检索结果-内蒙古大学图书馆

A Case Study Applying Mesoscience to Deep Learning

engineering 2024年第8期39卷 84-93页

作者： Li Guo Fanyong Meng Pengfei Qin Zhaojie Xia Qi Chang Jianhua Chen Jinghai Li State Key Laboratory of Multiphase Complex Systems Institute of Process EngineeringChinese Academy of SciencesBeijing 100190China School of Chemical Engineering University of Chinese Academy of SciencesBeijing 100049China

In this paper,we propose mesoscience-guided deep learning(MGDL),a deep learning modeling approach guided by mesoscience,to study complex *** establishing sample dataset based on the same system evolution data,different from the operation of conventional deep learning method,MGDL introduces the treatment of the dominant mechanisms of complex system and interactions between them according to the principle of compromise in competition(CIC)in *** constraints are then integrated into the loss function to guide the deep learning *** methods are proposed for the addition of mesoscience *** physical interpretability of the model-training process is improved by MGDL because guidance and constraints based on physical principles are *** was evaluated using a bubbling bed modeling case and compared with traditional *** a much smaller training dataset,the results indicate that mesoscience-constraint-based model training has distinct advantages in terms of convergence stability and prediction accuracy,and it can be widely applied to various neural network *** MGDL approach proposed in this paper is a novel method for utilizing the physical background information during deep learning model *** exploration of MGDL will be continued in the future.

关键词： Mesoscience Deep learning complex system Gas-solid system Bubbling bed

来源：评论

学校读者我要写书评

暂无评论

Adaptive Control for Stochastic Nonlinear systems with Input Saturation Based on Multi-dimensional Taylor Network 20

Adaptive Control for Stochastic Nonlinear Systems with Input...

引用

20th IEEE International Conference on Mechatronics and Automation, ICMA 2023

作者： Zheng, Xiao-Yi Yan, Hong-Sen Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Nanjing210096 China Southeast University School of Automation Nanjing210096 China

ISBN: (纸本)9798350320831

For stochastic nonlinear systems with input saturation, few of the existing control methods use a suitable auxiliary system to solve input saturation problem, and most of them are for deterministic systems. In this study, according to the backstepping approach and the Lyapunov stability theorem, utilizing multi-dimensional Taylor network (MTN), an adaptive controller is presented to secure that all closed-loop signals keep bounded in probability, in which MTN is implemented to approach the uncertain function during the controller design. Moreover, the impact raised by input saturation is mitigated by designing an appropriate auxiliary system, in which the order of the auxiliary system is the same as that of the controlled system. Furthermore, unlike the majority of prior findings, the final output signal could be constrained in a mean quartic sense by an explicit expression of the input saturation error (the difference between the control signal and the saturation input) and the parameters used for design. Finally, the efficiency of the suggested control methodology is confirmed by providing an example. © 2023 IEEE.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

LSTM-TD3-based control for delayed drone combat strategies 19th

LSTM-TD3-based control for delayed drone combat strategies

引用

19th Chinese Intelligent systems Conference, CISC 2023

作者： Ji, Bingyu Wang, Jun Zhang, Hailin Zhang, Ya School of Automation Southeast University Nanjing210096 China Key Laboratory Measurement and Control of Complex Systems of Engineering Ministry of Education Nanjing210096 China

ISBN: (纸本)9789819968817

Reinforcement learning has made great achievements in the field of game confrontation, and military rendition intelligence is also imminent. In this paper, we propose a game confrontation game model based on LSTM and deep reinforcement learning for the red and blue sides of the drones in the delayed *** action buffer pair is added to the original battlefield environment to create a large latency environment, a TD3 algorithm is designed to join LSTM to predict the state trend, and a historical data pool is built using PER. The training results in the simulation platform show that the model has a significant improvement in effectiveness in the large latency environment, with a 0.22 improvement in battle damage ratio and a 34.76% improvement in win rate over the original model. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Composite-mask GAN based on refined optical flow and disparity map for SLAM visual odometry

引用

Optoelectronics Letters 2025年

作者： JI Yuehui JIANG Jingwei LIU Junjie SONG Yu GAO Qiang School of Electrical Engineering and Automation， Tianjin University of Technology Tianjin Key Laboratory of Control Theory and Application for Complex Systems

