp rate have been obtained. Numeral testing results illustrate the tightening process and demonstrate the improvement of computational efficiency when using our tightened formulations. © 2021 IEEE.

关键词： Computational efficiency

来源：评论

学校读者我要写书评

暂无评论

Performance prediction of gasification-integrated solid oxide fuel cell and gas turbine cogeneration system based on PSO-BP neural network

引用

RENEWABLE ENERGY 2024年 237卷

作者： Wu, Xiao-long Yang, Yuxiao Li, keye Xu, Yuan-wu Peng, Jingxuan Chi, Bo Wang, Zhuo Li, Xi Nanchang Univ Sch Informat Engn Nanchang 330031 Jiangxi Peoples R China Wuhan Univ Sci & Technol Sch Informat Sci & Engn Wuhan 430081 Hubei Peoples R China Nanyang Technol Univ Sch Elect & Elect Engn Singapore 639798 Singapore Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Key Lab Minist Educ Image Proc & Intelligent Control Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Ctr Fuel Cell Innovat Sch Mat Sci & Engn Wuhan 430074 Peoples R China

By utilizing syngas produced from biomass gasification as fuel in a combined heat and power (CHP) system comprising solid oxide fuel cells (SOFC) and gas turbines (GT). This green technology uses syngas from biomass gasification in a CHP system with SOFC and GT. It integrates biomass gasification with new energy sources and has great potential for renewable energy. However, performance prediction remains challenging due to the complex multi-physics coupling between the bubbling fluidized bed gasifier (BFBG) and the SOFC-GT system. In addition, the system has various real-world applications and requires a universal method to comprehensively analyze its thermoelectric performance during operation. Thus, a robust theoretical model is essential for investigating the application of the BFBG-SOFC-GT system. In this study, a novel BFBG-SOFC-GT system utilizing 15 different types of biomass gas is constructed using Aspen Plus, and a particle swarm optimizationbackpropagation (PSO-BP) neural network model is developed for performance prediction. Numerical simulations are also carried out to validate and analyze the integrated system. The effects of gasification temperature, steam-to-biomass mass (S/B) value, and anode temperature on SOFC voltage and current density, as well as the system's electrical efficiency and CHP efficiency, are analyzed. The results demonstrate that increasing gasification temperature decreases SOFC voltage, increases current density, and reduces both electrical and cogeneration efficiency. Conversely, increasing the S/B value and anode temperature leads to higher SOFC voltage, lower current density, and improved electrical and cogeneration efficiency. Additionally, the PSO-BP neural network model's effectiveness in predicting system performance is validated. The results show that, in error evaluations of voltage, current density, electrical efficiency, and cogeneration efficiency, R2 exceeds 0.98, MAPE is below 0.06, and RMSE is less than 0.33.

关键词： Solid oxide fuel cell Biomass gasification Gas turbine Cogeneration Neural network

来源：评论

学校读者我要写书评

暂无评论

Resilient Multiagent Reinforcement Learning With Function Approximation

引用

IEEE TRANSACTIONS ON AUTOMATIC control 2024年第12期69卷 8497-8512页

作者： Ye, Lintao Figura, Martin Lin, Yixuan Pal, Mainak Das, Pranoy Liu, Ji Gupta, Vijay Huazhong Univ Sci & Technol Key Lab Image Proc & Intelligent Control Minist Educ Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Wuhan 430074 Peoples R China Fortna Denver CO 80110 USA SUNY Stony Brook Dept Appl Math & Stat Centereach NY 11720 USA Meta Menlo Pk CA 94025 USA Purdue Univ Elmore Family Sch Elect & Comp Engn W Lafayette IN 47906 USA SUNY Stony Brook Dept Elect & Comp Engn Centereach NY 11720 USA

Adversarial attacks during training can strongly influence the performance of multiagent reinforcement learning algorithms. It is, thus, highly desirable to augment existing algorithms such that the impact of adversarial attacks on cooperative networks is at least bounded. We consider a fully decentralized network, where each agent receives a local reward and observes the global state and action. We propose a resilient consensus-based actor-critic algorithm, whereby each agent estimates the team-average reward and value function, and communicates the associated parameter vectors to its immediate neighbors. We show that in the presence of Byzantine agents, whose estimation and communication strategies are completely arbitrary, the estimates of the cooperative agents converge to a bounded consensus value with probability one, provided that there are at most H Byzantine agents in the network that is (2H+1)-robust. Furthermore, we prove that the policy of the cooperative agents converges with probability one to a bounded neighborhood around a stationary point of their team-average objective function under the assumption that the policies of the adversarial agents asymptotically become stationary.

关键词： Approximation algorithms Vectors Linear programming Function approximation Training Convergence Resilience Adversarial attacks Byzantine-resilient learning consensus cooperative multiagent reinforcement learning (MARL)

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：