Although deep learning methods have been widely applied in slam visual odometry over the past decade with impressive improvements， the accuracy remains limited in complex dynamic environments. In this paper， a composite mask-based generative adversarial network is introduced to predict camera motion and binocular depth maps. Specifically， a perceptual generator is constructed to obtain the corresponding parallax map and optical flow from between two neighboring frames. Then， an iterative pose improvement strategy is proposed to improve the accuracy of pose estimation. Finally， a composite mask is embedded in the discriminator to sense structural deformation in the synthesized virtual image， thereby increasing the overall structural constraints of the network model， improving the accuracy of camera pose estimation， and reducing drift issues in the Visual Odometer. Detailed quantitative and qualitative evaluations on the KITTI dataset show that the proposed framework outperforms existing conventional， supervised learning and unsupervised depth VO methods， providing better results in both pose estimation and depth estimation.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Acetic acid-assisted mild dealloying of fine CuPd nanoalloys achieving compressive strain toward high-efficiency oxygen reduction and ethanol oxidation electrocatalysis

引用

Carbon Energy 2023年第7期5卷 112-120页

作者： Danye Liu Yu Zhang Hui Liu Peng Rao Lin Xu Dong Chen Xinlong Tian Jun Yang State Key Laboratory of Multiphase Complex Systems Institute of Process EngineeringChinese Academy of SciencesBeijingChina State Key Laboratory of Marine Resource Utilization in South China Sea Hainan Provincial Key Lab of Fine ChemistrySchool of Chemical Engineering and TechnologyHainan UniversityHaikouChina Jiangsu Key Laboratory of New Power Batteries School of Chemistry and Materials ScienceJiangsu Collaborative Innovation Centre of Biomedical Functional MaterialsNanjing Normal UniversityNanjingChina Center of Materials Science and Optoelectronics Engineering University of Chinese Academy of SciencesBeijingChina Nanjing IPE Institute of Green Manufacturing Industry NanjingJiangsuChina

Dealloying by which the transition metal is partially or completely leached from an alloy precursor is an effective way to optimize the fundamental effects for further enhancing the electrocatalysis of a ***,to address the deficiencies associated with the commonly used dealloying methods,for example,electrochemical and sulfuric acid/nitric acid treatment,we report an acetic acid-assisted mild strategy to dealloy Cu atoms from the outer surface layers of CuPd alloy nanoparticles to achieve high-efficiency electrocatalysis for oxygen reduction and ethanol oxidation in an alkaline *** leaching of Cu atoms by acetic acid exerts an additional compressive strain effect on the surface layers and exposes more active Pd atoms,which is beneficial for boosting the catalytic performance of a dealloyed catalyst for the oxygen reduction reaction(ORR)and the ethanol oxidation reaction(EOR).In particular,for ORR,the CuPd nanoparticles with a Pd/Cu molar ratio of 2:1 after acetic dealloying show a half-wave potential of 0.912 V(***)and a mass activity of 0.213 AmgPd^(-1) at 0.9 V,respectively,while for EOR,the same dealloyed sample has a mass activity and a specific activity of 8.4 Amg^(-1) and 8.23 mA cm^(-2),respectively,much better than their dealloyed counterparts at other temperatures and commercial Pd/C as well as a Pt/C catalyst.

关键词： compressive strain effect dealloying electrocatalysis ethanol oxidation reaction oxygen reduction reaction

来源：评论

学校读者我要写书评

暂无评论

Solar-powered miniature robot for on-site oil spill treatment under magnetic steering

引用

Science China(Technological Sciences) 2024年第7期67卷 2164-2175页

作者： WANG Ben CHEN YunRui WANG Yun JI FengTong LI HongYuan NIU ShiChao HAN ZhiWu ZHANG YaBin GUO ZhiGuang College of Chemistry and Environmental Engineering Shenzhen UniversityShenzhen 518060China State Key Laboratory of Featured Metal Materials and Life-cycle Safety for Composite Structures Ministry of Education Key Laboratory of New Processing Technology for Nonferrous Metals and Materialsand School of ResourcesEnvironment and MaterialsGuangxi UniversityNanning 530004China Key Laboratory for Bionic Engineering Ministry of Education Jilin UniversityChangchun 130022China Hubei Collaborative Innovation Centre for Advanced Organic Chemical Materials Ministry of Education Key Laboratory for the Green Preparation and Application of Functional MaterialsHubei UniversityWuhan 430062China Wellcome Trust/Cancer Research UK Gurdon Institute University of CambridgeCambridge CB21QNUK Department of Physiology Development and NeuroscienceUniversity of CambridgeCambridge CB21QNUK State Key Laboratory for Turbulence and Complex Systems Department of Mechanics and Engineering ScienceBeijing Innovation Center for Engineering Science and Advanced TechnologyCollege of EngineeringPeking UniversityBeijing 100871China

Walking on the water surface is an effective method for miniature robots to transport payloads with dramatically decreased interfacial drag. Current aquatic robots reported are generally actuated by a beam of focused light that can trigger asymmetrical deformation, enabling the directional movement through horizontal momentum transfer of photoinduced actuation force to the water. However, the operations are heavily dependent on manual manipulation of the focused light, making the long-term actuation and application of the aquatic robots in vast scenarios challenging. Herein, we developed a kind of water striderinspired robot that can autonomously manage the motion on the water surface under solar irradiation, with their direction steerable by a magnetic field. The motion of this bioinspired robot on the water surface was achieved by the use of a solar cell panel as a driving module to enable propulsive motion based on the conversion of light-electric-mechanical energies. The superhydrophobic design of its leg surfaces enables the aquatic robots with weight-bearing and drag-reducing abilities. With the assistance of magnetic navigation, the bioinspired robot can continuously and controllably locomote to the oily spill floating on the water body and collect them with high efficiency. For further demonstration, the treatment of oil spills in a campus pool with high efficiency has also been achieved. This on-site oil-spill treating strategy, taking advantage of a home-made bioinspired robot actuated by natural sunlight under magnetic steering, shows great potential applications in water-body remediation.

关键词： miniature robot superhydrophobic water strider interfacial friction magnetic actuation

来源：评论

学校读者我要写书评

暂无评论

Non-Smooth Speed Control of Switched Reluctance Motor Based on Generalized Proportional Integral Observer 42

Non-Smooth Speed Control of Switched Reluctance Motor Based ...

引用

42nd Chinese Control Conference, CCC 2023

作者： Yin, Qingzhao Tian, Zheng Li, Shihua School of Automation Southeast University Nanjing210096 China Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Nanjing210096 China

ISBN: (纸本)9789887581543

This paper investigates the speed regulation control of switched reluctance motor (SRM) systems. To improve the antidisturbance performance of SRM, a composite non-smooth control strategy is proposed. First, the structure of SRM is analyzed, and a simplified nonlinear model is obtained based on a segmented representation of the varying phase inductance. A virtual control function is introduced to represent the nonlinear part of the torque equation, whose inverse function is cascaded to linearized the nonlinear model. Second, a generalized proportional integral observer (GPIO) is constructed to estimate the lumped disturbance of the system, which is used for feedforward compensation design. Finally, a composite speed controller is designed based on a combination of finite time proportional feedback and feed-forward compensation based on GPIO (FTP+GPIO). The speed error closed-loop system can be regarded as a first-order finite time control system with bounded disturbances. Strict analysis shows that the proposed scheme can improve the anti-disturbance performance of the closed-loop system. The effectiveness of the proposed method is verified by simulation results. Moreover, it is compared with proportional feedback combining feed-forward compensation (P+GPIO) method and proportional integral (PI) control method. © 2023 Technical Committee on Control Theory, Chinese Association of Automation.

关键词： Closed loop systems

来源：评论

学校读者我要写书评

暂无评论

Stability Analysis of Load Frequency Control System with Time-Varying Delays

Stability Analysis of Load Frequency Control System with Tim...

引用

2023 China Automation Congress, CAC 2023

作者： Fan, Chunling Chen, Chaoyang Zhou, Shaowu Lian, Honghai Chen, Zuguo Li, Pei School of Information and Electrical Engineering Hunan Provincial Key Laboratory of Intelligent Control and Maintenance of Complex Systems Hunan University of Science and Technology Xiangtan China

ISBN: (纸本)9798350303759

In this paper, the stability analysis of Load frequency Control (LFC) systems with time-varying delay is conducted. Firstly, an augmented Lyapunov-Krasovskii (L-K) functional is designed to incorporate the relevant information of the delay state. Then, the functional derivatives are estimated using an auxiliary-function-based integral inequality in conjunction with an improved reciprocally convex inequality. Furthermore, the correlation function is processed through a quadratic function negative definite derivation. Finally, case studies established on a one-area LFC system are conducted to demonstrate the superiority of the method, and its accuracy is verified through a simulation case. © 2023 IEEE.

关键词： Electric control equipment

来源：评论

学校读者我要写书评

暂无评论

Accelerated Saddle Flow Dynamics for Bilinearly Coupled Minimax Problems 63

Accelerated Saddle Flow Dynamics for Bilinearly Coupled Mini...

引用

63rd IEEE Conference on Decision and Control, CDC 2024

作者： Liu, Yingzhu Mallada, Enrique Li, Zhongkui You, Pengcheng Peking University College of Engineering Beijing China Peking University State Key Laboratory for Turbulence and Complex Systems Beijing China Johns Hopkins University Department of Electrical and Computer Engineering BaltimoreMD United States

ISBN: (纸本)9798350316339

Minimax problems have attracted much attention due to various applications in constrained optimization problems and zero-sum games. Identifying saddle points within these problems is crucial, and saddle flow dynamics offer a straightforward yet useful approach. This study focuses on a class of bilinearly coupled minimax problems with strongly convex-linear objective functions. We design an accelerated algorithm based on saddle flow dynamics, achieving a convergence rate beyond the stereotype limit (the strong convexity constant). The algorithm is derived from a sequential two-step transformation of a given objective function. First, a change of variables is applied to render the objective function better-conditioned, introducing strong concavity (from linearity) while preserving strong convexity. Second, proximal regularization, when staggered with the first step, further enhances the strong convexity of the objective function by shifting some of the obtained strong concavity. After these transformations, saddle flow dynamics based on the new objective function can be tuned for accelerated exponential convergence. Besides, such an approach can be extended to weakly convex-weakly concave functions and still guarantees exponential convergence to one stationary point. The theory is verified by a numerical test on an affine equality-constrained convex optimization problem. © 2024 IEEE.

关键词： Constrained optimization

来源：评论

学校读者我要写书评

暂无评论

Manipulating d-d orbital hybridization induced by Mo-doped Co_(9)S_(8) nanorod arrays for high-efficiency water electrolysis

引用

Journal of Energy Chemistry 2024年第6期93卷 592-600,I0015页

作者： Xue Zhou Jing Li Guangyao Zhou Weiran Huang Yucan Zhang Jun Yang Huan Pang Mingyi Zhang Dongmei Sun Yawen Tang Lin Xu Jiangsu Key Laboratory of New Power Batteries Jiangsu Collaborative Innovation Centre of Biomedical Functional MaterialsSchool of Chemistry and Materials ScienceNanjing Normal UniversityNanjing 210023JiangsuChina 'College of Science Jinling Institute of TechnologyNanjing 211169JiangsuChina State Key Laboratory of Multiphase Complex Systems Institute of Process EngineeringChinese Academy of SciencesBeijing 100190China Nanjing IPE Institute of Green Manufacturing Industry Nanjing 211100JiangsuChina School of Chemistry and Chemical Engineering Yangzhou UniversityYangzhou 225009JiangsuChina Key Laboratory for Photonic and Electronic Bandgap Materials Ministry of EducationSchool of Physics and Electronic EngineeringHarbin Normal UniversityHarbin 150025HeilongjiangChina

Precisely refining the electronic structure of electrocatalysts represents a powerful approach to further optimize the electrocatalytic ***,we demonstrate an ingenious d-d orbital hybridization concept to construct Mo-doped Co_(9)S_(8) nanorod arrays aligned on carbon cloth(CC)substrate(abbreviated as Mo-Co_(9)S_(8)@CC hereafter)as a high-efficiency bifunctional electrocatalyst toward water *** has experimentally and theoretically validated that the 4d-3d orbital coupling between Mo dopant and Co site can effectively optimize the H_(2)O activation energy and lower H^(*)adsorption energy barrier,thereby leading to enhanced hydrogen evolution reaction(HER)and oxygen evolution reaction(OER)*** to the unique electronic and geometrical advantages,the optimized Mo-Co_(9)S_(8)@CC with appropriate Mo content exhibits outstanding bifunctional performance in alkaline solution,with the overpotentials of 75 and 234 mV for the delivery of a current density of 10 mA cm^(-2),small Tafel slopes of 53.8 and 39.9 mV dec~(-1)and long-term stabilities for at least 32 and 30 h for HER and OER,*** impressively,a water splitting electrolylzer assembled by the self-supported Mo-Co_(9)S_(8)@CC electrode requires a low cell voltage of 1.53 V at 10 mA cm^(-2)and shows excellent stability and splendid reversibility,demonstrating a huge potential for affordable and scalable electrochemical H_(2) *** innovational orbital hybridization strategy for electronic regulation herein provides an inspirable avenue for developing progressive electrocatalysts toward new energy systems.

关键词： d-d orbital hybridization Transition metal sulfides Nanorods arrays Water electrolysis

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：